Articles Archives – Page 3 of 57 – Southern California Law Review

Fair Use and Fair Price

July 2025July 2025Roy Baharad* & Gideon Parchomovsky†

In this Article, we present and develop a new justification for the fair use doctrine. The accepted lore among copyright law scholars is that fair use is a means for overcoming a market failure in the form of high transaction costs. According to this view, the doctrine sanctions unauthorized use of copyrighted works in cases where transaction costs hinder voluntary, mutually beneficial exchanges.

Departing from conventional wisdom, we argue that the fair use doctrine serves as an important empowerment even in fully functional markets. Fair use enables users to secure more favorable licensing terms from copyright owners by endowing users with a threat point in their negotiations. Without fair use, users would have to pay the price demanded by copyright owners or not use the work. With fair use, many users can credibly assert that their intended use of copyrighted content is privileged by the fair use doctrine and thus they can use the desired content without authorization. The fair use doctrine, therefore, gives users leverage in their negotiations with copyright owners.

We illustrate our thesis by applying it to the landmark fair use decisions of the Supreme Court, including the recent ruling in Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith, that determined the bargaining standpoint of users for decades to come. We demonstrate the distributive effects of the Supreme Court’s fair use jurisprudence and explain how the Court must act to preserve and augment the empowering effect of fair use. The theory presented in this Article proves that the reach of fair use goes well beyond market failures and that the impact of the doctrine is much more significant than previously thought.

INTRODUCTION

Considered by many as the most important doctrine in our copyright law system,¹ fair use sanctions certain unauthorized uses of copyrighted works that would otherwise constitute a copyright infringement. Under the fair use doctrine, when a use is considered fair, the user is relieved of liability and need not pay compensation to the copyright owner whose content they used. Therefore, fair use may be conceptualized as a doctrine that confers upon deserving users a private taking power over copyrighted content that can be exercised at a zero price.²

Since its inception in the English common law and equity courts, the fair use doctrine has never ceased to fascinate theorists and students.³ The existence of the doctrine raised two critical challenges for scholars. First, under what circumstances should a use be considered fair? Second, why should fair users be fully relieved of the duty to compensate copyright owners?

The answer to both these questions has been provided by Professor Wendy Gordon. In a pathbreaking article authored almost forty years ago, Gordon conceptualized fair use as a means for overcoming a market failure in the form of high transaction costs.⁴ Gordon persuasively argued that courts should recognize fair use when three cumulative conditions obtain: first, high transaction costs prevent voluntary market exchange between copyright owners and users; second, the allegedly fair use is socially beneficial; and, third, a fair use finding would not unduly undermine incentives to create.⁵ Gordon’s key insight was that when transaction costs are prohibitive, there will be no voluntary trade between copyright owners and users. In this scenario, allowing users to use copyright content for free benefits the user without harming copyright owners, for the latter would not be able to collect payments from users as transaction costs bar voluntary exchanges. Gordon correctly submitted that under these circumstances, allowing users to use copyright content for free, by classifying their use as fair, is welfare enhancing.

By tying fair use to the level of transaction costs, Gordon, at once, provided a cogent defense of the fair use doctrine and exposed its vulnerability. Professor Tom Bell and other scholars pointed out that in an interconnected world, where technological advancements constantly lower transaction costs, there may no longer be a need for fair use, and at a minimum, courts’ willingness to recognize fair uses should diminish.⁶ Importantly, this skeptical view of fair use has not been universally endorsed by other scholars, who advanced other utilitarian and non-utilitarian justifications for fair use.⁷

In this Article, we develop a new justification for fair use that is radically different from prior theorizing. Our theory seeks to complement and reinvigorate the theories of scholars who view fair use as an essential component of our copyright system. Yet, our outlook places fair use on a very different ground from past scholarship. We contend that fair use’s most significant yet overlooked role is to facilitate bargaining between copyright owners and users, even when transaction costs are low. We argue that fair use helps users not only in those unrepresentative and rare cases where transaction costs prevent consensual transactions between copyright owners and users, but also in the more common case where transaction costs are low or nonexistent. Fair use does this by improving the bargaining power of users and giving them leverage or a threat point vis-à-vis copyright owners.

Although fair use has been traditionally considered an open-ended and unpredictable doctrine,⁸ two recent developments have infused a certain level of predictability into this area of the law. First, a close reading of fair use cases uncovers, what we call, two fair use clusters: transformative uses and uses that yield a substantive public benefit.⁹ Second, an empirical study by Professor Barton Beebe established a meaningful positive correlation between two of the statutory factors that courts are instructed to weigh in making fair use determinations—the purpose of the use and the effect of the use on the market for the copyrighted work.¹⁰ This emergence of clusters of uses or activities that have a legitimate claim for fair use is a welcome development for users, as it manifests in increased licensing leverage.

Consider transformativeness. Transformativeness covers a wide range of uses. While the level of transformativeness varies among uses, all users who engage in transformative uses of copyrighted works have a colorable fair use claim. Obviously, not all of them would win a fair use ruling in court, but each can credibly argue in negotiations with copyright owners that their use would be found fair with a certain probability. Since transformativeness, as recently established by the Supreme Court, “is a matter of degree,”¹¹ the level of transformativeness can be represented on a unit interval, essentially reflecting its probability of enjoying a fair use defense. This probability, in turn, immediately translates into licensing leverage when negotiating with a monopolistic rightsholder.

To illustrate, imagine that Anne plans to make a transformative use of a photograph in which Bob owns a copyright. Assume that there is 0.5 probability that Anne’s use would be found fair if she uses Bob’s photograph without his permission and a lawsuit ensues. Assume further that Bob typically demands $100 per license from users who wish to use his photograph. Anne, however, owing to her potential fair use claim, should be able to secure a license for $50—a price that reflects the strength of her fair use claim.

Now consider a different user, Carol, whose planned use of Bob’s work is slightly less transformative than Anne’s. The probability of Carol’s use being fair is only 0.3. Yet, all things being equal, she, too, should be able to receive a license from Bob for a lower price than his original asking price—$70, instead of $100. As we shall demonstrate throughout this Article, these examples are representative.

The same is true for a host of other transformative users—users who can claim that their derivative works bestow a significant benefit on the public, and users who appropriate only a small portion from copyrighted works. All of them have a certain individual probability of succeeding on a fair use claim when sued by the copyright owner. The license price each would be able to negotiate depends on the strength of their fair use claim, as well as on their bargaining power. But critically, each should be able to negotiate a license for a price lower than the asking price of the copyright owner. We use a formal model to demonstrate this result and identify its robustness over a wide range of scenarios. As we will show, even a small probability of obtaining a fair use ruling may well change the bargaining outcome between users and copyright owners.

The Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith decision that was recently issued by the Supreme Court provides a powerful example of our theory.¹² In this case, the Supreme Court had to determine whether fourteen unauthorized silkscreens and two unauthorized pencil drawings that were produced based on a Lynn Goldsmith photograph qualify as fair uses of the photograph.¹³ At the heart of the matter lies the definition of the term “transformativeness.” In rejecting the foundation’s fair use claim, the Court held that a transformative work must have a fundamentally different and new artistic purpose and character.¹⁴ This definition is clearly narrower than that endorsed by the Supreme Court in Campbell v. Acuff-Rose Music, Inc., according to which a transformative work is one that has a different purpose or conveys new message or meaning.¹⁵ The Supreme Court’s adherence to a rather narrow interpretation of transformativeness will carry far-reaching implications not only for the parties to the case, but also for a multitude of users who negotiate with copyright owners.

It bears emphasis that negotiation is often a prelude to litigation. In fact, many of the celebrated fair use cases, including Google LLC v. Oracle America, Inc.,¹⁶ Authors Guild, Inc. v. Google Inc.,¹⁷ and Campbell,¹⁸ were filed only after the parties tried, and failed, to negotiate a consensual agreement. Furthermore, even after a case is brought to court, the litigating parties attempt to negotiate a settlement while the case is pending. Indeed, such was the case in Oracle and Authors Guild. Naturally, not all negotiations result in a successful outcome, and some copyright disputes will inevitably end up in court. But it should be underscored that in any regular case, in which the existence of transaction costs does not hinder trade, the rightsholder-user negotiation becomes the primary apparatus for allocating use of copyrighted content; litigation is nothing but a complementary mechanism. This understanding implies that fair use theory must account for the doctrine’s impact on the negotiation process.

Our analysis shows that irrespective of its actual invocation in court, the fair use doctrine provides considerable benefits to users by improving their bargaining position. In economic parlance, the theory advanced by this Article views fair use as empowering significant categories of users by improving what negotiation theorists call the “best alternative to a negotiated agreement” (“BATNA”).¹⁹ By diminishing the BATNA of copyright holders and bolstering that of users, the fair use doctrine redesigns the licensor-licensee relationship to promote the use and distribution of copyrighted content.

The present Article thus conceptualizes fair use not as a mechanism that overcomes transaction-costs-related market failures, but rather, as a doctrine that rests the foundations for just and efficient bargaining framework. In this regard, we analyze the behavior of relevant economic actors that operate “in the shadow” of the fair use doctrine.²⁰

We develop our argument in accordance with the following structure. In Part I, we explain the fair use doctrine and discuss the theories that have been developed to justify it. In Part II, we introduce a new theory of fair use and explicate how it differs from the extant literature. In Part III, we present the normative attractiveness of our theory by highlighting its positive welfare and distributive implications. A short Conclusion ensues.

I. Fair Use: Law and Theory

A. The Fair Use Doctrine

The fair use doctrine made its first appearance in the U.S. in the 1841 case of Folsom v. Marsh.²¹ According to scholars, however, its origins are far more ancient. Professor Matthew Sag, for example, suggests that the fair use doctrine “predate[d] Folsom v. Marsh by at least 100 years.”²² Professor William Patry went even further and dated the doctrine back to the year 1710 when the Statute of Anne was enacted.²³ Other researchers trace the roots of the fair use doctrine to the common law and natural rights conceptions.²⁴

In 1976, the fair use doctrine was codified and became part of the Copyright Act. 17 U.S.C. § 107 (“section 107”), in which the fair use doctrine is enshrined, opens with a preamble that offers a non-exhaustive list of presumptively fair uses, including, “criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research.”²⁵ Then, it proceeds to enumerate four factors that courts ought to consider when making fair use determinations: “(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; [and] (4) the effect of the use upon the potential market for or value of the copyrighted work.”²⁶

It is important to note that the codification of the fair use doctrine did not purport to change its nature as an equitable doctrine. In the accompanying House Report, Congress referred to the fair use doctrine as “an equitable rule of reason.”²⁷ The statutory formulation of the doctrine, therefore, preserved the flexibility and open-endedness that have become the hallmark of the fair use doctrine. As Judge Pierre Leval wrote in his 1990 classic article on fair use:

What is most curious about this doctrine is that neither the decisions that have applied it for nearly 300 years, nor its eventual statutory formulation, undertook to define or explain its contours or objectives. . . . [They] furnish little guidance on how to recognize fair use. The statute, for example, directs us to examine the “purpose and character” of the secondary use as well as “the nature of the copyrighted work.”²⁸

Even more remarkable was Judge Leval’s admission that although “courts have treated the definition of the doctrine as assumed common ground[, t]he assumption of common ground is mistaken. Judges do not dineshare a consensus on the meaning of fair use.”²⁹ Indeed, the courts’ inability to converge on a common understanding of fair use has frustrated copyright scholars, leading Professor Larry Lessig to conclude that the fair use doctrine amounts to nothing other than “the right to hire a lawyer to defend your right to create.”³⁰

While scholars have bemoaned the uncertainty that shrouds the fair use doctrine³¹ and, over the years, have advanced various proposals in order to cabin the unpredictability of fair use,³² in this Article, we take a different tack. We argue, contrary to conventional wisdom, that the inherent uncertainty of the fair use doctrine might actually help, rather than harm, users. The possibly virtuous effect of uncertainty has been overlooked by theorists as they have focused exclusively on the litigation arena. We, by contrast, are interested in the effect of fair use outside of the courtroom, in negotiations between copyright owners and users. We develop this argument fully in Part II of this Article, as part of our transactional model of fair use. But before elaborating on the potential virtues of uncertainty, it behooves us to complete our discussion of the development of the fair use doctrine and highlight some critical recent developments.

Four years after Judge Leval’s 1990 portrayal of fair use as a helplessly underminable doctrine, the fair use landscape was reshaped. The turning point was the Supreme Court’s 1994 decision in Campbell v. Acuff-Rose Music, Inc.. In finding 2 Live Crew’s version of Roy Orbison’s and Bill Dees’s copyrighted song “Oh Pretty Woman” a fair use, the Court stated that “the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.”³³ As we will show, lower courts were quick to follow the language and spirit of the Campbell opinion. Since Campbell, transformativeness has become the currency of the fair use realm.³⁴

The importance of transformativeness to fair use determinations was highlighted by Judge Leval in his classic article.³⁵ Yet, it was the Supreme Court’s endorsement of transformativeness in Campbell that officially conferred upon it an elevated status.³⁶ Subsequent cases have substantially increased the weight granted to transformativeness in fair use determinations. Among others, courts held that a challenged work is worthy of protection whenever it “contains significant transformative elements,”³⁷ is “sufficiently transformative,”³⁸ or is endowed with a “patently transformative character.”³⁹ It has been further emphasized that a central purpose of fair use investigations is to establish “whether and to what extent the new work is ‘transformative.’ ”⁴⁰ In other instances, courts submitted that copyright infringement, notwithstanding the commercial character of the work in question, may not be determined given that the original work is “used for a transformative purpose.”⁴¹

The effect of the rise of transformativeness can be best seen in the context of fair use cases involving appropriation art. Appropriation art is an art form predicated on the use of existing objects with subtle modifications.⁴² Cases on appropriation art include the Second Circuit’s famous Blanch v. Koons⁴³ and Cariou v. Prince.⁴⁴ In both cases, the court was persuaded of the sufficient transformation that the process of appropriation can embed. In Blanch, the court explained that Jeff Koons’s appropriation of Blanch’s photograph “was intended to be—and appears to be—‘transformative.’ ”⁴⁵ In Cariou, the court was more hesitant but nonetheless concluded that Richard Prince’s appropriating work “could be transformative even without commenting on Cariou’s work or on culture, and even without Prince’s stated intention to do so.”⁴⁶

A closely related development in fair use jurisprudence that can also be traced back to Campbell involves parodies. A parody is defined as a derivative work whose purpose is to criticize or comment on a preexisting copyrighted work.⁴⁷ After classifying 2 Live Crew’s unauthorized rendition of “Oh Pretty Woman” as a parody, the Court likewise ruled that parodic uses are especially transformative and therefore have an especially strong claim for fair use.⁴⁸ The Court added that copyright owners are unlikely to authorize parodies that are of unflattering commentary or mockery to their works,⁴⁹ and thus, the production of parodies critically depends on fair use.

The Supreme Court’s ruling in Campbell has been understood by lower courts and commentators to create something of a safe harbor for parodies.⁵⁰ As Professor Pamela Samuelson observed, “[n]otwithstanding the Court’s unwillingness in Campbell to presume that parodies are fair, every subsequent parody case has been adjudged a fair use.”⁵¹ For example, in Suntrust Bank v. Houghton Mifflin Co., the Eleventh Circuit ruled that Alice Randall’s “The Wind Done Gone,” a critical literary account of Margaret Mitchell’s “Gone With the Wind,” was a parody of the original and was therefore a fair use of it.⁵² Likewise, in Burnett v. Twentieth Century Fox Film Corp., the Court ruled that an audiovisual work produced by the creators of Family Guy, which poked fun at the figure of Carol Burnett, constituted a fair use on account of its parodic nature.⁵³ Similarly, in Brownmark Films, LLC v. Comedy Partners, the Seventh Circuit held that a music video by the creators of South Park which parodied the plaintiff’s music video has “obvious transformative value.”⁵⁴ Recently, the Second Circuit openly acknowledged that in the aftermath of Campbell, “parody, which ‘needs to mimic an original to make its point,’ . . . is routinely held transformative.”⁵⁵

Another line of cases, beginning with Kelly v. Arriba Soft Corp.,⁵⁶ highlighted another key determinant of fair use: social benefit. Importantly, this category, too, grew out of the Supreme Court’s Campbell decision, with its emphasis on transformativeness. In Campbell, the Court suggested that a “parody has an obvious claim to transformative value” because “it can provide social benefit, by shedding light on an earlier work, and, in the process, creating a new one.”⁵⁷ The fair use cases that thrust public benefit to the forefront of the fair use analysis predominantly involved users from the technology sector. In Kelly, the Ninth Circuit ruled that the public display of photographs in the form of thumbnails as part of the operation of a visual search engine constituted fair use.⁵⁸ In reaching this conclusion, the court mentioned the benefit conferred on the public by the appellee’s search engine.⁵⁹ Approximately four years later, in 2007, in Perfect 10, Inc. v. Amazon.com, Inc., the Ninth Circuit was asked to revisit the issue, when an adult content company sued Google, alleging that its authorized display of its copyrighted photos as thumbnails in response to users’ searches constituted copyright infringement.⁶⁰ Finding that Google’s use was fair, the Ninth Circuit emphasized the “significant public benefit” of Google’s search system, explaining:

[A] search engine provides social benefit by incorporating an original work into a new work, namely, an electronic reference tool. Indeed, a search engine may be more transformative than a parody because a search engine provides an entirely new use for the original work, while a parody typically has the same entertainment purpose as the original work.⁶¹

The social benefit factor quickly found its way to the decisions of other circuits.⁶² It played an important role in the Second Circuit’s ruling in Authors Guild v. Google, Inc..⁶³ The case was brought after Google decided to establish a large, searchable digital repository of literary works by scanning the books in several libraries around the world, including the libraries of Harvard University, The University of California, Stanford University, The University of Michigan, Columbia University, Princeton University and the New York Public Library.⁶⁴ The goal was to allow users to search the content of the books, yet the content of books could not be copied and only snippets from books were displayed to users in response to their searches.⁶⁵ Nonetheless, after prolonged negotiations between the parties failed to yield a settlement, the Authors Guild sued Google for copious copyright infringements.⁶⁶ In reaching the conclusion that Google’s use was fair, the district court dedicated a full section of its decision to a discussion of the myriad public benefits arising from Google’s use.⁶⁷ On appeal, the Second Circuit affirmed the district court’s decision, awarding Google an important legal victory, but toned down the importance of the public benefits provided by Google.⁶⁸

The final imprimatur of public benefit as a key determinant of fair use was given by the Supreme Court in its 2021 decision in Google LLC v. Oracle America, Inc.⁶⁹ In a 6-2 decision, the Supreme Court ruled that Google’s unauthorized appropriation of 11,500 lines of Oracle’s Java Application Programming Interface (“API”) for the Android operating system constitutes fair use.⁷⁰ Writing for the majority, Justice Breyer noted the added value created by Google’s use for third parties:

Here Google’s use of the Sun Java API seeks to create new products. It seeks to expand the use and usefulness of Android-based smartphones. Its new product offers programmers a highly creative and innovative tool for a smartphone environment. To the extent that Google used parts of the Sun Java API to create a new platform that could be readily used by programmers, its use was consistent with that creative “progress” that is the basic constitutional objective of copyright itself.⁷¹

Importantly, it was not only the courts that helped dispel some of the doctrinal mist that enveloped the fair use doctrine for centuries. Academics, too, have managed to point to overarching principles that affect fair use determinations. Two such efforts are worthy of special note.

In two separate projects, the first published in 2008⁷² and the second in 2020,⁷³ Professor Barton Beebe empirically examined fair use case law to find correlations between various fair use factors and case results. His method allowed him to pierce the judicial rhetoric and examine which statutory factors are outcome determinative in fair use cases. In his 2008 article, which surveyed the opinions from all fair use cases issued between 1978 and 2005, Beebe noted that “[i]t appears . . . that courts and commentators have exaggerated the influence of transformativeness doctrine on our fair use case law.”⁷⁴ However, in his 2020 article that analyzed the opinions from all fair use cases issued between 2005 and 2019, Beebe reports that “while the transformativeness test appeared to be waning in influence by 2005, it has since recovered its previous level of influence, even in the lower-profile, workaday fair use opinions that make up the majority of the data.”⁷⁵ Along similar lines, Clark Asay, Arielle Sloan and Dean Sobczak have empirically established that if courts perceive a certain use as transformative, it would almost invariably qualify for fair use protection, which implies that transformativeness is essentially a sufficient condition for enjoying the fair use defense.⁷⁶ These recent findings suggest an important confluence between judicial rhetoric and reality.

The Supreme Court’s recent decision in Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith⁷⁷ unfolded the most recent development in fair use jurisprudence. Issued in May 2023, the ruling is the first application of fair use in the arts and entertainment industry since the 1994 Campbell ruling. The Warhol case involves a series of Prince’s photos taken by photographer Lynn Goldsmith in 1981 to accompany a Vanity Fair article concerning Prince.⁷⁸ Unbeknownst to Goldsmith, the magazine solicited Andy Warhol to create a stylized painted version of the photograph that became known as the “Orange Prince” silkscreen.⁷⁹ Vanity Fair published the Orange Prince portrait in its November 1984 edition, and Goldsmith was co-credited for this work.⁸⁰ Orange Prince, however, was only one of sixteen painted versions created by Warhol, collectively known as Warhol’s “Prince Series.”⁸¹ Pursuant to its completion, the Prince Series has been routinely displayed in museums and galleries.⁸² After Prince’s passing in 2016, Condé Nast, the official publisher of Vanity Fair, published a commemorative magazine entitled “The Genius of Prince,” with the Orange Prince portrait embellishing its cover.⁸³ Alleging copyright infringement, Goldsmith argued that notwithstanding her initial 1984 licensing agreement with Condé Nast to use her photo one time as an “artistic reference,” she was unaware of the Orange Prince silkscreen—as well as of the Prince Series at large—until its reintroduction as part of the commemorative 2016 edition.⁸⁴

Once Goldsmith learned of the aforementioned facts, she brought an infringement suit against the foundation. The Southern District of New York granted the foundation’s motion for a preliminary ruling that Warhol’s work, though it incorporated Goldsmith’s photograph, was sufficiently transformative to qualify as a fair use.⁸⁵ The ruling was then overturned by the Second Circuit.⁸⁶ Finding that Warhol’s portrait incorporated no significant addition or alteration upon Goldsmith’s original photograph, the court concluded that the portrait was infringing.⁸⁷ In a 7-2 decision, the Supreme Court elected to affirm the Second Circuit’s conservative interpretation of transformativeness, thus rejecting the appeal and siding with Goldsmith.⁸⁸ The Warhol case thus marks a potential deviation from status quo interpretation of the fair use doctrine. We analyze the implications of the rulings to our theory in Part II, below.

The preceding discussion teaches that the Supreme Court’s approach to fair use has interjected a certain degree of certainty and predictability into the doctrine. This is not to say that fair use has become a clear doctrine—far from it—but it can be generally stated that in the aftermath of Campbell, parodic uses, sufficiently transformative uses, and uses that produce significant public benefits are likely to be found fair. As we will explain in Part II, the creation of these fair use clusters, or silos, is of vital importance to our theory of fair use. But before introducing our own theory of fair use, we must give credit to prior theorists and discuss their contributions. It is to this task that we next turn.

B. Theoretical Justifications of the Fair Use Doctrine

As befits a doctrine of its significance, scholars have advanced several theories to justify fair use. In this Section, we review the leading theoretical justifications of fair use. It should be emphasized at the outset that our goal is not to discredit other theories or even criticize them. As we will show, our theory of fair use complements existing theories by elucidating a central function of fair use that has hitherto evaded other scholars. The goal of the proceeding discussion is twofold: first, we wish to map the theoretical landscape of fair use, so we can precisely locate our own theory within it. Second, we seek to show how our justification of fair use interacts with the extant theoretic literature.

The most dominant theory has been put forth by Wendy Gordon.⁸⁹ In an immensely influential article, Gordon argued that the fair use doctrine is a mechanism for allowing the use of copyrighted content when voluntary transactions between copyright owners and users are barred by high transaction costs.⁹⁰ Specifically, Gordon postulated that fair use should be recognized when three cumulative conditions are met: (1) high transaction costs prevent consensual bargaining between copyright owners and users; (2) the unauthorized use is socially desirable; and (3) legitimizing the disputed use would not undermine incentives to create.⁹¹

To illustrate the operation of Gordon’s theory, imagine a student who wishes to quote a copyrighted manuscript. Assume that the author of the manuscript charges $20 for the requested use and that is also the maximum price that the student is willing to pay. However, transaction costs—defined as the cost of identifying the counterparty to the transaction, negotiating and formalizing an agreement with them, and enforcing the agreement—would bar the exchange from taking place. Under these circumstances, allowing the student to quote the manuscript without permission would make them (and society at large) better off without harming the author. The author, in our example, could not receive payment from the student, not because they refused to pay, but rather owing to the fact that the level of transaction costs made payment impossible. Hence, the author stands to lose nothing if the student’s use is considered fair and the student receives a benefit of $20. Society, too, is benefitted by the award of fair use to the student since the realm of creativity is enriched, while incentives to create future works are not harmed.

Gordon’s approach grounded fair use in economic theory, proving that the doctrine can be welfare enhancing when applied properly. However, the tie Gordon created between fair use and high transaction costs has proven to be a double-edged sword. As the title of her article, Fair Use as a Market Failure suggests, Gordon justified fair use as a means of overcoming a market failure in the form of high transaction costs. This was a great strength of Gordon’s theory, but also a potential weakness. Critically, Gordon’s theory established a direct correlation between the level of transaction costs and the cases in which fair use should be recognized. The problem is that improvements in telecommunications and computer technologies have dramatically lowered transaction costs since Gordon published her article in 1982, putting a lot of strain on Gordon’s theory. Critics of Gordon’s theory suggested that in an interconnected world, fair use has outlived its “raison d’être” and all uses must be secured via payment.⁹² It bears emphasis that Gordon responded to her critics by broadening and sharpening the role of market failure in her account. Yet, Gordon’s account remains inextricably related to the concept of market failure, and it conceives fair use as a mechanism that operates in non-transactional settings, i.e., settings in which voluntary transactions are vitiated.⁹³

A slightly different justification of fair use that complements Gordon’s original justification has been offered by Ben Depoorter and Francesco Parisi. Depoorter and Parisi correctly pointed out that technological advancements do not lower all transaction costs.⁹⁴ The problem of strategic holdouts remains, even in the face of technological advancements. The holdout problem is especially acute for users, such as creators of documentary works, who need to secure permission from multiple copyright owners. In such cases, each copyright owner possesses veto power over the planned use and may strategically exercise it to the detriment of users. The fair use doctrine allows users to carry out their creative projects despite strategic attempts by copyright owners to hamper the enterprise.⁹⁵ Depoorter’s and Parisi’s account is both persuasive and elegant, but “it potentially limits the usefulness of the market-failure theory to only cases in which a user must clear multiple rights and has no other alternatives.”⁹⁶

Two additional justifications of fair use focus on allocative efficiency. The first, associated with William Fisher, calls on courts to use fair use to promote the goals of copyright protections.⁹⁷ To this end, Fisher asks courts to think of all possible uses of copyrighted works. Then, for each use, he calls on courts to design what can be called an “efficiency ratio,” with the numerator representing the profit an author could realize if fair use is denied and the denominator representing the loss to society if the copyright owner refuses to license the work.⁹⁸ A high efficiency ratio suggests, per Fisher, that the use is probably unfair, while a low one indicates that the use is fair.⁹⁹ Fisher sets the cutoff at the use with the highest marginal aggregate social gain (gain to society from creation minus loss to society from monopoly control over uses).¹⁰⁰ As Fisher himself admits, his approach to analyzing fair use, at least with respect to some sorts of works, “is nearly coterminus [sic] with economic analysis of the copyright system as a whole.”¹⁰¹ Fisher likewise acknowledges that his approach is probably impractical in light of the informational burden it imposes on judges.¹⁰²

A different allocative justification, termed the “dual-grant theory” of fair use, was constructed by Professor Abraham Bell together with one of this article’s authors.¹⁰³ The dual-grant theory maintains that in designing our copyright system, Congress created two blocs of uses, allocating one to authors and the other to the public. Each group was given the uses which it values most. Accordingly, authors received “standard commercial uses,” while the public received “uses of highly dispersed social value,”¹⁰⁴ such as political speech, and uses that promote the pursuit of science, knowledge and truth.¹⁰⁵ By contrast to Fisher’s theory of fair use, which requires courts to make individual case-by-case determinations of fair use, the dual-grant theory calls on courts to make fair use determinations based on the category of uses to which the challenged use belongs. This, in turn, renders the approach a lot more practical but less precise compared to Fisher’s framework of analysis.

II. Fair Use as Licensing Leverage

As our discussion in Part I demonstrates, extant theories of fair use perceive it as a mechanism for bypassing the market and a substitute to owner-user bargaining. Importantly for the purpose of our analysis, all existing justifications of fair use focus exclusively on the courts. The underlying assumption in all four accounts is that fair use serves users exclusively in litigation or as a tool for overcoming negotiation breakdowns or high transaction costs that prevent negotiations from occurring ab initio.

The justification we develop in this Part focuses on the effect of fair use on enabling voluntary transactions between copyright owners and users. By contrast to prior justifications of fair use, we show that the main function of fair use is to facilitate bargaining, rather than replace it. We call this effect “the hidden function of fair use.” As we demonstrate, the fair use doctrine allows users to secure more favorable licensing terms from copyright owners. Importantly, our theory does not compete with any of the prior theories. Rather, it complements all four of them. This complementarity suggests that the effect of fair use is broader and deeper than previously believed.

A. The Theory

In their acclaimed bestseller on negotiation theory, Getting to Yes, Professors Roger Fisher and William Ury famously stress that “the relative negotiating power of two parties depends primarily upon how attractive to each is the option of not reaching agreement.”¹⁰⁶ They therefore submit that a sine qua non¹⁰⁷ for A’s bargaining advantage is that their benefit from a state of disagreement exceeds B’s, since, in such case, B would be willing to sacrifice more in order to reach an agreement, which diverts the terms of agreement formation in A’s favor. Normally, the party with the upper hand in a bargaining setting is the one who possesses a stronger BATNA.¹⁰⁸ To exemplify, suppose that A is interested in selling B a used car, which B values at $10,000. Realizing this, A can require up to $10,000 to secure an agreement between the two. Assume now that C offers a similar car for sale, in exchange for only $5,000. C’s entry affects the negotiations between A and B: the maximum price that B would be willing to pay for A’s car now drops to $5,000. C’s entry, by providing B with an alternative to negotiating with A, enhances B’s BATNA.

The fair use doctrine has a similar effect to that of C’s entry in the previous example: it elevates users’ BATNA and thereby confers upon them significant leverage in negotiations with copyright owners. A helpful way to see this effect of fair use is to think of it as a call option the law gives to users over copyrighted content with a strike price of zero if their use is found fair. Recall that a fair use finding means that a user is not only free to use copyrighted content without permission, but also that they are allowed to do so free of charge.

To illustrate, suppose that Ella, an artist, is interested in using Francine’s copyrighted work for a transformative purpose and contacts her to acquire authorization. Assume that Ella values Francine’s work at $8,000 and is willing to pay Francine up to this sum in order to secure authorization. If Ella were to use Francine’s work without permission, she would be sued for copyright infringement and a court would order her to pay Francine $10,000 in damages.¹⁰⁹

To see the effect of fair use on negotiation dynamics, consider first a world without fair use. In this world, users can only use copyrighted content permissively. Any attempt at bypassing the market would be remedied by damages, an injunction, or both. Under a legal regime that does not recognize fair uses, Ella has no feasible alternative to negotiating with Francine. Using the work without authorization is not a viable option from Ella’s perspective, as it represents a negative net value (-$2,000): while Ella would receive a benefit of $8,000 from her use, she ought to pay $10,000 in damages. Therefore, in economic parlance, Ella has no credible threat of using without authorization: both parties know that if negotiations fail, Ella will just have to forgo her planned use. Consequently, Ella is willing to pay any price up to $8,000 (her valuation of Francine’s work), eliciting a positive benefit, instead of zero. Francine, in turn, would take advantage of her monopolistic status as an exclusive rightsholder and accord by setting up an asking price of $8,000—the maximum Ella is willing to pay.¹¹⁰ In such a world, therefore, Ella confronts an asking price of $8,000, which means that Francine gets to pocket the entire bargaining surplus.

Now consider a world with fair use. Assume that, given the high transformativeness of Ella’s intended use, there is a 0.5 probability that a court will find the use fair. This means that if negotiations fail, Ella is better off using Francine’s work without authorization than with forgoing the use: using the work without permission provides her with a benefit of $8,000 while the expected costs are only $5,000 (0.5 × $10,000 = $5,000). Critically, the introduction of fair use changes the expected value of unauthorized use from -$2,000 to $3,000 ($8,000 – $5,000). The existence of the fair use doctrine, thus, dramatically empowers users vis-à-vis copyright owners. This implies that Ella’s threat of using Francine’s work without permission becomes credible: if Francine’s asking price would exceed $5,000, Ella would walk off the negotiation table and use the work without authorization. This is because any asking price that crosses this threshold makes her benefit from authorization less than $3,000, which should lead her to refuse to pay this amount and, instead, use the work without authorization. In such a world, therefore, Ella confronts an asking price of $5,000.¹¹¹ We summarize the results of our examples in Table 1, below.

Table 1. The Effect of Fair Use on Users’ Licensing Leverage
World	Ella’s Benefit from Using	Expected Cost of Unauthorized Use	Expected Value of Unauthorized Use	Ella’s Value of Not Using	Ella’s BATNA	Francine’s Asking Price
Without Fair Use	$8,000	$10,000	-$2,000	0	Non-existent	$8,000
With Fair Use	$8,000	$5,000	$3,000	0	Increased	$5,000

Assume next that Ella’s desired use is not only highly transformative, but a clear parody, thus representing a 0.8 probability of a fair use finding by a court. In that case, there is only a 0.2 probability that a court will find Ella liable for copyright infringement and award Francine damages. This, in turn, renders the expected value of unauthorized use for Ella even higher: $8,000 – 0.2 × $10,000 = $6,000. In light of this fact, Ella will never accept any asking price that surpasses $2,000. Realizing this, Francine would agree to license the work to Ella for $2,000. The results are summarized in Table 2.

Table 2. The Effect of Increased Fair Use on Users’ Licensing Leverage
World	Ella’s Benefit from Using	Expected Cost of Unauthorized Use	Expected Value of Unauthorized Use	Ella’s Value of Not Using	Ella’s BATNA	Francine’s Asking Price
With Increased Fair Use Probability	$8,000	$2,000	$6,000	0	Further increased	$2,000

The logic that underlies this finding is as follows. When facing a copyrighted work, any user entertains a trichotomous choice: (1) they may pay a licensing fee and use the work unhinderedly; (2) they may avoid using it altogether; or (3) they may refrain from licensing, use the work without acquiring authorization, and face the expected costs of a copyright infringement lawsuit. Users’ BATNA essentially depends on the feasibility of the third alternative. Herein lies the significance of fair use. Fair use bestows upon users the power to credibly threat to use copyrighted content without authorization and thereby leads rightsholders to adjust their asking price downwards. A world without fair use, by contrast, would allow any rightsholder to prevail in an infringement suit against any unauthorized user, which obviates the user’s third alternative. In such a world, users are left with options (1) and (2): they can either pay copyright owners the licensing fees they post or refrain from using the work altogether.

The fair use doctrine not only engenders alternative (3) but also makes it viable. As in the example discussed above, knowing that Ella reaps a positive expected value of $3,000 from unauthorized use, Francine realizes that for any asking price that exceeds $5,000 (leaving Ella with a benefit of less than $3,000), Ella has a credible threat of not taking the deal and using the work without permission.

B. The Model

Our theory can be generalized in a simple formal fashion. Let v denote a given user’s valuation of a given copyrighted work. Let A denote the owner’s asking price. A is the maximum licensing fee that the user finds attractive, i.e., the maximum price under which the user does not have a credible threat to engage in unauthorized use as a substitute for acquiring authorization. Denote by p the probability that a court will find that the user’s unauthorized use is protected by the fair use doctrine, and denote by D the damages the court is expected to award the rightsholder if fair use protection is not granted.

The owner’s asking price is given either by the user’s valuation of the work, v, or, if the user obtains a positive expected value from an unauthorized use (namely, if v – (1 –p)D > 0), by deducting this sum from v:

Note that in a world without fair use, p = 0. This means that if v < D, namely, as long as the court is expected to award damages that exceed the user’s valuation of the copyrighted work, unauthorized use is never a worthwhile alternative to licensing. The user thus possesses no leverage vis-à-vis the owner, and the latter will therefore require payment of v, the user’s valuation of the work, in order to grant authorization.

The fair use doctrine invariably increases the expected value of unauthorized use: raising the value of p from p = 0 to p > 0, which results in a higher p – (1 –p)D. Furthermore, for some threshold values of and , it increases v – (1 –p)Dto satisfy v – (1 –p)D > 0, and in such case, the existence of the fair use doctrine results in positive expected value from unauthorized use. From the user’s perspective, this means that v > (v – (1 –p)D), and thus A, their negotiated authorization price, decreases. Figure 1 exemplifies the relationship between A, the asking price, and p, the probability of fair use, for given values of v and D that uphold D > v.

Figure 1. The Effect of Fair Use on Copyright Pricing

This stylized model unfolds the technical bargaining mechanism that underlies our theory. Thus far, the accepted lore perceived fair use as offering users an effective bypass from licensing negotiations—but this statement can be confidently made only when fair use is granted with certainty (p = 1), which is hardly the regular case. The model emphasizes that even under the conventional setting where a fair use ruling is uncertain, the doctrine plays a critical role in the market for copyrighted content—the uncertainty translates into decreased asking price by monopolistic rightsholders, hence advantaging users’ bargaining standpoint.

We now turn to introduce evidence from licensing practices that supports our theory, and then we discuss the normative implications.

C. Illustrations

Our theoretical exposition of the hidden function of fair use finds support in real world cases. In this Section, we provide various examples that substantiate our theoretical predictions. These cases show that users are aware of the bargaining leverage they can get from the fair use doctrine and take advantage of its more favorable licensing terms.

Oracle and the Future of Licensing in Technologies

As noted, our theory likewise applies to providers of technological applications that generate a benefit to the public. A case in point can be found in Justice Thomas’s dissenting opinion in Google LLC v. Oracle America, Inc..¹¹² In support of his view that Google’s use of Oracle’s code was not fair, he noted that Google’s use created a reality where “device manufacturers no longer saw much reason to pay to embed the Java platform.”¹¹³ Justice Thomas proceeded to emphasize that:

[B]efore Google released Android, Amazon paid for a license to embed the Java platform in Kindle devices. But after Google released Android, Amazon used the cost-free availability of Android to negotiate a 97.5% discount on its license fee with Oracle. Evidence at trial similarly showed that right after Google released Android, Samsung’s contract with Oracle dropped from $40 million to about $1 million.¹¹⁴

The sharp decline in Oracle’s revenues from licensing its Java code provides a powerful illustration of the impact of fair use on market transactions. As the figures show, the bargaining leverage of Oracle in all of its licensing transactions critically depended on the licensees’ assessment of the strength of Google’s actual fair use claim and their own hypothetical fair use claims. For although Google was at the forefront of this legal battle, many other technology companies were similarly situated to Google insofar as their status as fair users.¹¹⁵ As the sentiment that Google would prevail gained purchase among technology companies, the revenues of Oracle sharply decreased, precisely as our theory predicts.

In analyzing the Supreme Court’s decision, commentators have noted that “[i]n the future, this decision may prompt more disrupters to use fair use as a shield in releasing new products or services that build off of older functional technologies, or otherwise influence negotiations as some potential licensees may find the value of certain functional code to be devalued by the Oracle.”¹¹⁶

Documentary Filmmakers

Copyright scholars tend to refer to the documentary filmmakers’ population as a primary beneficiary of fair use. Documentary filmmakers must incorporate prior works, many of which are subject to copyright protection. Hence, the fair use doctrine is critical to the operation of the industry. It may come as no surprise, therefore, that documentary filmmakers are cognizant of their ability to leverage on fair use in negotiations with copyright owners. A 2020 study by the Center for Media and Social Impact (“CMSI”) concerning the state of the documentary field¹¹⁷ indicates that 76% of all participating U.S. documentary directors and producers have utilized or leveraged the fair use doctrine in the making of their most recent film.¹¹⁸ This corresponds to other surveys, wherein 70% of the filmmakers have rated their understanding of fair use as “good or excellent,”¹¹⁹ while 73% of those with more than a decade of experience found fair use a “very useful” doctrine.¹²⁰ Likewise, an elaborate manual published by the Archive Valley company, which provides archival services for documentarists, introduces the underlying strands of fair use by explaining to authors that “if you meet the fair use guidelines . . . you can use footage for free,” but complements this straightforward statement by informing authors that “[e]ven if you know that you do want to license footage, knowing your rights about fair use can put you in a stronger negotiating position.”¹²¹

Finally, it seems that the leverage fair use bestows upon filmmakers changes the boundaries of the bilateral owner-user bargaining. Crucially, a strong fair use claim appears to pave authors’ way to efficiently negotiate with relevant third parties, too. In this regard, 95% of filmmakers have reported to persuade broadcasters regarding the applicability of fair use to their case when equipped with a lawyer’s letter.¹²² 99% responded identically when asked about negotiations with insurers.¹²³ This means that fair use not only leverages users when bargaining with rightsholders, but, as importantly, alleviates the entire chain of negotiations necessary for pursuing one’s artistic vision.

Parodies and Satires

The story of the musician “Weird Al” Yankovic, who is known for his humoristic commercial adaptations of popular music, provides yet another real-world example of the effect of ambiguity on broadening the population of users that can enjoy increased licensing leverage in light of fair use. Whether Yankovic’s spoof songs are in fact parodies¹²⁴ or satires¹²⁵ is a controversy that has never been, and probably never will be, settled, as Yankovic always acquires authorization from the owners of the rights to the original songs¹²⁶—be it because of industry norms¹²⁷ or in order to be on the “safe side” of copyright law. Interestingly, however, Yankovic’s practice has been described as “[l]icensing [w]ith [l]everage.”¹²⁸

Yankovic, whose parodic use adjoins satire, bargains for authorization under a fairly reasonable possibility of fair use protection. And Yankovic is not alone. Other artists who produce adaptations of copyrighted works have the same leverage that he does: they possess a colorable fair use claim.

It should be underscored that this broadening effect is not limited to musical adaptations. Rather, it applies to copyrighted subject matter categories, from videoclips to literary works. As importantly, the effect is not confined to adaptations that straddle the line separating parodies and satires. In the case of transformative uses, the effect is much broader. Consider, for example, the case of Gregg Gillis, better known as the popular D.J. Girl Talk, whose foremost artistic contribution has been described as “mak[ing] danceable musical collages out of short clips from other people’s songs.”¹²⁹ Gillis insists that his sampling work is protected by the fair use doctrine, and therefore, as opposed to Yankovic, he consistently refuses to solicit authorization from rightsholders to the original works.¹³⁰ Markedly, while many have accused Gillis of infringing upon their exclusive rights, a lawsuit has yet to be filed.¹³¹ In short, thus, it appears that both Gillis and his accusers believe that the musician enjoys a solid case, garnering a sufficiently high fair use probability that in turn nullifies the credibility of threats to sue. This translates into a reservation price of zero.

Warhol and the Future of Transformativeness

Another exemplification of the utilization of the transformativeness threshold into licensing leverage can be driven from Cariou v. Prince, in which the Second Circuit defined transformativeness as any alteration of the original’s “expression, meaning, or message,” finding that even appropriation art can be considered fair use.¹³² Under this expansive definition of fair use, any user who adapts a work can raise a fair use defense. Obviously, not every adaptation would be ruled fair. Similarly, not every appropriation artist will meet the same faith as Richard Prince. But every transformative user has bargaining leverage thanks to this broad definition.

As noted, however, the Supreme Court’s recent decision in Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith diverges from preceding rulings on transformativeness that advantaged users, and it seems to reallocate bargaining power between users and rightsholders.¹³³ Owing to Warhol, we might confront a shift toward a new era that introduces a new equilibrium in the market for copyrighted content.

Notwithstanding their absence from both popular discourse and academic scholarship on fair use, the Justices—both members of the majority and dissenters—were far from oblivious to this outcome. First, this understanding manifests in Justice Sotomayor’s articulation of the majority opinion. Justice Sotomayor first warns us that the broad interpretation of transformativeness would favor users with utterly disproportionate leverage over the original rightsholder. Upon recognizing that transformativeness “is a matter of degree,”¹³⁴ Justice Sotomayor maintains that:

[Holding for the Plaintiff] would potentially authorize a range of commercial copying of photographs, to be used for purposes that are substantially the same as those of the originals. As long as the user somehow portrays the subject of the photograph differently, he could make modest alterations to the original . . . and claim transformative use.¹³⁵

Justice Sotomayor’s words implicitly capture our theory, acknowledging the simple fact that the degree of transformativeness that suffices for the Court to bestow a fair use defense would shape market interactions between owners and users. Determining that changes reminiscent to those made by Andy Warhol are not significant enough to meet this threshold, particularly when the use is of commercial nature, Justice Sotomayor precludes users from leveraging a fair use claim and consequently—when viewed via an economic lens—submitting that their intended use does not deserve a reduced licensing fee.

More explicit are the statements made in Justice Kagan’s dissenting opinion. Justice Kagan distinctly highlights the pivotal role of the Supreme Court in allocating bargaining surplus between owners and users, noting that by refusing to side with Warhol’s deeds, the Court would frustrate users’ ability to license under favorable terms:

Still more troubling are the consequences of today’s ruling for other artists. If Warhol does not get credit for transformative copying, who will? And when artists less famous than Warhol cannot benefit from fair use, it will matter even more. . . . [A]s our precedents show, licensors sometimes place stringent limits on follow-on uses, especially to prevent kinds of expression they disapprove. And licensors may charge fees that prevent many or most artists from gaining access to original works.¹³⁶

The transactional role of the fair use doctrine and its effect on licensing negotiations has been properly identified and adequately considered. Despite that, the Court has upset the longstanding understanding of market actors, potentially requiring the market to form a new equilibrium that is less favorable to users. This leads us to conclude that the reallocation of bargaining power that we expect to witness in the near future—together with the inevitable increase of copyright pricing—is the Court’s fully conscious decision, rather than an unforeseen economic side effect.

III. Normative Implications

In this Part, we explain how the licensing leverage created by fair use ameliorates the efficiency losses and inequities associated with copyright protection. In particular, we demonstrate the doctrine’s ability to increase the number of voluntary transactions between copyright owners and users and to redistribute wealth from rightsholders to users.

A. Countervailing Allocative Inefficiencies

Being quintessential public goods, intellectual works cannot be supplied efficiently by the free market.¹³⁷ In the absence of a legal prohibition on copying, users would be able to copy expressive works with impunity. In such a world, the market price of copyrighted works would rapidly drop to zero, and so would the incentives to create and supply intellectual goods. Unable to recoup their initial investment in the creation of original content, authors may well decide to put their creative skills to rest. The need to maintain authors’ incentives to create warrants legal intervention that would grant them exclusive control over the distribution of their works.¹³⁸

Yet, the legal exclusivity copyright law bestows upon authors creates several costs. Chief among them is monopolistic pricing.¹³⁹ Two major effects are customarily attributed to monopolistic pricing: allocative inefficiency and distributional inequity.

The allocative inefficiency arising from exclusivity is well established.¹⁴⁰ Monopolistic pricing invariably creates a deadweight loss.¹⁴¹ This loss emanates from the fact that some consumers who were willing to pay the competitive price for the good or service would not be able to pay the higher, monopolistic price, and would simply forgo the good or service. As a result, certain welfare-enhancing transactions that would have occurred in a competitive market would not take place under monopolistic pricing.¹⁴² To illustrate, suppose that a good is offered by a monopolistic manufacturer for $20. Assume, however, that had the production process been perfectly competitive, the same good would have been offered for a price of $12. In this example, the deadweight loss is represented by the foregone transactions of the buyers who value the good at more than $12 but less than $20. As economist William McEachern points out, this phenomenon “is called the deadweight loss of monopoly because it is a loss to consumers but a gain to nobody.”¹⁴³

Perhaps even more worrisome than its adverse effect on allocative efficiency is the distributional inequities imposed by monopolistic pricing. Since the monopolistic price is higher than the competitive price, it works to transfer wealth from consumers to the monopolist. Monopolistic pricing, in other words, allows monopolists to capture consumer surplus.¹⁴⁴ To see this, let’s return to the above example, and consider a consumer who values the good at $22. Recall that the monopolist sets the price at $20, whereas in a competitive market, the price of the good would be $12. Under perfect competition the consumer would have paid $12 for the good, thus deriving a surplus of $10 ($22 – $12) from the transaction. Under monopolistic pricing the surplus of the consumer shrinks to $2 ($22 – $20), with the $8 lost to the consumer being transferred to the monopolist—all relative to a competitive market.

To understand how fair use remedies the allocative inefficiencies resulting from the exclusivity granted by copyright protection, consider a copyright owner who offers their work for a fixed license fee, which represents the profit-maximizing amount for them. As opposed to the general analysis unfolded in Part II, we now assume that, for whatever reason, price differentiation is impracticable from the rightsholder’s perspective.¹⁴⁵ Our theory submits that under such circumstances, the mere existence of fair use for certain users creates a positive externality for other users who have no fair use claim.¹⁴⁶ In this respect, the number of voluntary transactions may increase, decline, or remain unchanged, but what bears emphasis is that fair use avails the entire universe of users, including those users who would never persuade the court that their use is considered fair.

To see this counterintuitive outcome, suppose that the relevant users are five parodists and two satirists. For this example, assume that the parodists’ valuation of using the work is $10,000, while the satirists’ valuation is, say, $6,000.¹⁴⁷ For simplicity’s sake, assume that the damages in case of infringement are likewise $10,000. It is easy to see that if the monopolist has to determine a fixed, undifferentiated asking price, this price will be either $10,000 (thus authorizing only parodists) or $6,000 (thus authorizing all users, satirists included). The profit-maximizing price in that case is $10,000, since 5 × $10,000 > 7 × $6,000. Therefore, only parodists manage to obtain authorization from the owner, while satirists are precluded from using the work in light of their unwillingness to pay $10,000, neither as licensing fees nor as damage awards.

Things change dramatically, however, in a world with fair use. To begin, suppose any parodist holds 0.5 probability of obtaining a fair use ruling in their favor, whereas satirists have no chance of convincing the court that they are entitled to a fair use ruling. But note, while satirists will rarely enjoy a fair use ruling in their favor, they in fact manage to obtain user authorization in a world in which the fair use doctrine exists. In this world, the owner can never sell at $10,000: if they were to stick to this asking price, all parodists are incentivized to use the work without permission, as the expected value of using without authorization is $5,000 (0.5 probability multiplied by $10,000 in damages). Therefore, the two relevant pricing schemes are either $6,000 (authorizing only satirists) or $5,000 (authorizing all users). The

rightsholder is expected to set an asking price of $5,000 since 7 × $5,000 > 2 × $6,000.¹⁴⁸

If, on the other hand, any parodist holds 0.9 probability of obtaining fair use, this increased probability simply results in the rightsholder focusing exclusively on negotiations with satirists, setting an asking price of $6,000 as 7 × $1,000 < 2 × $6,000.

Interestingly enough, this example illustrates not our primary argument that fair use bolsters users’ bargaining power, but rather, that if rightsholders are subject to a fixed licensing fee, the doctrine may avail even users with zero probability of obtaining fair use protection. The hitherto undiscussed impact of fair use on copyright pricing allows them to enjoy copyrighted content they would not have in a world without fair use.

It should be noted that while parodists in the latter example will not be willing to pay the fixed price required by the rightsholder for authorization, they will manage to reach a favorable settlement if the copyright owner decides to sue them for infringement. Due to their enhanced probability of enjoying a fair use ruling, a settlement agreement is expected to feature extremely low payment on their behalf.¹⁴⁹

B. Redistributing Bargaining Surplus

In addition to improving allocative efficiency by increasing the number of users who enjoy owner authorization, fair use also has the effect of ameliorating the distributive distortions arising from copyright protection. As already noted, without fair use, rightsholders are positioned to capture the lion’s share of the users’ bargaining surplus by setting up prices that approximate users’ maximal valuations. In our previous example, even the five parodists who are willing to pay $10,000 for the right to use a work would have to fork over to the copyright owner—who sets an asking price of $10,000—all of their bargaining surplus.

This unfortunate outcome exemplifies monopolistic holdout. More specifically, monopolistic rightsholders manage to extract increased payments particularly from those users who consider the work extremely valuable. Pertinent examples include young technological entrepreneurs who are in need of particular copyrighted software in order to advance their startup initiative, or documentary filmmakers at the dawn of their career who are required to negotiate with a myriad of rightsholders in order to actualize their cinematic vision.

These users, and others, confront a significant barrier erected by copyright owners, who strategically hold out to extract exorbitant licensing fees. On our theory, the fair use doctrine counterbalances such strategic behavior. In addition to improving licensing terms for all users who possess a fair use claim (and, as exemplified earlier, even for those who do not), the doctrine helps reduce monopolistic holdouts in negotiations. This insight is not merely theoretical. It can be seen at work in the aftermath of the Supreme Court’s recent decision in Google LLC v. Oracle America, Inc..¹⁵⁰ The ruling has won commentators’ accolades for lowering entry barriers for smaller startup initiatives: consistent with our analysis, the ruling eroded the strategic incentive of powerful software owners to prevent smaller competitors, seeking interoperability with industry standards, from entering the market.¹⁵¹ Smaller developers can currently draw on the Google decision to acquire what is needed from a preexisting application programing interface, to provide consumers an innovative, transformative technological program.¹⁵²

The Warhol decision withheld the user-advantaging trend that bestows fair use protection even on mildly transformative users. Ample criticism has been directed against the Court since the decision was handed, primarily spotlighting the confinement of artistic freedom inevitably imposed by their ruling.¹⁵³ But one crucial aspect keeps eluding the commentators. As our analysis shows, the Court’s decision not only restricts artistic expression, but also diminishes the bargaining power of users in negotiations with rightsholders and thereby increases the price of copyrighted content. A narrow view of fair use makes one miss this aspect.

C. The Role of Damage Awards

To equip users with the full force of fair use, it is imperative to ensure that other copyright law doctrines may chill the effect this Article identifies. In particular, exaggerated damage awards may undesirably offset the economic virtues of the fair use doctrine. This is an especially grave concern in the copyright domain, as copyright owners typically sue for statutory damages. As Ben Depoorter noted, “[o]nce infringement has been established, a plaintiff may elect a statutory damage award. In doing so, 17 U.S.C. § 504(c) of the 1976 Copyright Act relieves the copyright holder from the burden of providing any evidence whatsoever of actual harm. Among developed Western democracies, the U.S. copyright statutory framework is exceptional.”¹⁵⁴ Indeed, under the current regime, statutory damages for watching an illegal livestream may amount to $150,000.¹⁵⁵ Oren Bracha and Talha Syed have similarly pointed out that “[t]he most troubling [aspects of statutory damages] are those cases in which massive supracompensatory damages are inflicted on ordinary individuals who are unable to spread the impact of such awards through limited liability or dispersed ownership.”¹⁵⁶

Apart from the usual pitfalls scholars associate with excessive copyright damages—namely, overdeterrence and suboptimal use of copyrighted works—we wish to emphasize its adverse impact on the licensing leverage that fair use provides to users. To see this, return to the main example on which we draw to convey our conception of fair use. Assume that rather than $10,000, Francine is expected to enjoy $50,000 in damages if she prevails in court. In this scenario, as in the case in a world without fair use, Ella is devoid of any feasible alternative to negotiations: the value of unauthorized use to her is 0.5 × (-$50,000) + $8,000 = -$17,000, which means that even in a world with fair use, she has no credible threat of using Francine’s work without authorization. Under these circumstances, the fair use doctrine does not bestow upon Ella a credible threat (and licensing leverage thereof) as long as her probability of prevailing in court falls short of 0.84.

By the same token, courts’ commitment to relatively modest awards is expected to significantly enhance the impact of fair use on users’ bargaining standpoint. Modest awards imply that even a user with a low probability of persuading the court of their entitlement to fair use will enjoy licensing leverage in negotiations with copyright owners. Taking this idea to the extreme, suppose that Francine is expected to be awarded $5,000 if she prevails in court. Recall that Ella values the use of Francine’s work at $8,000. In this case, Francine’s a-priori asking price is reduced irrespective of fair use: even without the doctrine, Ella will not be willing to pay any licensing fee that surpasses $5,000, since this would make unauthorized use more beneficial. Under the fair use doctrine, this price will drop even further.

Excessive damage awards imply that even users with a strong fair use claim might rationally refrain from an unauthorized use to avoid the insurmountable expected loss in future copyright litigation. We, therefore, call upon judges and legislators to determine damages in copyright cases when accounting for the possible erosion of users’ negotiation standpoint. In other words, we submit that by exhibiting restraint in the determination of damages, decisionmakers can bolster the hidden function of fair use; yet, by superfluously awarding damages to rightsholders, they might nullify this function in its entirety.

D. The Latent Virtue of Ambiguity

Our theory of fair use also puts ambiguity in a different light. A recurring criticism of the fair use jurisprudence concerns its uncertainty. The ambiguity of the fair use doctrine has been accused of overdeterring individuals from using copyrighted content. As James Gibson argued, even users with a strong fair use claim may rationally elect to license copyrighted content or refrain from using it altogether in order to avoid the inherent uncertainty that attends the fair use doctrine.¹⁵⁷ Losing on a fair use claim comes at a high price: a court may order the unlucky defendant to pay high statutory damages. Anticipating this, users may rationally choose not to assert their fair use claim in court, and instead, buy a license from the copyright owner. Gibson termed this dynamic “rights accretion.”¹⁵⁸ Furthermore, he described it as an ongoing process by which copyright holders broaden the scope of their protection.¹⁵⁹

Our theory points to a potentially countervailing effect. The uncertainty that characterizes the fair use doctrine may actually help users in negotiations with copyright owners. This is because the inherent ambiguity embedded in the fair use doctrine enhances the licensing leverage of users, relative to a bright-line-rule fair use doctrine. To demonstrate the argument, we first provide a brief description of the ambiguity inherent in extant fair use jurisprudence, and then we point out its advantages compared to a world in which the fair use ambiguity does not exist.

In its current form, the fair use doctrine is an open-ended standard, rather than a rule.¹⁶⁰ As the time-honored distinction suggests, rules are well-defined legal commands that offer a precise definition of a proscribed action or conduct. Conversely, standards are legal provisions that employ a more general, obscure description, thus subjecting the lawfulness of a particular behavior to judicial discretion ex post. While standards allow the legal system to function with greater flexibility, rules provide actors with a greater degree of clarity and certainty. As has been noted before:

[T]he ex post guidance provided by courts is often confined to the specifics of the case at hand and does little to clarify the realm of legitimate behavior for other actors. The unpredictability associated with standards affects not only wrongdoers, but also law-abiding citizens who wish to act in accordance with the law but cannot readily discern what acts are permissible.¹⁶¹

As noted above, fair use is a prototypical standard.¹⁶² Fair use jurisprudence is ridden with terms that may purport to be synonymous to fair use, but in fact provide little or no ex ante clarity with respect to their applicability to a given case. Parodic use, for example, was tautologically defined by the Supreme Court as anything that may be reasonably perceived as a parody.¹⁶³ The concept of transformative use has likewise eluded any consistent definition,¹⁶⁴ and the “public benefit” test, to which courts increasingly reference, is inherently uninformative.¹⁶⁵ All of this implies that even when the user possesses a strong claim, fair use protection is hardly guaranteed. Nor can it be completely ruled out even in the presence of users with relatively weak claims. In its current conception as a quintessential standard, therefore, the user and the owner confront a given probability that fair use will be recognized. But as noted earlier, this probability—created by the malleable, open-ended environment in which fair use determinations are being made—may render a user’s threat of resorting to unauthorized use credible, which in turn creates a favorable bargaining framework that allows them to secure authorization with reduced licensing fees.

To see this, compare the current fair use regime with an alternative hypothetical regime in which fair use is governed by rules, namely, all permitted uses are well-defined on an ex ante basis. This alternative design leaves no room for uncertainty as to the defense’s applicability: the probability that fair use will be recognized in a given case is either zero or one. While allowing clarity and complete reliance, we wish to stress the adverse effect that such dichotomy carries on parties’ bargaining. Begin with cases wherein fair use will be conferred upon users with complete certainty. It is evident that whenever a fair use ruling is guaranteed, users are unwilling to pay for licensing, and negotiations will never take place. On the other hand, if there is no chance for a user to enjoy a fair use judgment, they are essentially confronting a world without fair use, hence the doctrine fails to bestow licensing leverage upon them.

This comparison of a standard-based versus rule-based fair use illuminates the surprising virtue of the current design of the fair use doctrine. To be sure, each user individually, if asked, would say they prefer to obtain fair use with certainty, but the actual comparison that needs to be considered here is whether, behind a veil of ignorance, the right to use copyrighted content without permission should be allocated with certainty to a distinct, small class of users, or with uncertainty to a class of a broader scope. Users may reasonably subscribe to the latter. As noted earlier, a well-defined fair use doctrine not only allows fewer users to invoke the right in court, but more importantly, it also deprives all the rest from the licensing leverage they would have enjoyed under uncertain fair use, thereby subjecting them to the choice of paying the monopolistic asking price or avoid using.

Indeed, the current design of fair use facilitates a bargaining framework in which users enjoy discounts in asking prices on account of the possibility of a future fair use judgment in their favor. Conversely, a clearly and fully specified fair use doctrine will have no effect on owner-user bargaining: it advances fair use as a mere substitute to negotiations, rather than a complement. If parties foresee a fair use ruling with probability 1, bargaining will not take place; if a fair use judgment is not feasible, then parties essentially reside in a world without fair use, and the doctrine does not therefore impact their negotiated licensing terms. Either way, predicating the fair use doctrine as a system of well-defined bright line rules precludes its function in assisting users.

This insight corresponds to a handful of scholarly works that have stressed the merit of standards, compared to rules, at facilitating trade. The 1995 works of Jason Scott Johnston,¹⁶⁶ and Ian Ayres and Eric Talley,¹⁶⁷ have independently demonstrated that when it comes to bargaining over an entitlement, uncertainty is bliss. Both articles showed that standards have the effect of splitting entitlements between parties, which thereby enhance bargaining and negotiations. When entitlements are split, each party can buy or sell their share of the entitlement to the other.¹⁶⁸ The same is true of fair use. Fair use has the effect of splitting rights in expressive content between rightsholders and potential fair users.¹⁶⁹ Accordingly, in negotiations between them, the copyright owner sells to the potential fair user the right to use the work while the potential fair user sells away their privilege to assert fair use.

Interestingly, fair use’s ambiguity not only aids users with a strong fair use claim, but it also broadens the scope of user population that enjoys licensing leverage. To see this, consider the widely accepted distinction between parodies and satires, which prevails since Campbell.¹⁷⁰ Traditionally, it has been argued that this distinction favors parodists by exempting them from any authorization requirements, while disadvantaging satirists who enjoy no similar privilege.¹⁷¹ But this argument, too, overlooks the hidden benefit that users at large—parodists and satirists alike—can reap from fair use. To see this, it is necessary to take a step back and picture a world without fair use protection granted to parodies. In such a world, the author of the original work holds the upper hand by exercising complete control over the licensing process and can exclusively decide which subsequent works to authorize. The vestment of fair use privileges on parodies spills over to hybrid works that combine critical commentary on the original work as well as more general statements that amount to satire.¹⁷² This, in turn, means that satirists, too, reap some benefit from fair use. The effect of fair use on satirists is never zero. Because there is no clear distinction between parodies and satires, the fair use doctrine—on account of this very ambiguity—bestows bargaining leverage not only on parodists, but also on satirists.

Conclusion

In this Article, we uncovered a crucial, yet overlooked, function of the fair use doctrine: the empowerment of users in negotiations with copyright owners. The fair use doctrine endows users with a credible threat to leave the negotiation table and use the work without permission. This credible threat allows users to force copyright owners to lower their licensing fees, thus enhancing the total number of uses and increasing the share of the bargaining surplus kept by users. Our theory maintains that the principal effect of fair use is to facilitate market transactions involving copyrighted content, and not to bypass them as other theorists have suggested. Our analysis thus shows that the impact of fair use in the copyright domain is far greater than previously thought. Importantly, our theory of fair use is fully consistent with preexisting scholarly contributions; it complements them, not competes them. The addition of our theory to the existing literature on fair use demonstrates the full prowess of fair use in protecting users interests and needs in the copyright world.

98 S. Cal. L. Rev. 761

Download

Roy Baharad *

* Doctoral Candidate, University of Chicago Law School.

Gideon Parchomovsky†

Robert G. Fuller, Jr. Professor of Law, University of Pennsylvania Law School; Edward Silver Professor, Faculty of Law, The Hebrew University of Jerusalem. We thank Shiri Alon, Clark Asay, Ian Ayres, Shyamkrishna Balganesh, Oren Bar-Gill, Stefan Bechtold, Abraham Bell, Bari Britvan, Mala Chatterjee, Melissa Eckhause, Shira Ephron, Kristelia Garcia, Yehonatan Givati, Ehud Guttel, Thomas Haley, Lital Helman, Laura Heymann, Justin Hughes, Adi Libson, Peter Menell, David Nimmer, Shmuel Nitzan, Sarah Polcz, Matthew Sag, Peter Siegelman, Christopher Sprigman, Alex Stein, Alfred Yen, Eyal Zamir, and participants at the American Law and Economics Association Annual Conference and at the Seventh Copyright Scholarship Roundtable at Columbia Law School, for valuable comments and suggestions. We are especially indebted to Joseph Fishman for his enlightening remarks. For excellent research assistance, we thank Ron Bechar, Guy Cohen, Rachel Hujsa, Ruoyu Ji, Noam Kozlov, Ariel Melitz, Jeremy Rapaport-Stein, Ido Schlesinger, Yoav Stoler, Jude Taragin, Amir Tzur and Miriam Weinstock. Parchomovsky thanks the Israel Science Foundation for research support.

Mind the Gap(s): Mitigating Harassment in a Post-#METOO Workplace

July 2025July 2025Jamillah Bowman Williams,* Elizabeth C. Tippett† & Anu Ramdin‡

In a post-#MeToo workplace, harassment remains pervasive, and harassment law still fails to provide protection for the harms experienced by many workers—particularly those in the most vulnerable jobs. Even when reform efforts are introduced through legislation, courts, and agency guidance, it often does not provide greater power, autonomy, and dignity to women in ways that would more meaningfully protect them from workplace abuse. We are the first to create a database of state legislation, including over three thousand bills, that allows us to empirically analyze the extent to which lawmakers comprehensively address harassment following the rise of the #MeToo movement. We assess comprehensiveness by examining how responsive legislation is to existing gaps in legal protection during the five years following the 2017 tweet that took #MeToo activism global, relative to the 2016 baseline period. We found that states introduced a wide breadth of reforms to combat harassment and gender inequality, including some changes that address longstanding gaps in legal protection. Gaps persist, however, and in some cases worsened post #MeToo. Going forward, reform efforts by state legislatures—and all legal stakeholders—will prove most effective if they move away from narrow conceptions of sexual harassment and follow the voice of workers, pursuing a broad, multi-layered agenda around gender equity that is responsive to the realities of our evolving workplace and society. This is especially true in 2025 as the Trump administration redirects the EEOC’s resources towards attacking diversity, equity and inclusion measures.

Introduction

Tarana Burke coined the phrase “Me Too” in 2006 in support of Black women and girls of color who had survived sexual violence, encouraging them to share their stories despite the many pressures they faced to remain silent.¹ Social media galvanized the movement in 2017 after Alyssa Milano took to Twitter, posting: “If you’ve been sexually harassed or assaulted write ‘me too’ as a reply to this tweet . . . ‘we might give people a sense of the magnitude of the problem.’ ”² The #MeToo³ hashtag was used 19 million times between 2017 and 2018.⁴ Sustained digital participation increased the movement’s visibility and maintained its momentum beyond the boom and bust trend typically observed on social media.⁵ Online engagement surrounding the #MeToo movement also served as a catalyst for offline action, evolving into highly publicized protests, employee walkouts, and historic strikes.⁶ For example, in 2018, McDonald’s employees organized a historic multi-state strike against the company’s sexual harassment policies.⁷ Protesting workers wrote “#MeToo” on signs and covered their mouths with tape, bridging the gap between social media activism and traditional means of protest.⁸ After union-led campaigns and employee-staged walkouts, prominent companies in industries from technology to hospitality changed existing workplace policies such as mandatory arbitration.⁹

Subsequently, the hashtag #TimesUp, referencing the TIME’S UP Legal Defense Fund (“TULDF”), was created to solve the issues #MeToo revealed.¹⁰ TULDF sought to support women of color and low-wage women in the workplace, raising over $24 million to connect these women to attorneys and media specialists.¹¹ The movement also brought an increase in enforcement activity at the Equal Employment Opportunity Commission (“EEOC”), the government agency responsible for enforcing workplace discrimination law. In 2018, the EEOC reported that sexual harassment charges were up nationwide—the first increase observed in a decade.¹² In 2018, the EEOC received 7,609 sexual harassment charges compared with the 6,696 charges received in 2017.¹³ The EEOC recovered almost $104 million more for those with sexual harassment claims between 2018 and 2021 than between 2014 and 2017.¹⁴ The agency capitalized on #MeToo momentum by increasing lawsuits to enforce sexual harassment law and holding more employers accountable.¹⁵

This widespread activism also influenced legislatures. In the five years after #MeToo went viral, thousands of bills were introduced—mainly at the state level. These bills covered a wide range of topics related to gender equity in the workplace, including harassment training, nondisclosure agreements, forced arbitration for harassment claims, pay equity, and leave law. This surge of legislative activity was aimed at eliminating harassment and addressing gender equity in the workplace.¹⁶

While it was clear that legislatures were becoming more attentive to gender equity following the increased activism, it remained unclear whether the new reforms were getting to the root of the issues that cause harassment and addressing the legal gaps that allow harassment to persist. #MeToo media coverage tended to focus on certain victims—namely, the white women in the entertainment industry and the nondisclosure agreements (“NDAs”) that kept them silent.¹⁷ We argue that a more holistic approach is required both for progress over time and for reaching a broader range of working women. This is not unlike the “Swiss Cheese Model” adapted for use to control the spread of infection during the COVID-19 pandemic.¹⁸ The Swiss Cheese Model of risk reduction, developed by James Reason in the 1990s, visually demonstrates how a variety of strategies and actors can work together to reduce risk of harm.¹⁹ In the context of harassment, no single intervention such as restricting NDAs or increasing training can prevent all harassment, but multiple interventions can make a lasting difference. This is particularly true when the approach aims to reduce gendered power dynamics that make harassment more likely.²⁰

To empirically analyze the extent to which lawmakers were comprehensively addressing harassment and gender equity following #MeToo, our team of researchers and lawyers at Georgetown University collected and coded over 3,000 federal and state bills introduced from 2016 to 2022, some of which passed into law. We also analyzed how timing, geography, and political factors shaped the legal activity following #MeToo, relative to the 2016 baseline period. Our dataset is defined broadly and includes all harassment-specific legislation, as well as legislation addressing other gender equity issues including: pay equity, expanded coverage under Title VII of the Civil Rights Act of 1964 (“Title VII”), leave and accommodation, and occupational protections that address intersectional subordination. We found that the variety of legislation introduced by state legislators was consistent with the Swiss Cheese Model of risk reduction; however, there was wide variation across states.

Figure 1. Swiss Cheese Model for Reducing Harassment

This Article focuses primarily on state legislation because virtually all reform activity during the sample period occurred at the state level. Harassment law and gender-based reforms more broadly tend to follow a similar pattern of state-led rights protection and enhancement due to the gradual trend of federal courts contracting civil rights in the workplace among other venues. Federal lower courts, in particular, have increasingly issued employer-friendly Title VII and Americans with Disabilities Act (“ADA”) decisions that have been both formally and informally codified into law.²¹ Congressional gridlock has also led to a standstill with respect to anti-harassment legislation.²² Moreover, extreme partisanship in both Congress and the judiciary diminishes the likelihood of relying on Congress to act as a check when courts roll back rights, as they have in the past.²³ Only six federal reforms related to harassment and workplace gender equity passed during the five years following #MeToo; of those, only two had system-wide effects: the Ending Forced Arbitration of Sexual Assault and Sexual Harassment Act of 2021 and the Speak Out Act of 2022. By contrast, states have been a hotbed of legislative activity.²⁴

State legislation continues to play a central role in law reform, particularly under a second Trump administration that has signaled its hostility to transgender rights and diversity and inclusion programs. For example, Trump’s Executive Order 14168 declared that “ ‘sex’ is not a synonym for and does not include the concept of ‘gender identity.’ ”²⁵ Although the Order does not supersede the Supreme Court’s 2022 Bostock v. Clayton County decision recognizing sexual orientation and gender identity as sex discrimination under Title VII,²⁶ the Executive Order attempts to rescind the EEOC’s 2024 harassment guidance regarding gender identity.²⁷ In this context, the Trump administration seems more likely to roll back available civil rights coverage and enforcement than engage in meaningful reform.

Thus, states now, more than ever, remain important “laboratories of democracy,” a term Justice Brandeis coined in 1932, for achieving nationwide rights’ protection.²⁸ A close look at state legislative activity reveals the state-specific innovation and experimentation apparent in the variety of bills introduced. Nevertheless, the path from state-specific legislative innovation to systemic federal reform is not linear. Rather, broader effects depend on state-specific efforts influencing a combination of players from different arenas, including state courts, state agencies, federal courts, executive action, and even voluntary action by private entities.

We hypothesized that post #MeToo, the most significant activity at the state level would focus on harassment training and NDAs, which were scrutinized in the extensive media coverage. While high-profile support for these kinds of reforms did catch the attention of state legislators, with many bills introduced and passed, state legislators also introduced and passed many bills addressing systemic issues such as pay equity, leave law, and reforms to fill gaps in protection left in the legal landscape prior to #MeToo.

Although media coverage of the #MeToo movement may have overlooked many of the biggest and more systemic obstacles to workplace gender equity—such as lack of coverage under anti-discrimination law, employer retaliation, Title VII interpretation, and underenforcement—many states began to tackle these issues. We argue that states must continue this momentum not only to mitigate harassment more broadly, but also to address intersectional issues of economic precarity and racism that make harassment more frequent and particularly harmful to low-wage workers, who are disproportionately women of color.²⁹ For example, while women of color face higher rates of harassment due to compounding systems of subordination, most of the legislation proposed only strengthens protection for claims of sexual harassment or assault, and not intersectional subordination based on sex and race that is a common experience for women of color.³⁰ As a result, those experiencing intersectional harassment or discrimination based on multiple protected characteristics, including race, ethnicity, or national origin, are left vulnerable.³¹ Legal remedies that fail to consider experiences unique to women of color or other groups with multiple marginalized identities will ultimately fail to root out discrimination and harassment for those individuals, leading to underenforcement.

Nonetheless, our data reveals a broad array of state bills over a sustained period of time, which is a promising development. Complex social problems like harassment have numerous causes, and no single reform will prevent all instances of harassment, encourage employers to respond appropriately, or ensure a just legal remedy for the harm. We expect the net effect of the legislation, along with the social and attitudinal changes resulting from the #MeToo movement, to be positive and lasting. Not all legislation, however, had an equal chance of success to reduce broader trends of harassment. A close examination of the proposed and enacted legislation presents a mixed picture, which we discuss in depth in the empirical analysis that follows.

This Article is organized as follows: Part I explains how and why harassment persists today by looking at its historical origins, the current landscape, and how harassment relates to a growing trend of workplace violence. In Part II, we will discuss the specific gaps in legal protection that allow harassment and gender disparities to persist. In Part III, we offer original empirical analysis that illustrates trends in the anti-harassment and gender equity legislation that emerged during the five years post #MeToo. Finally in Part IV, we discuss the implications of our data for courts, lawmakers, and agencies, including what is still needed to mitigate workplace harassment post #MeToo.

I. The Persistent Nature of Workplace Harassment

This Part provides an overview of the underlying social determinants that yield high rates of workplace harassment over time.

A. The Historical Origins of Harassment

As various scholars have argued, harassment is not principally about sexual desire, but rather a reflection of the distribution of power within the particular workplace and society more broadly.³² It is a form of social behavior that reflects the dominant group’s—in this case men’s—ability to exclude, marginalize, or dominate less powerful individuals within the workplace.³³ It can often function to undermine the competence and confidence of marginalized groups and interfere with work performance, while also setting the norm of what is accepted or acceptable behavior.³⁴ Even harassment that is motivated in whole or in part by sexual desire reflects the perpetrator’s power to impose his demands and desires on others who are poorly positioned to rebuff those demands without consequence.³⁵

Harassment therefore serves as a mirror for systemic hierarchy, within the microcosm of the workplace and within the culture and larger historical context of a nation. The subordination of women was a central feature of nineteenth and twentieth century law and work arrangements. Women lost their legal personhood upon marriage through a principle known as coverture, wherein they could only acquire property or sign contracts through the legal personhood of their husbands.³⁶ Women did not secure the constitutional right to vote until 1920.³⁷ Many factories in the early decades of the twentieth century were explicitly segregated by gender, with women and men working in separate departments at separate pay rates, eating in separate cafeterias, and even entering and exiting from different entrances or on separate schedules.³⁸ Until 1974, banks could deny loans and credit cards to married women unless their husbands co-signed the application.³⁹ The legal and social foundation of the United States is one where men have power over women, are superior in status, and are free to control them.

America’s history of slavery, segregation, and white supremacy is also inextricably intertwined with the history of work, which shapes the experiences of Black and other racialized women.⁴⁰ Slavery was in part a workplace hierarchy predicated on violence and white supremacy.⁴¹ Black women were chattel assets, whose purpose in the economy and social order was to labor and serve white slave masters and their families.⁴² White slave masters frequently raped them, bred them to produce more workers, and subjected them to physical and psychological terror to maintain control.⁴³ After slaves were emancipated, segregation became a workplace reality, with Black women being relegated to the most menial and low-wage jobs.⁴⁴ Title VII offered reform, but it did not fully dismantle job segregation and workplace hierarchies predicated on white privilege.⁴⁵

Immigrant workers in the United States have also faced shifting, yet persistent, forms of discrimination. Although immigrants in the nineteenth and early twentieth century received privileged treatment compared with Black workers,⁴⁶ there is a long history of discrimination and subordination of immigrant populations, particularly Latino and Asian Americans. These include, for example, the Chinese Exclusion Act⁴⁷ and the internment of Japanese Americans during World War II, which kept Asian women workers excluded, marginalized, and exploited.⁴⁸ Continued subordination of both Black and immigrant women is also evidenced by the exclusion of domestic and agricultural workers from the 1938 Fair Labor Standards Act and the 1935 National Labor Relations Act.⁴⁹ These positions are overwhelmingly occupied by immigrant women and Black women, and many are still excluded from these legal protections today, making them more vulnerable to discrimination and abuse. Further, while Title VII prohibits discrimination on the basis of national origin, employers are permitted to discriminate against workers on the basis of immigration status.⁵⁰

Workers who belong to more than one subordinated group—such as Black women, immigrant women of color, or Black immigrant women—face compounded marginalization, as Kimberlé Crenshaw originally argued in her landmark article on intersectionality.⁵¹ Such discrimination is also replete within the historical record, in which a combination of race and sex-based discrimination operated in tandem. For example, the legislative history of Title VII suggests “sex” was included as a protected category based in part on arguments that protecting race, but not sex, would give Black women an advantage over White women.⁵² Likewise, during debates in the 1970s over whether to finally include domestic workers in the federal minimum wage law, opponents painted the work itself—performed primarily by women of color—as undeserving of minimum wage, while the White housewives who employed them were depicted as unqualified to calculate wages and hours.⁵³

Throughout United States history, women and people of color have also been vastly underrepresented in Congress, as well as within state legislatures.⁵⁴ Underrepresentation of these groups as lawmakers—both now and historically—reinforce hierarchies and subordination over time, by affecting which types of bills get introduced and passed, and which end up on the cutting room floor. Representation can shape values, priorities, and debate around who is deemed worthy of protection, in what ways, and whether legal interventions are needed to reduce inequality. Men—specifically White men—are also markedly overrepresented in the judiciary, where the laws are enforced.⁵⁵ The life experiences and perspectives of the judiciary affects their approach to cases and the parties with whom they identify. Biases influenced by one’s social position can then get baked into precedent, which later constrains and influences subsequent rulings.⁵⁶ Thus, lack of representation in both legislatures and courts can facilitate and exacerbate gaps in the law, ultimately failing to provide accountability and leaving women workers vulnerable to exploitation and abuse.

B. The Current Landscape of Workplace Harassment

Current national statistics reflect the continued influence of these historical patterns of subordination. Hispanic women earn fifty-seven cents for every dollar earned by White, non-Hispanic men.⁵⁷ In 2020, Black women earned sixty-four cents for every dollar earned by White, non-Hispanic men.⁵⁸ Over time, this gap fuels income and wealth disparities; Black women are estimated to lose nearly one million dollars over the course of their careers.⁵⁹

Relatedly, broad societal hierarchies and unevenly distributed power within organizations also affect workplace culture, behavior, and policies.⁶⁰ Power can take many forms; however, economic power or lack thereof tends to be at the foundation. Thus, when one’s labor is devalued—as occurs most substantially with Black and Hispanic women wage gaps compared with White, non-Hispanic men—this fundamentally affects the bargaining power one has with respect to protecting themselves from harassment at work.⁶¹ Economic disparities in bargaining power are compounded by continuing racism and sexism in low-wage industries where women of color are overrepresented.⁶²

The EEOC has published a list of workplace conditions that increase the likelihood of harassment. The vast majority of the conditions noted are common features of low-wage industries.⁶³ For example, low-wage and precarious industries such as hospitality, food service, modeling, agriculture, construction, and custodial work often: (1) lack effective policies and procedures; (2) have minimal to no oversight; (3) require working in isolation; and (4) prioritize customer and client well-being over the worker.⁶⁴ It is thus unsurprising that the women working in these industries experience the highest rates of harassment, among other forms of workplace violence.⁶⁵ For example, surveys have found that 65% of casino workers reported unwanted touching by guests, while over half of restaurant workers reported sexual harassment as occurring on a weekly basis.⁶⁶ One casino worker described how the harassment was intertwined with her wages, stating that a guest wanted to “put the tip on [her] ass” and took back the tip when she refused.⁶⁷ Another worker, this time at a hotel, described having to jump over the beds to escape the room of a guest who exposed themselves to her while requesting shampoo.⁶⁸ COVID-19 has only worsened these rates; one nationwide study found that 78% of surveyed restaurant workers reported an increase in hostility and harassment as they enforced COVID-19-related policies.⁶⁹

Harrowing accounts of harassment are also extremely prevalent among domestic workers and farmworkers, who are among the least protected due to the working conditions and the historical factors discussed above. Studies have found that 48% of domestic workers have had clients expose themselves, while 80% of farmworkers have experienced some form of sexual violence while at work.⁷⁰ In many of these cases, low-wage workers’ bargaining power is further diminished by language barriers or distrust of government agencies due to immigration status.⁷¹

Gender and racial hierarchies also shape who has access to leadership positions that hold power in the workplace and in government. For example, women and people of color remain vastly underrepresented within business leadership ranks.⁷² These statistics remain stubbornly sticky despite nearly sixty years of anti-discrimination law under Title VII. Reviewing decades of national employment data, sociologists Kevin Stainback and Donald Tomaskovic-Devey found that Black men and women, as well as White women, made far fewer gains in managerial jobs than in professional jobs.⁷³ White men retained a disproportionate share of managerial jobs, which these authors attributed to continued White male advantage that led them to be “pushed up in organizational hierarchies.”⁷⁴

Women also face gender-based harassment even when they break social and historical norms and move up to the top of the power hierarchy. In these cases, social and historical hierarchies are disrupted, and men resist and engage in demeaning behaviors to restore the status quo. Sociologist Raewyn Connell might characterize the problem in terms of “hegemonic masculinity”—“practice[s] . . . that allowed men’s dominance over women to continue,” that “embod[y] the currently most honored way of being a man” and “requir[e] all other men to position themselves in relation to it,” and that “ideologically legitimat[e] the global subordination of women to men.”⁷⁵ This type of harassment is not necessarily sexual in nature and can consist of taunting, insubordination, and use of demeaning slurs and images. This is less discussed than the harassment that is driven by sexual desire and targets low-status women. However, at its core, it is driven by systemic power hierarchies that exploit women or justify their differential treatment, similar to other types of harassment.⁷⁶

A poignant example are the numerous accounts of workplace harassment in law enforcement and the military. In 2018, it was reported that almost 25% of women in active-duty military experienced sexual harassment while in the military.⁷⁷ For women veterans, the percentage who reported experiencing sexual harassment rises to 80%, suggesting underreporting may be a significant problem for the military.⁷⁸ One particularly horrifying example comes from a female Navy lieutenant. In 1992, she went public with a personal story of being sexually assaulted by “the gauntlet,” which was described as a “nightly, coordinated, and systematic sexual assault of women who found themselves on the third floor of the hotel.”⁷⁹ High rates of harassment are also commonplace in law enforcement. One nationally representative survey of law enforcement officers found that 71% of female law enforcement officers have experienced sexual harassment or sexual violence in the workplace.⁸⁰ These male-dominated environments are high-risk, as women are devalued and targeted for breaking gender stereotypes and hierarchies of power.

C. Harassment as a Workplace Hazard

Based on historical and current social realities, we argue that harassment is a workplace hazard that is disproportionately distributed to women—particularly low-wage workers and women of color—creating a health and safety issue with civil rights and economic justice issues at the core. As such, harassment is best suited to be addressed by a multi-layered hazard prevention model, such as the Swiss Cheese Model we proposed above.⁸¹

Harassment is a significant hazard to workers’ mental, physical, and economic safety following historical trends of gendered and racialized subordination. Studies show workplace harassment is associated with increased rates of (1) stress, (2) anxiety and depression, and (3) post-traumatic stress disorder (“PTSD”).⁸² This effect may be particularly pronounced for women of color as they are vulnerable to harassment on the basis of both race and gender.⁸³ All three of these mental health effects are risk factors for increased rates of suicidal ideation, which also links workplace harassment to increased rates of suicide.⁸⁴

Workplace harassment frequently occurs with, and may even precipitate, other forms of workplace violence.⁸⁵ Physical violence can occur in any workplace and among any type of worker, even women in leadership; but the risk for fatal violence is highest among sales, protective service, and transportation workers, while the risk for nonfatal violence resulting in days away from work is greatest for healthcare and social assistance workers.⁸⁶ For women working in low-wage industries, the economic harms associated with workplace harassment can be particularly devastating. Economic violence can take many forms such as constructive discharge or retaliation for reporting or opposing workplace harassment, and can occur via termination, demotion, pay cuts, wage theft, or detrimental changes to job terms and conditions such as scheduling.

The detrimental effects of harassment are dangerously cyclical, leaving victims at risk of remaining stuck in these cycles. Like other health and safety hazards, preventing harassment demand multifaceted interventions, and many gaps remain. In Section II, we identify these gaps before turning to our data, which we analyze to examine the effectiveness of the overall policy response post #MeToo.

II. Legal Constraints Fail to Provide Accountability

Title VII contains no reference to the word “harassment.” Harassment law is instead a product of regulatory and judicial interpretation of the statutory language stating it is unlawful for an employer “to discriminate . . . with respect to . . . conditions, or privileges of employment.”⁸⁷ Lower courts began to recognize racial, religious and sex-based harassment claims in the 1970s,⁸⁸ while the EEOC recognized sexual harassment as a form of discrimination in 1980.⁸⁹ The Supreme Court first solidified the cause of action in the 1986 decision, Meritor Savings Bank v. Vinson.⁹⁰ The foundation set forth in Meritor would be elaborated in subsequent Supreme Court rulings.⁹¹

In this Part, we discuss ten major limitations of harassment doctrine, which leave significant gaps and fail to protect many women who are subjected to harassment at work. If the laws are ineffective, this leaves many victims without remedy, while also failing to deter future harassment due to the lack of accountability, weak enforcement of existing law, and restricted access to justice. In Part III, we examine whether the flurry of state reforms during the height of the #MeToo movement began to acknowledge and address these gaps to meaningfully improve protection for working women. In the absence of federal reform, state reform serves as both a laboratory of innovation as well as a way to gauge bipartisan support.

A. The Individuation of Harassment Evidence

As early as 1986 in Meritor v. Vinson, the Supreme Court imposed a compartmentalized frame on harassment cases, treating it as a matter of individual offenders and targeted victims as opposed to broader systemic harms in an organization. Meritor v. Vinson was brought by a Black woman, Mechelle Vinson, against her bank employer. Vinson was subject to sexual advances and repeated demands for sexual favors, fondled in front of other workers, followed to the restroom, and forcibly raped on several occasions by the bank vice president.⁹² The bank had a grievance procedure, but it would have required Vinson to report the misconduct to her supervisor—the perpetrator of the harassment.⁹³

From the outset, the Court failed to assess Vinson’s story comprehensively. It treated her harassment as an isolated circumstance and refused to consider the evidence of harassment other women at the bank were experiencing alongside Vinson as relevant to the question of whether the harasser cultivated an environment that violates Title VII.⁹⁴ This was despite the lower court’s holding that even where a woman is not the object of harassment herself, there may be a valid Title VII claim if the atmosphere fostered pervasive harassment. Instead, Justice Rehnquist concluded that the District Court did not allow the presentation of a “wholesale evidence of a pattern and practice relating to sexual advances to other female employees in her case in chief.”⁹⁵

This blinkered view of harassment is at odds with the concept of a hostile work environment, which can pervade and impact the workplace beyond one individual victim. The Court’s individualized approach also represents a marked departure from other claims available under Title VII. As early as the 1971 case of Griggs v. Dukes, the Court acknowledged that discrimination can operate systemically within a workplace and therefore permitted a wide variety of evidence from multiple sources.⁹⁶ In these Title VII discrimination cases, including “disparate impact,” “pattern or practice,” or “systemic” cases, the law acknowledges group offenses, even if individual experiences of discrimination vary and are effectuated by multiple actors and systems within an organization.⁹⁷ Consequently, the Court does not begin from the presumption that discrimination is limited to a single bad actor within the organization who has engaged in misconduct targeting a single individual.

Courts further compound the individuation of harassment claims through the routine enforcement of settlement and severance agreements containing non-disclosure provisions.⁹⁸ Such non-disclosure provisions conceal harassment, making it difficult to detect and root out patterns of abuse.⁹⁹ Settlement agreements commonly include non-disclosure provisions, and employers and their counsel often refuse to settle a case without some form of confidentiality provision. While some victims may want confidentiality,¹⁰⁰ this requirement often pressures victims to sign and remain silent in order to achieve some type of closure. This manner in which non-disclosure provisions are included in settlement agreements serves to limit collective grievances, opportunities to negotiate, and accountability for repeat perpetrators and organizations that foster hostile work environments.

B. Mandatory Arbitration Provisions

Access to legal justice under Title VII has also historically been barred for many workers by mandatory arbitration provisions. Approximately 56% of non-union private sector workers are required to sign mandatory private arbitration agreements,¹⁰¹ which represents roughly sixty million American workers.¹⁰² Of those, 30% have signed agreements that include class-action waivers.¹⁰³ Arbitration agreements have become particularly prevalent since the 2011 Supreme Court ruling, AT&T v. Concepcion, which enabled companies to enforce class and collective action waivers through arbitration.¹⁰⁴ Arbitration agreements divert claims from the public court system to private arbitration, in which neither the filings, rulings, nor proceedings are open to the public.¹⁰⁵ Even if the arbitration agreement does not require the victim to maintain the secrecy of the proceedings, arbitration is to a much greater extent shielded from public view, including media coverage and scholarly research.

Arbitration often offers poor remedies for the plaintiffs. When the arbitration agreement includes a class action waiver, the employee cannot file a collective claim in court or in arbitration, forcing them to bring an individual claim no matter the fact pattern.¹⁰⁶ Arbitration also offers few avenues for appeal, leaving workers little recourse if the arbitrator issues a bad ruling. Arbitration can further inhibit access to justice because claimants then have difficulty finding a lawyer willing to file a claim in arbitration when the deck is so heavily stacked against them. Research shows women of color are more likely to be denied access to courts than White women, due to mandatory arbitration.¹⁰⁷ This is because of the particular prevalence of these clauses in low-wage industries where women of color are overrepresented.¹⁰⁸ Thus, low-wage workers, who are already uniquely vulnerable to workplace violations including harassment and retaliation, suffer the most from contracts restricting their ability to access a court of law.

C. Coverage Gaps Constrain Workers’ Access to Courts

Many workers are not covered by key antidiscrimination and labor law statutes, leaving them with no legal recourse regardless of the merits of the claim. Title VII only covers claims against “employers,” and by extension claims brought by “employees.”¹⁰⁹ This coverage restriction means that independent contractors lack any form of protection under Title VII. Independent contractors make up over a third of the nation’s workforce,¹¹⁰ and almost half of these unprotected independent contractors are women.¹¹¹ Many of these are low-paid jobs in industries such as personal services, transportation, and educational services.¹¹² Women of color frequently land in these jobs due to the low barriers to entry, discrimination in other parts of the labor market, and the need for supplemental income.¹¹³ Research has shown that women and people of color are also overrepresented in most industries that tend to misclassify their workers as independent contractors.¹¹⁴

Title VII also largely excludes the most physically vulnerable low-wage workers from protection—often immigrants and women of color. For example, domestic workers who serve as housekeepers, nannies, babysitters, or home health care aides are often excluded from coverage if they are employed directly by individuals rather than agencies, as those individuals generally do not employ more than fifteen employees.¹¹⁵ This means that women like live-in worker, Etelbina Hauser, report having nowhere to turn for legal protection despite being “consistently groped” and pressured for “sexual services.”¹¹⁶ Domestic workers are often physically isolated from other workers, which further removes sources of social support and solidarity that might otherwise facilitate internal complaints and remediation. The physical isolation of private homes also tends to limit transparency and oversight of workplace practices. Thus, it is no surprise that studies indicate one-third of domestic workers face gender, race, language, or immigration-based abuse.¹¹⁷

Undocumented workers are nominally covered by Title VII,¹¹⁸ but the law does not prohibit workers from being fired—or deported—for their immigration status.¹¹⁹ These workers are also unlikely to be awarded back pay due to their immigration status, which reduces the legal risk for employers.¹²⁰ Immigrant workers make up a majority of the workforce in specific industries, including agricultural work.¹²¹ In these cases, the threat of termination or deportation largely cuts off any meaningful access to justice and makes them particularly vulnerable targets for harassment.

Coverage gaps that predominantly affect women of color and immigrants are no historical accident.¹²² As previously discussed, occupations in which women of color and immigrants predominated were intentionally excluded from landmark federal employment legislation in the twentieth century.¹²³ Employers have continued to treat women of color both as invisible and as their labor to control. Roles like nannies and maids are disproportionately held by immigrant women of color, traditionally employed in private homes of the mostly White middle- and upper-class, where a lack of transparency and adequate oversight gives employers the liberty to take advantage of these workers.¹²⁴

D. Retaliation Law and Its Effect on Underreporting

The legal standard for Title VII retaliation claims, which requires the plaintiffs to show that the retaliatory conduct was “materially adverse,” fails to deter low-level and informal retaliation.¹²⁵ Most courts find neither ostracizing nor harassing conduct to rise to the level of an adverse employment action for purposes of a retaliation claim under Title VII.¹²⁶ They construe the harm as not significant enough to deter someone from filing a charge.¹²⁷ Courts have even held that a negative performance review is not sufficiently “material” to support a retaliation claim.¹²⁸ And, as previously noted, judges with lifetime or fixed-term appointments may not recognize the threat that would deter a reasonable worker who lacks such security. One study, for example, revealed that many of the employment actions courts have held not “materially adverse” would actually dissuade participants from reporting.¹²⁹

The threat of retaliation can be a particularly powerful deterrent for marginalized populations.¹³⁰ Underrepresented groups within a particular workplace—such as women in non-traditional occupations, or women of color in majority White occupations—have less access to internal social networks and political capital within the workplace, which both increases the likelihood and the detrimental effect of informal social sanctions. The fear of job loss and unemployment for a low-wage worker can be so economically threatening to the employee’s livelihood that it deters them from reporting even extreme misconduct. Concerns about retaliatory deportation can weigh even more heavily on an employee’s decision-making. Moreover, even where actual retaliation is absent, workplace culture that fosters the threat of retaliation alone can deter victims from reporting.

Fears about retaliation for at-will employees who complain about harassment are particularly well-founded, despite whatever assurances human resources may provide about the company’s policy regarding retaliation. Even though retaliation for speaking up against harassment and discrimination is prohibited by law, it is a common workplace reality.¹³¹ Complainants may face formal action, such as termination, demotion, or pay cuts, as well as informal social sanctions. These repercussions originate not only from the perpetrator, but also from co-workers or supervisors who side with the perpetrator or perceive the complaint as disruptive. Studies have demonstrated that these negative consequences, such as being ostracized by coworkers, more often than not follow harassment reports.

E. Rigid Administrative Exhaustion Requirements Block Access to Justice

Title VII’s administrative filing requirement imposes a notably short time window—less than a year—for the plaintiffs to bring a harassment claim. Before Title VII claimants can file a lawsuit against an employer, they must file an administrative claim with the EEOC within 180 days of the last occurrence of harassment.¹³² Any lawsuit brought by an employee that has failed to timely file an administrative claim will be dismissed for failing to exhaust administrative remedies. This aggressive window can be unrealistic for traumatized workers who may be afraid to speak up about harassment or who may not recognize that the workplace harms they suffered qualified as unlawful harassment until years later. The narrow administrative filing window can be particularly detrimental for the vulnerable workers previously described, who may be concerned about retaliation and job loss.

These workers might reasonably choose job security over the possibility of a lawsuit, such that they may not be ready to file a legal claim until they have secured adequate support or alternate employment. Moreover, gaps in access to legal services and information often preclude workers from even knowing about the administrative filing requirement.

F. Severe or Pervasive Requirement Excludes Strong Claims

To prove harassment under Title VII, the plaintiffs must show that they were subject to unwelcome comments or conduct on the basis of a protected category (race, sex, religion, color, or national origin) that was so “severe or pervasive” as to alter the conditions of the plaintiffs’ employment and create a “hostile or abusive work environment.”¹³³ The “severe or pervasive” language originated in Meritor, although the court did not initially define the term.¹³⁴ In a subsequent 1993 ruling, Harris v. Forklift, the Supreme Court elaborated on the meaning of the phrase by listing several non-exhaustive factors that affect whether conduct is deemed severe or persuasive: (1) the frequency of the conduct; (2) the severity of the conduct; (3) whether is the conduct is physically threatening, humiliating, or “a mere offensive utterance”; and (4) whether the conduct unreasonably interferes with work performance.¹³⁵

As several commentators have observed, the “severe or pervasive” requirement has evolved to impose a very high burden of proof on the victim.¹³⁶ For example, lower courts have inconsistently and often improperly interpreted the type of conduct necessary for a violation. In some courts, the standard has been deemed so high that it may reject claims for conduct that may be egregious, offensive, and in some cases even criminal.¹³⁷ This significantly impacts outcomes of cases, as the “severe or pervasive” requirement has become a common basis on which courts grant summary judgment against the plaintiffs. Judges have ruled that things like indecent exposure, being threatened and referred to as a Black bi[***], and being offered pornography by one’s boss, are not sufficiently severe or pervasive, thus dismissing the cases on summary judgement.¹³⁸ This leaves the plaintiff demoralized and without remedy and allows the employer to shield itself from accountability.¹³⁹

Other lower courts have misinterpreted the Harris v. Forklift opinion to require that conduct be “severe, frequent and physically threatening,” effectively requiring severe and pervasive conduct that is also physical in nature.¹⁴⁰ In McGraw v. Wyeth-Ayerst Laboratories, for example, the court held that repeated propositions, yelling, and non-consensual kissing by a supervisor was neither severe nor pervasive.¹⁴¹ Extreme lower court rulings can have a lasting effect, as courts later rely on those fact patterns and judicial interpretations to justify outcomes in favor of employers in subsequent cases.¹⁴²

G. The “Objectively” Hostile or Abusive Standard is Out of Touch

Closely related to the “severe or pervasive” requirement is the requirement that the working environment be both “subjectively” and “objectively” hostile or abusive. This means that the plaintiff perceived the conduct as hostile or abusive, and that a “reasonable person” in that situation would have found it hostile or abusive.¹⁴³

Defining reasonableness has proven difficult as courts appear to lack a clear standard.¹⁴⁴ In a 1998 case, Oncale v. Sundowner Offshore Services, Inc., the Court further noted that ensuring whether or not something is objectively hostile or abusive is crucial to “ensure that courts and juries do not mistake ordinary socializing in the workplace . . . for discriminatory ‘conditions of employment.’ ”¹⁴⁵ Here, the Supreme Court clarified that determining severity “requires careful consideration of the social context in which particular behavior occurs.”¹⁴⁶

The “objective” component of the hostile or abusive standard has also proven problematic. In the summary judgment context, judges superimpose or extrapolate from their own experiences when deciding what a “reasonable person” would consider hostile and abusive.¹⁴⁷ Yet, the judge’s own assumptions about tolerable behavior can be tainted by white and male privilege in ways they may not recognize. As previously discussed, the pervasive influence of hegemonic masculinity can lead judges to discount hostile aspects of the work environment simply because they are commonplace or fit within traditional workplace norms.¹⁴⁸

For example, in Oncale, Justice Scalia recounted a variety of workplace behaviors that he considered inoffensive, such as a football player being smacked on the buttocks by his coach—which very well could be experienced as hostile.¹⁴⁹ However, when courts—and juries—are instructed to ignore the plaintiff’s actual (subjective) experience and focus on what a hypothesized “reasonable” (objective) other would consider harassment, it is an implicit invitation to default to a framework wherein workplace culture is largely defined by and governed by White men.¹⁵⁰

Broader representation in the judiciary may begin to address this bias. In other research, we have found that there is a significant disconnect between judges’ assessments of what is “objectively” abusive and hostile and a layperson’s assessment, which may mean that judges are disconnected from social realities and evolving social norms.¹⁵¹ However, bias does not only impact judges. Racialized and sex stereotypes can also color perceptions of witnesses, fact finders, and others, clouding their view whether the plaintiff contributed to the harassment, the extent to which they feel the plaintiff is harmed by the perpetrator, and whether enduring the conduct is deemed acceptable and within the realm of their role as worker.¹⁵²

H. The Standard for Vicarious Liability

As a general matter, when an employee commits a Title VII violation, the law imposes strict liability.¹⁵³ For example, when an employee engages in sex-based discrimination or retaliates against an employee for speaking out about it, courts do not inquire whether the employer should be held vicariously liable for the conduct. Employers are simply liable for the violation. Harassment, however, is the exception. In Meritor, the Supreme Court first raised the question of whether there might be some circumstances in which employers might not be held vicariously liable for harassment.¹⁵⁴ This question was settled in two 1998 cases decided together, Burlington Industries v. Ellerth and Faragher v. City of Boca Raton (“Faragher/Ellerth”).¹⁵⁵ In these two cases, the Supreme Court imported tort principles into the employment discrimination context. The majority opinions held that employers would only be held strictly liable for harassment committed by a supervisor when the plaintiff experienced a tangible employment action, such as a demotion, firing, or pay cut.¹⁵⁶ The Court also imposed a negligence standard when coworkers were responsible for harassment. Under this rule, employers would only be held liable for harassment that they knew or should have known about, and they failed to take action to correct it.¹⁵⁷

In cases involving supervisory harassment, in which the plaintiff suffered no tangible employment action, the Court created a new affirmative defense for employers. Even when a plaintiff proves a successful harassment case—which is an uphill battle due to the constraints discussed in this section—the employer can assert an affirmative defense to evade liability. Joanna Grossman argued that the defense effectively insulates employers from liability following an initial complaint about harassment.¹⁵⁸ To assert the defense, an employer must show that it “exercised reasonable care to prevent and correct promptly any sexually harassing behavior, and that the plaintiff employee unreasonably failed to take advantage of any preventive or corrective opportunities provided by the employer or to avoid harm otherwise” (the “Faragher/Ellerth defense”).¹⁵⁹

Beyond the exceptional nature of the Court’s departure from a strict liability framework, courts have also interpreted the Faragher/Ellerth defense in an expansive manner, favoring employers. In many cases, courts merely require employers to maintain anti-harassment and complaint policies, without inquiring into the effectiveness or reviewing the culture broadly.¹⁶⁰ This trend of “judicial deference” is largely why anti-harassment policies and practices have proliferated in the workplace, many of which have been found to be ineffective at curbing harassment.¹⁶¹

Likewise, the Faragher/Ellerth defense directs the blame for a hostile work environment on a plaintiff who was slow or reluctant to complain internally, rather than on the perpetrator or the employer. However, high rates of retaliation give many employees a valid reason to pause when reporting harassment or otherwise using grievance procedures.¹⁶² Only one in four women subjected to sex-based harassment reported it using an internal grievance procedure; even fewer filed a charge with the EEOC.¹⁶³ At present, it is possible for a plaintiff to file a hostile work environment claim, and, despite evidence to her benefit, lose merely because she herself failed to utilize the internal grievance procedures created by the employer.¹⁶⁴

I. Damage Caps Fail to Remedy Harms

The remedies available in Title VII cases include injunctive relief, reinstatement,¹⁶⁵ back pay, compensatory damages, punitive damages, and attorneys’ fees and costs.¹⁶⁶ Compensatory damages—that is, damages for pain and suffering—are subject to a statutory cap according to employer size.¹⁶⁷ For employers with fewer than one hundred employees, compensatory and punitive damages cannot exceed $50 thousand; for two hundred or fewer employees, the cap is $100 thousand; for five hundred or fewer, the cap is $200 thousand; and for five hundred employees or more, the cap is $300 thousand.¹⁶⁸

Damage caps for compensatory and punitive damages can be particularly harmful in harassment cases when devastating psychological effects are common.¹⁶⁹ Such effects can include suffering from depression, anxiety, stress, post-traumatic stress disorder, suicidal ideations, and adjustment disorders, among others.¹⁷⁰ One study found that nurses who experience sexual harassment are three to eight times more likely to suffer from depression than women who were not harassed.¹⁷¹ These negative mental health outcomes have also been shown to have profound impact on long-term job-related outcomes, including lack of initiative, lower job satisfaction, increased propensity to leave, and financial problems.¹⁷²

In addition, other forms of available relief may not be especially meaningful in harassment cases. Although some employees quit or are fired in connection with workplace harassment, many harassment victims continue to work throughout the abuse, making them ineligible for back pay. Ironically, staying longer and enduring more prolonged harassment and abuse may end up resulting in lower damages. This is particularly problematic for low-wage workers who are economically vulnerable and cannot afford to quit their jobs. Conversely, if the employee was terminated, reinstatement may not be a useful remedy, as the plaintiff may be reluctant to return to an abusive workplace. Compensatory and punitive damages, along with attorneys’ fees and costs, may be the most important forms of relief available to harassment claimants. Yet, damage caps force courts and juries to limit relief to the plaintiffs to whom they might have made a much larger award to compensate for pain and suffering and to punish the employer for maintaining a hostile work environment.¹⁷³

Damage caps also limit access to justice. Because many plaintiff-side lawyers operate on contingency fees, a harassment claimant suing a small or even mid-size employer may have difficulty finding a lawyer willing to sue when the maximum recovery is less than $100 thousand. The effect of damage caps can be especially pronounced for low-wage workers, women of color, and immigrant workers engaged in domestic work or agricultural labor. When the size of recoverable wages is low due to a low base wage, plaintiffs are even more reliant on compensatory and punitive damages to attract the interest of a potential lawyer. Domestic workers, agricultural workers, and even restaurant and food-service workers may find themselves on the low end of the damage caps because they work for smaller operations. In such cases, a worker’s ability to find legal representation may ultimately depend on whether they can allege a separate tort or statutory claim not subject to the damage caps or whether they live in a state that offers more generous discrimination remedies under state law.

Damage caps also reduce deterrent effects on delinquent employers.¹⁷⁴ With minimal penalties, many employers are disincentivized to improve their responses to sexual misconduct in the workplace or to change workplace culture. Once again, vulnerable workers are doubly cursed—their employers know that their workers may not be covered by the statute, and even if they are covered, they may not find an attorney or recover very much. Facing little prospect of a big-ticket lawsuit, employers in these industries can turn a blind eye to harassment with little fear of accountability.

J. Implied Hierarchy of Harassment Claims

Even prior to the #MeToo movement, the dominant narrative surrounding workplace harassment involved a subordinate White woman subjected to unwelcome sexual conduct or requests for sexual favors by a high-ranking White man. This scenario, for example, commonly appeared in early harassment training from the 1980s and 1990s.¹⁷⁵ It was also, to some extent, reflected in the earliest EEOC regulations, which defined harassment in terms of sexual conduct and specifically referenced “quid pro quo” harassment, in which a supervisor requests sexual favors in exchange for some workplace benefit or the avoidance of harm.¹⁷⁶

This frame was not, however, compelled by the case law,¹⁷⁷ and it operates to the disadvantage of all other harassment claims that do not involve sexual conduct or that are brought on the basis of other protected classes. Intersectional claims brought on the basis of more than one protected category—often women of color experiencing racism and sexism—are at particular disadvantage.

Nowhere in the case law does the Supreme Court assert that sexual conduct forms a necessary part of a harassment claim or that certain protected classes are more deserving of relief than others.¹⁷⁸ Nevertheless, the narrative of harassment as primarily a problem of sexual misconduct toward White women has had a measurable effect on lower court jurisprudence over time. This effect is well-documented in scholarly literature. Indeed, Pat Chew and Robert Kelly’s empirical study of harassment claims concluded that judges tend to discount race-based harassment claims.¹⁷⁹ Women of color pursuing litigation are further marginalized due to the courts’ pressure to separate out experiences of harassment into the false dichotomy of “Because of Race” or “Because of Sex,” when these are commonly intertwined.¹⁸⁰ Empirical research has found that plaintiffs bringing intersectional claims are less than half as likely as plaintiffs bringing single claims to win their cases.¹⁸¹ Even within those statistics, Black women are more likely to lose their cases than Black men who bring intersectional claims, such as those based on both race and age.¹⁸²

In summary, harassment law is subject to a variety of gaps that enabled workplace harassment to continue in the decades leading up to the #MeToo movement. Next, we turn to the question of whether the many legislative reforms wrought by the #MeToo movement addressed these gaps.

III. An Empirical Analysis of State and Federal Gender Equity Legislative Activity, 2016–2022

In previous scholarship, we discussed the ability of social movements to promote legal change.¹⁸³ While the window of opportunity may be small, #MeToo, like the mass Black Lives Matter protests during the summer of 2020, has the potential to generate staying power and remain influential on legal policy.¹⁸⁴ This is especially true if lawmakers, courts, and agencies follow the lead of workers who have been organizing for harassment-free workplaces long before #MeToo.¹⁸⁵ Doing so will provide stakeholders an appropriate goalpost that they can measure their efforts against as they attempt to mitigate workplace harassment.

This Article builds on prior work that has begun to investigate the legal implications of #MeToo, but does so with an emphasis on empirical analysis.¹⁸⁶ The questions we explore include: (1) Has #MeToo effectively shifted the law forward in addition to raising awareness and sparking debate? (2) What topics were centered in bills versus introduced as secondary topics within proposed legislation? (3) To what extent did political factors such as state party lines and representation of women lawmakers influence the volume of bills introduced and passed? (4) To what extent did legislators take a comprehensive versus a narrow approach to workplace harassment?

A. Methodology

To examine the actual and potential policy changes following #MeToo, our research team collected a corpus of 3,916 state bills¹⁸⁷ and 255 federal bills relating to workplace gender equity, discrimination, and harassment that were introduced between 2016 and 2022. The corpus was collected through a legislative search of Legiscan, NexisUni, and Westlaw using seventy-five different search terms intended to identify relevant legislation.¹⁸⁸

The search terms were generated to identify a broad range of workplace gender equity legislation rather than more narrowly-focused harassment bills for several reasons. As articulated in Part II, harassment is a complex systemic problem stemming from a variety of social, political, and cultural factors. To the extent legislation helps to address the underlying inequity—such as pay equity—the intervention may ultimately help to reduce harassment indirectly over time.¹⁸⁹ We therefore sought to capture the full range of legislative interventions to enable us to evaluate them collectively. Taking a broader approach to harassment reform also allows us to identify patterns in and across legislative bills, including the extent to which the bills are: (a) individualized, offering siloed remedies for individual cases; (b) institutionalized, delegating responsibility to employer training, policy or practices; or (c) systemic, which are broader in scope and affecting many workers across industries.

The initial corpus contained a substantial number of duplicative bills, typically as a result of near-identical legislation that was introduced separately in both the House of Representatives and the Senate of a particular state legislature. Potential duplicates were flagged manually based on similar or identical bill names, numbers, or descriptions. The textual similarity of potential duplicates were then assessed using the “compare” function in Adobe Acrobat Pro, which counts and highlights all textual differences. Pairs of bills where 80% or more of the text was identical were deemed duplicates, and one copy of the duplicate bill was discarded for the substantive coding and quantitative analysis. The final corpus analyzed herein contained 3,012 state bills and 255 federal bills (“Legislative Corpus”).

We then coded for the substantive topic of the bills pulled into our dataset through the criteria listed in Table 1.¹⁹⁰ Bills fell into one of eleven topic groups, as defined below.

Table 1. State Legislative Topics
Topic	Definition
Transparency	Bills that implement new reporting or recordkeeping on employers regarding harassment or assault, or that restrict the use of non-disclosure provisions in employment contracts or settlement agreements.
Government Officials & Contracts	Bills that regulate the conduct of government officials, lobbyists or government contractors, such as requirements that legislatures adopt harassment policies for members; prohibiting public funds from being used to settle harassment claims; or requiring state contractors to adopt certain employment practices.
Anti-Harassment Practices	Bills that require some or all employers to adopt anti-harassment practices, such as training, policies, notices, or procedures to investigate harassment. Also includes bills that extend unemployment benefits to employees who quit due to harassment.
Anti-Discrimination Law	Bills that extend or expand anti-discrimination protections, such as expanding coverage or adding new protected categories such as sexual orientation, family status, marital status, or victims of domestic violence.
Leave	Bills that mandate paid or unpaid leave or that provide for accommodation for pregnant or nursing mothers.
Pay Equity	Bills that alter legal rules, hiring practices, or compensation practices relating to pay disparities or withholding of promotion or opportunities on the basis of sex, such as equal pay laws, prohibitions on requesting salary history, protection for pay discussions in the workplace, or mandatory pay disclosures.
Occupational	Bills that expand or create legal protections for sectors or occupations that predominantly employ women such as domestic work, hospitality, hotels, and janitorial services.

Enforcement	Bills that alter or expand legal rules or remedies relating to harassment, discrimination, leave/accommodation, equal pay, non-disclosure agreements, private arbitration, or workplace bullying.
Mandatory Arbitration	Bills that attempt to alter the enforceability of private arbitration provisions in employment contracts.
Equal Rights Amendment	Bills that ratify or pass the Equal Rights Amendment.
Other^a	Bills that do not fit into the above categories, many of which are symbolic in nature, such as designating “Equal Pay Day,” establishing commissions or reports. Also includes laws regarding vocational training in non-traditional occupations, and gender diversity in boards of directors.

Note: ^a The bulk of the “Other” category consists of bills that engage with harassment or gender equity symbolically rather than substantively. Examples include bills that create Days or Months to honor gender inequities that other bills in the dataset are seeking to redress. Other bills considered symbolic are those that commission Task Forces or Studies on harassment or gender equity reform more broadly. The rest of the “Other” category were outliers such as grants or vocational training for women in “High-Wage, High-Demand” jobs, regulation on law enforcement and military, and bills requiring the appointment and disclosure of women on boards.

Each bill summary was reviewed and coded by a member of our five-person research team, which included four researchers with a J.D. and an economics Ph.D. fellow. Where the summary was not sufficient to code the bill into one of the above-listed categories, the researcher reviewed the full text of the bill. Bills that fell outside the scope of the study—such as laws relating to cyberbullying, rape kits, or “street harassment”—were removed from the Legislative Corpus.

Many of the bills in the Legislative Corpus were “bundled,” that is, they sometimes included a variety of separate measures.¹⁹¹ Where a given bill covered more than one of the relevant topics, it was coded using a “primary” topic and any additional topics received a “secondary” classification. For example in 2019, the Connecticut legislature introduced a bill that would amend multiple statutes to establish “economic equality for women,” including by requiring equal pay for equal work, increasing the minimum wage, requiring paid family and medical leave, and requiring annual anti-harassment training for companies with fewer than twenty employees among others.¹⁹² Thus, while this bill was coded primarily as a Pay Equity bill, it also was secondarily classified under Anti-Harassment Practices, Leave, and others.

Where bills presented a close case regarding the applicable category, members of the team discussed the bill and made a final decision. Coding decisions were also cross-checked using keyword searches within the Legislative Corpus specific to each subcategory.¹⁹³ Keyword searches within the Legislative Corpus were also used to populate subcategories of bills within the “Enforcement” category relating to harassment law reforms.¹⁹⁴

B. Volume of Gender Equity Legislation Post #MeToo

Federal Reforms

The #MeToo movement produced few legislative successes at the federal level between 2016 and 2022. Although many bills were introduced, few of them passed. By contrast, there has been a massive wave of #MeToo-related legislation at the state level. This flurry of legislative activity took place in statehouses across the country and continued for a sustained period—extending throughout 2022.

In the U.S. Congress, from 2016 to 2022, 255 bills were introduced relating to sexual harassment, sexual assault, and gender equity in employment.¹⁹⁵ There was a sharp rise in bills between 2016 and 2017, from twenty to fifty-nine. Congress continued to introduce legislation at a high rate in 2018 and 2019—reaching thirty-eight bills in 2019. Proposed legislation then dropped sharply in 2020, with five bills introduced that year.

Figure 2. Federal Gender Equity Bills (2016–2022)

At the federal level, only six #MeToo related bills passed during the sample period, and three of those bills are very limited in scope. The most wide-reaching bill was the Ending Force Arbitration of Sexual Assault and Sexual Harassment Act of 2021, discussed in greater detail in Part V.E.¹⁹⁶ In 2022, Congress also passed the Speak Out Act of 2022, which limits the enforceability of non-disclosure and non-disparagement clauses relating to sexual harassment and assault.¹⁹⁷ Another substantive change came from the National Defense Authorization Act for Fiscal Year 2022, which created additional reporting requirements for sexual harassment in the military—an industry at high risk of workplace harassment, among other forms of workplace violence.¹⁹⁸

The remaining laws that passed at the federal level were quite limited in scope. Two such bills related to lawmakers themselves: one mandates anti-harassment training for Senators and Senate employees¹⁹⁹ and the other makes lawmakers financially liable for harassment settlements.²⁰⁰ A third

bill was symbolic in nature, designating April as National Sexual Assault Awareness and Prevention Month.²⁰¹

State Reforms

As aforementioned, the vast majority of harassment and gender equity legislation was introduced and passed by state legislatures rather than Congress. We will begin the state legislation analysis by looking at state-specific and political trends in introduction and passage rates of gender equity bills between 2016 and 2022. We will then take a closer look at the potential for bipartisan support of gender equity bills before moving to topical analysis; there, our goal is primarily to examine how well legislation has been able to fill the gaps in legal protection that leave so many workers vulnerable to harassment and other gender-based workplace harm.

i. General Trends

Broadly speaking, states introduced—and passed—a large number of #MeToo related bills between 2016 and 2022. Since 2016, states introduced approximately 3,000 such bills, of which 382 passed—a passage rate of 12.7%. As Figure 3 illustrates, some amount of gender-related legislative activity predated the viral spread of the #MeToo movement in 2017, with 244 relevant bills introduced in 2016, though only 19 such bills passed. The volume of #MeToo related legislation nearly doubled between 2016 and 2017, with 427 bills introduced, of which 42 passed. Legislative activity continued to rise year over year even after the extensive media coverage faded, peaking in 2019 at 715 bills introduced, of which 96 passed. Legislative activity was somewhat lower between 2020 and 2022 but remained above the 2016 baseline level. These trends suggest that the legislative momentum of the #MeToo movement continued for several years.

Figure 3. State Gender Equity Bills (2016–2022)

There was substantial variation between states in the volume of legislation introduced, as illustrated in Figure 4. The most active state legislatures were New York (340 bills), followed by New Jersey (208 bills), Mississippi (151 bills), California (113 bills), Illinois (112 bills), and West Virginia (112 bills). Other states with a relatively high volumes of legislative activity were not confined to consistently Democratic “blue” states,²⁰² and included “swing”²⁰³ and Republican leading states, including Virginia, Michigan, Minnesota, Pennsylvania, Texas, Hawaii, and Missouri.

However, the states with very little legislative activity—those that introduced ten or fewer #MeToo related bills between 2016 and 2022—were “red” states: Arkansas (four bills), South Dakota (six bills), North Dakota (seven bills), Montana (seven bills), Idaho (nine bills), and Wyoming (ten bills). Nevertheless, the sustained engagement of most states in #MeToo legislation suggests that there was more sustained legislative interest and energy around #MeToo-related gender equity reform than the partisan gridlock in Congress would suggest.

Figure 4 also highlights the years in which bills were introduced in each state, indicated by color within each bar. Although each state had more legislative activity in some years than others, they generally depict some legislative activity during each year in the sample period. This further suggests sustained legislative activity over time, not just across states writ large, but within each state.

The states that passed the most #MeToo related bills were similar—but not identical—to those that introduced the most #MeToo related legislation, with California in the lead (seventy-three bills), followed by Illinois (thirty-one bills), New York (twenty-seven bills), Washington (twenty-five bills), New Jersey (seventeen bills), Virginia and Maryland (fifteen bills each), Nevada (thirteen bills), then Louisiana, Oregon, and Maine (twelve bills each).²⁰⁴

Although blue states dominate the list, the most prolific bill-passing states also included a solidly red state (Louisiana), and two swing states (Nevada and Virginia). The group of states that passed between five and ten bills also included a handful of swing or red states, including Pennsylvania, Texas, Arizona, Tennessee, and Kentucky, each of which passed five bills. Ten states failed to pass any #MeToo bills. Of these, seven were red states, and three were swing states.²⁰⁵ In other words, the volume of bills passed in any given state is somewhat more reflective of the partisan divides between red states and blue states than the volume of bills introduced.

Figure 4. Bills Introduced by State (2016–2022)^a

Note: ^a Only states with at least 20 introduced bills were included in Figure 4.

Figure 5. Bills Passed by State (2016–2022)^a

Note: ^a Only states that passed at least one bill were included Figure 5.

C. Political Trends in Gender Equity Legislation Post #MeToo

Now that we have a sense of the general landscape with respect to state legislation post #MeToo, we can take a closer look at the political trends. While all states introduced some harassment and gender equity reform legislation, there are trends in region, party, and even representation of women in legislatures, that provide information about who is leading the charge with respect to introducing and passing legislation that produces effective change.

Political Party

Figure 6 presents a more direct breakdown of the relationship between partisanship and #MeToo legislation. Throughout much of the sample period, more #MeToo-related bills were introduced in Republican-led legislatures, relative to Democrat-led legislatures.²⁰⁶ However, Democratic legislatures passed far more bills than Republican-led legislatures by a margin of 3.6 to 1. These results suggest that there was substantial legislative interest in #MeToo related topics in both Republican and Democratic state legislatures—and perhaps even a greater level of experimentation in Republican legislatures—but far less momentum to pass those bills in Republican states.

Figure 6. Bills Introduced and Passed by Party Majority and Year

a. Introduced bills

b. Passed bills

Note: ^aThis figure does not include seven laws introduced in Alaska that had “N/A” for party majority

Representation of Women Lawmakers

Part I discussed the historical exclusion and underrepresentation of women and racialized minorities in the lawmaking process. From 2016 to 2021, as the conversation around #MeToo and gender equity broadened, the percentage of women in state legislatures grew to record numbers in many states.²⁰⁷

Figure 7. Percent of State Legislators Who Are Women, 2012–2021

We next analyzed the relationship between the average representation of women in each state legislature from 2016 to 2022, and the volume of gender equity legislation the state passed during that same period. The results suggest that states with more women lawmakers tended to pass more gender equity legislation in the years following #MeToo than states with fewer women lawmakers.²⁰⁸

Most states tended to cluster together along party lines. For example, traditionally red states including West Virginia, Wyoming, Alabama, and South Carolina had state legislatures with less than 20% of women lawmakers and also passed less than ten gender equity bills between 2016 and 2022. By contrast, Democratic strongholds such as Illinois, New York, and Washington had legislatures with between 30% and 40% women lawmakers, as well as over twenty gender equity bills passed in each of these states between 2016 and 2022. This suggests that Democratic states may be leading the way with respect to maximizing anti-discrimination legislation put forth—in line with the Swiss Cheese Model of risk reduction. However, the substance of the bills passed is another critical part of the analysis.

Figure 8. Frequency of Gender Equity Legislation Passed, by Mean % Women Lawmakers, 2016–2022

D. Topical Analysis of Gender Equity Legislation Post #MeToo

This Section takes a closer look at the types of bill topics introduced or passed and examines whether the activity was responsive to the gaps in protection previously identified in Parts I and II. We found that states attempted to pass a variety of legislative measures on a number of topics related to gender equity and workplace harassment. The breadth of introduced bill topics is promising, as it suggests legislators are open to taking a multi-layered approach to increasing protection and minimizing harassment—especially in industries at particular risk.

Figures 9 and 10 display the introduced and passed bills according to the topic of the legislation. As Figure 9 illustrates, pay equity was the most common topic of proposed legislation, with more than 700 bills introduced. Leave laws were the second most common topic, with 582 bills proposed. Enforcement bills—which included a variety of reforms to strengthen protections by changing legal rules and remedies—were the third most prevalent, with 376 bills. Laws mandating changes to employer practices and policies (“Anti-Harassment Practices”), such as harassment training and grievance procedures, came in fourth, with 332 bills.

The topics most commonly passed in the legislature did not correspond to the topics that were most frequently introduced. The most commonly passed topic was “Other,” which predominantly consisted of symbolic legislation, such as designating an “Equal Pay Day,” expressing a policy position, or establishing a task force. The second most commonly passed topic was laws relating to Anti-Harassment Practices, closely followed by laws relating to Pay Equity.

Figure 9. Number of State Gender Equity Bills, By Primary Topic

The volume of bills introduced on specific topics could ultimately be viewed as a proxy for broad legislative interest in a topic, with pay equity and leave garnering the most interest. Legislative volume may also be an indicator of legislative creativity and innovation around a particular topic. Pay Equity bills in particular produced a wide variety of proposals to address the underlying problem, ranging from pay disclosure rules, protection from retaliation, new forms of equal pay mandates, or restrictions on the types of information employers can use to calculate pay rates. Republican led states were often as creative in this regard as Democratic ones.

For example, West Virginia introduced a bill entitled the “Katherine Johnson Fair Pay Act of 2019” in honor of Katherine Coleman Johnson, an African American mathematician born and educated in West Virginia.²⁰⁹ She was awarded the Presidential Medal of Freedom despite facing segregation and wage discrimination along with the rest of her Black female crew. The bill would prohibit an employer from (1) banning pay discussion in the workplace formally through waiver or informally and (2) inquiring about prospective employees’ wage or salary history.²¹⁰ Utah and Wyoming also took steps to bolster equal pay protections; in 2016 and 2019, respectively, both states passed legislation increasing fines for employers and legal remedies for victims of wage discrimination.²¹¹

Many states also used a bundling approach, by adding in additional topics that were “secondary” to a primary bill topic.²¹² Once secondary topics are included, the “Other” category remains the most commonly passed topic. However, other types of reforms received a substantial boost by including “secondary” topic codes, particularly the “Enforcement” category, which rises to second place when secondary topics are included. Likewise, anti-discrimination law reforms were much more frequent when adding analysis of secondary topics. This suggests that legislators are strategizing—proposing a bill on a primary topic, such as pay equity, that may garner wider support, and then in the same bill, adding other protections that are less likely to be presented and supported in standalone bills.

These secondary “hidden” topics are of note because they tend to add substance to bills that may otherwise lack the ability to promote effective change. We measure the effectiveness of a bill by assessing its ability to fill at least one of the many gaps identified in harassment and anti-discrimination law. The ability to fill these “holes,” to use the Swiss Cheese Model analogy, indicates an appropriate focus on how harassment and other gender inequities in the workplace are actually experienced by today’s most vulnerable workers. For example, many of the workers most vulnerable to harassment also have extremely limited economic bargaining power due to various factors including the racial and gender pay gap. To bolster these workers’ ability to fight existing wage discrimination, many state legislators added a cause of action into pay equity reform bills that would have otherwise lacked an enforcement mechanism that allowed workers to take their claims in court.

Figure 10. Total Number of State Gender Equity Bills Passed, By Topic

Secondary topics were a common feature of the dataset: almost 40% of bills had one or more secondary topics.²¹³ The most common topic bundle combined Pay Equity and Enforcement (307 bills), followed by Leave and Enforcement (200 bills). This bundle tended to reflect that new leave or pay equity laws were typically more substantive in nature and drafted with some sort of enforcement remedy beyond fines for employer violations—thereby increasing workers’ access to courts. For example, in 2016, even before the #MeToo activism, California passed an Equal Pay for Equal Work bill that was both systemic and substantive and included a bundle of enhanced protections.²¹⁴ This new law prohibits wage discrimination, creates a cause of action for employees aggrieved by wage discrimination, and prohibits retaliation against employees seeking enforcement of the law.²¹⁵

Table 2 also reveals less intuitive combinations, such as bundling Anti-Discrimination Law along with other bill topics. One example of this strategy comes from New York, where the 2020 legislature introduced an extensive bill that bolstered discrimination and harassment enforcement remedies and expanded protected classes to include individuals experiencing these harms on the basis of their sexual orientation and gender identity, marital status, familial status, and more.²¹⁶ This strategy was also adopted in many Leave laws, which were frequently accompanied by reforms to Anti-Discrimination Law (seventy-five bills). Pay Equity was also commonly paired with reforms to Anti-Discrimination Law (forty-one bills). This bundled approach may have served as an effective strategy to effect systemic broadening of statutory protection without attracting undue attention and opposition.

Indeed, it appears that strategically “bundling” topics increased the ability of legislatures to pass reforms that either (1) create or strengthen a variety of enforcement mechanisms for harassment and other gender equity claims, or (2) broaden coverage under anti-discrimination statutes. Although standalone bills involving these reforms had less success, legislators appear to have successfully tacked them on to bills involving other subjects.

Table 2. Common Bundles of Workplace Gender Equity Topics
Primary topic	Secondary topic
Primary topic	Anti-Discrimination Law	Anti-Harassment Practices	Enforcement	Anti-Discrimination Law	Anti-Harassment Practices	Mandatory Arbitration	Anti-Discrimination Law	Anti-Harassment Practices	Pay Equity	Anti-Discrimination Law
Anti-Discrim. Law		5	60	21	10		1	2	3	1
Anti-Harassment Practices	25		29	11	2	4	9	13		37
Enforcement	25	12		5	12		3	4	6	9
Govt. Officials and Contracts	8	58	11		1			1	1	19
Leave	75		200				18	6	2	27
Mandatory Arbitration	2	3	20	10				1		12
Occupational	1	12	12		3				1	5
Other	7	8	3	1	7			16		8
Pay Equity	41	1	307	47	17		22	51		54
Transparency	6	10	36	28	1	14	8	2	1

Reform initiatives that aim to strengthen enforcement mechanisms and broaden coverage under Title VII to protect more workers are essential because they increase access to the courts. Reform will not reduce the prevalence of harassment and other gender inequities if impacted workers are unable have their claim addressed by the court system. It is also true, however, that the judiciary needs reform itself, particularly with respect to its interpretation of the severe or pervasive standard, the objectively hostile standard, the Faragher/Ellerth defense, and retaliation. Until the judiciary is more in touch with the realities of the workplace, it is unclear how well this new wave of workers with access to the courts will fare.

Progressive coastal states such as California and New York provide examples of legislative agendas that substantially shift how workplace harassment is conceptualized by courts.²¹⁷ This legislation attempts to remedy some of the shortcomings in harassment jurisprudence discussed in Part II.²¹⁸ Both of these states also amended laws to expand protections for harassment and discrimination to include more protected categories as well.²¹⁹

Some states also improve enforcement by extending the administrative filing period of harassment and discrimination claims. Five states (California, Connecticut, Maryland, New York, and Oregon) extended the administrative filing deadline.²²⁰ Of these, three states extended the deadline for all discrimination claims.²²¹ This is particularly important for low-wage workers, for whom inadequate filing deadlines exacerbate existing pressure to use limited time and resources to at once both find a job and seek legal recourse for harm suffered.

A relatively small number of bills sought to increase the damages available in harassment claims. These bills were also somewhat successful, with four states—Virginia, Nevada, New York, and Connecticut—increasing available damages. Many—but not all—who passed bills did so by (1) allowing the discrimination plaintiffs to recover both compensatory and punitive damages and (2) removing damage caps for victims of discrimination based on employer size. In some states—such as Nevada—damage caps remain, limiting the impact of recent increases in available damages.²²²

IV. Discussion & Implications

When analyzing the legislative activity over time, we found that proposed bills began to address a wider range of systemic gender equity issues over the time period;²²³ however, as with many movements, reform efforts gradually fizzled out over time. Importantly, however, they never returned below the 2016 baseline activity. These empirical results suggest that while the #MeToo movement may have sustained some of its initial impact on harassment and gender equity reform over the past five years, progress may be stagnating. Avoiding further stagnation requires legal stakeholders to offer workers consistent harassment and gender equity reform that are responsive to the changing realities of the current workplace. This includes acting swiftly to respond to the 2025 Trump administration’s attempts to reverse harassment-related protections for transgender workers under Title VII.²²⁴

A. Did States Fill Any Gaps?

State legislation partially filled gaps left by federal law. While state legislators took a broad approach to introducing harassment and gender equity reform, they took a narrower approach to amending harassment law. While we did see efforts to combat noted gaps in protection such as mandatory arbitration at both the state–and even federal level–these efforts are typically limited to banning mandatory arbitration of sexual harassment claims—leaving harassment on other bases, as well as other discrimination claims, unprotected. This creates challenges for the plaintiffs experiencing harassment based on multiple categories or those experiencing both discrimination and harassment, which often occur together.

At the same time, state legislatures seriously attempted—and, in many cases, succeeded—in expanding the list of protected classes protected by anti-discrimination law. For example, some states have expanded protections to cover workers excluded from Title VII coverage, such as those working for small employers, independent contractors, and unpaid interns, among many others.²²⁵ Some have also expanded liability and remedies for harassment by removing Title VII’s affirmative defense against harassment as well as its caps on compensatory and punitive damages for victims of harassment.²²⁶

The most unexpected finding came from the Enforcement bill category. Enforcement was the third most introduced bill topic, suggesting stronger legislative support than expected. Moreover, when we considered secondary topics, the number of Enforcement bills nearly doubled. This suggests a strategy among state legislators to increase access to the courts and to strengthen relief available for harassment and gender equity issues, including pay equity, leave and accommodation, and anti-discrimination law. As a result, there will likely be an influx of the new plaintiffs who are able to have their day in state court. While this is generally good news, it also highlights the continuing and glaring failure of the judiciary to remedy the harassment that is experienced and brought before the courts. Expanding coverage to additional workers and increasing their access to the courts will only work if they are encountering a judiciary that is responsive to power hierarchies and realities of the workplace, while also being serious about curtailing harassment. It is clear that there is still work to be done in this respect, and we look forward to conducting further research examining the topical granularity of the Enforcement bills in our database and how future clams fare in court.

On the federal level, several bills failed in their attempt to remedy the judicially created harassment doctrine outlined in Part II. House Bill 8698, for example, introduced in October 2020, sought to amend Title VII to overturn Supreme Court jurisprudence that is unfavorable to victims of harassment, including Alexander v. Sandoval,²²⁷ Vance v. Ball State,²²⁸ and Faragher/Ellerth.²²⁹ This bill would enhance enforcement by allowing the plaintiffs to sue under Title VII based on evidence of disparate impact, strengthening the plaintiff’s ability to sue under a theory of vicarious liability, and reducing the likelihood of judicial deference to employer anti-harassment policies and procedures, despite ineffectiveness.²³⁰ The bill also created a broader exception to the Federal Arbitration Act for all employment-related rights and remedies under federal and state law, not just sexual harassment.²³¹ It would also have expanded available remedies by allowing the plaintiffs to collect pre-trial attorney’s fees.²³² House Bill 8698 would prohibit employment discrimination on the basis of sexual orientation and gender identity,²³³ a proposal that was ultimately rendered moot by the Supreme Court’s 2020 decision in Bostock v. Clayton.²³⁴

Knowing the effect of the #MeToo movement on legislative agendas and bill passage rates is important for those who wish to maintain and expand the momentum that was fueled and, in some cases, created by these movements. More specifically, legislators and other legal decisionmakers, including the courts and government agencies, should make workers’ voices central to their work. Following the lead of worker organizers whose anti-harassment advocacy efforts predated the groundswell of attention brought by the #MeToo movement will not only bring consistency, but also timeliness and specificity to the anti-harassment agendas set by various legal stakeholders. Below, we outline four specific issues that we believe should be central to advocacy efforts in particular, due to their deleterious effects on workers most vulnerable to workplace harassment.

B. Legal Gaps in Need of Greater Attention

While the past five years following the #MeToo movement have seen reforms that were more systemic and bipartisan than anticipated, major gaps persist, allowing harassment to remain a threat. Harassment and gender equity reform is particularly in need of (1) more systemic changes with greater oversight, transparency, and accountability, (2) greater focus on intersectional harms, (3) well-informed enforcement efforts by judges, attorneys, and agencies alike, and (4) more creative strategies to deal with retaliation.

Symbolic Action and Individuation of Harassment Continues, and Even Worsens?

Our analysis revealed a substantial number of reforms that were symbolic in nature, which encourages institutional “window dressing,” to merely signal enhanced rights, but without effective policies or cultural shifts.²³⁵ Some states created days of honor or remembrance, without adding substance of additional protections or enforcement, which we do not expect to make a meaningful difference beyond raising public awareness about harassment and other gender equity issues. Others condoned behavior of a government official or stated policy support for an issue, without actually making changes that change circumstances for women on the ground. Many states also created task forces and commissions to study issues of harassment and gender equity, but subsequent legislation would be needed to effectuate any changes proposed by these groups. Symbolic reforms may even be counterproductive to the extent they “check the box” by passing a related bill, creating the appearance of having addressed the problem. This can provide political cover while failing to implement more meaningful systemic reform.

Even more troubling is the continuing popularity of bills imposing harassment training requirements. While these bills can have substantive impact, research has shown that unless certain conditions—such as transformative leadership or bystander-specific training—are met, trainings are likely to be ineffective at promoting changes in workplace behavior. Moreover, training requirements can even be counterproductive. Legal stakeholders, thus, need to be aware of how training requirements tend to individuate harassment, making it about individual perpetrators rather than broader shifts in workplace culture, power dynamics, and worker dignity that are required. Training, grievance procedures, and investigations that focus on damage control may obscure the systemic patterns so often present in harassment and discrimination cases. In the current landscape, given the central role of the Faragher/Ellerth defense, these employer practices and procedures do more to insulate the employer from liability than they do to mitigate workers’ risk of harassment or other gender-based harms.

Greater Focus on Intersectional Harms

Congress and state legislatures can, and should, also introduce more legislation protecting specific industries where women of color are overrepresented. This includes not only domestic workers, but also independent contractors, farmworkers, healthcare, hospitality, retail, and restaurant workers.²³⁶ Although state legislatures have stepped up to fill substantial federal gaps in coverage post #MeToo, barriers to enforcement remain and limit potential impact. To be effective, more systemic reform efforts are required that go beyond harassment law to also broaden protections across legal doctrines to the benefit of all women workers. This means, in particular, supporting state and federal legislative agendas that close gaps for the most vulnerable women workers, including low-wage workers, women of color, workers with disabilities, and LGBTQIA+ workers.

Importantly, this requires moving away from the narrow interpretation of sexual harassment by judicial precedent and envisioning a new legal agenda around gender equity reform that is responsive to the reality of our evolving workplace and society. Doing so will require not only our policymakers, but also agencies and courts, to internalize gender inequity—rather than sexual desire—as the foundation of workplace harassment, highlight how it is exacerbated by low-wage employment, and acknowledge how its impact, in many cases, compounds for women with additional protected characteristics, in precarious work, segregated working conditions, and occupationally isolating jobs.²³⁷

Navigating an Employer-Friendly Judiciary

As aforementioned, our legislative analysis suggests that plaintiffs will have greater access to the courts in the near future; however, we must further question and examine what that means if the judiciary continues to constrain harassment law and gender equity doctrine. This is a key moment for employment and civil rights lawyers to advocate effectively for their clients and in doing so, set new precedents, for the court. Doctrinal areas in need of innovation include the “severe and pervasive” requirement, the “objectively” hostile or abusive standard, and the Faragher/Ellerth affirmative defense.

Dealing with Retaliation is Required for Reforms to Work

Strengthening judicial enforcement is also important in the context of retaliation, which continues to go hand in hand with harassment. Despite high rates of retaliation and its role in deterring victims from speaking up and using existing protections, state legislative agendas failed to pass measures that would more effectively deal with this issue. Solving the problem of retaliation requires more than changes to anti-retaliation laws; it also requires stronger enforcement that would incentivize systemic changes to workplace culture. In addition to legislation strengthening anti-retaliation protections generally, more bills should focus on identifying and mandating what specific behaviors are considered retaliatory. These reforms could include retaliatory actions such as substantive changes to terms of employment, disclosing personnel files, contacting immigration authorities, or threatening to report an employee’s immigration status.²³⁸ Doing so would signal the legislature’s recognition that retaliation is rooted in systemic power imbalances and is particularly threatening for workers in low-wage industries or workers with intersectional identities including race and national origin.

Conclusion

Considerable legislative progress has been made in the first five years after #MeToo went viral. Overall, more harassment and gender equity reform has occurred than expected, and through a more varied and comprehensive approach than first predicted. This is particularly true at the state level where new laws also offer greater coverage and enforcement remedies under civil rights statutes regardless of worker classification and with consideration for their occupational hazards and intersectional vulnerabilities.

While it is clear that the #MeToo movement moved the law forward with respect to harassment and gender equity, more work needs to be done, particularly with respect to centering the interests of workers most vulnerable to harassment. Various institutions–including judges, lawmakers, and agencies–contributed to the dissonance between the legal enforcement of, and the realized experience of, workplace harassment. This dissonance is then distributed widely via media, leading to public apathy and confusion about the current landscape of workplace harassment and related gender equity issues. These legal stakeholders thus have a responsibility to ensure that access to justice is equitably and consistently distributed to all workers, especially those most vulnerable. Fulfilling this responsibility grows more

urgent everyday under an increasingly hostile second-term Trump administration.

This does not mean, however, that effective and responsive reform efforts are out of reach. Legislators and regulators alike should consult anti-harassment worker organizers in crafting future reform. Doing so will enable legal branches to establish greater consistency in their treatment of harassment and remain more up to date with the rapidly changing realities of workplace harassment. This, in turn, will have the desired effect of closing the gaps between how harassment is enforced on the books, how it is enforced in the workplace, and how it is experienced by working women.

Federal actors should also make fruitful use of state-level innovation to ensure that state-level successes achieve a broader impact. Now that states have provided evidence of bipartisan support for these issues, Congress and federal agencies including the EEOC, the Department of Health and Human Services, and even the Occupation Safety and Health Administration, should step in and offer a more systemic approach that covers the multitude of gender equity issues that contribute to harassment.

Lastly, our goal is for our dataset to be used as a public resource, from which further research can be developed with the hope of providing concrete findings regarding the effectiveness of state-specific harassment and gender equity legislation. In turn, this research can be used to bolster the advocacy efforts by worker organizers. Doing so will enable legal stakeholders to remain in conversation with those best situated to determine the needs of a given workplace with respect to preventing harassment: the workers themselves.

Appendix A. Search Terms Used to Identify #MeToo Related Legislation
Abusive Work Environment	Gender	Pay Parity	Sexual Orientation
anti-SLAPP	Gender Expression	Predispute Arbitration	Sexual penetration
Antidiscrimination	Gender Identity	Pregnant Workers	Sexual Violence
Compensation History	Gender Representation	Public Right to Know	Sodomy
Confidentiality Agreement	Gratuities	Rape	Stalking
Confidentiality Clause	Harassment	Retaliation	Statute of Limitations
Confidentiality Provision	Harassment Complaints	Salary Experience	Statutory Right
Cyberbullying	Harassment Prevention	Salary History	Unauthorized Disclosure
Denim Day	NDA	Sex	Wage Disclosure
Discrimination	Non-biased Compensation	Sexual Abuse	Wage Discrimination
Discrimination Complaints	Non-consensual	Sexual Arousal	Wage Disparities
Domestic Violence	Nonconsensual dissemination	Sexual Assault	Wage Disparity
Equal Pay	Nondisclosure Agreement	Sexual Assault Awareness	Wage History
Equal Rights	Nondisparagement Agreement	Sexual battery	Wage Secrecy
Equal Rights	Paid Family Leave	Sexual Discrimination	Workplace Bullying
Equal Rights Amendment	Panic Button	Sexual Gratification	Workplace Climate
Ethics Violation	Panic Device	Sexual Harassment	Workplace Misconduct
Family Leave	Pay Disparity	Sexual Intimidation	Workplace Protections
Forced Arbitration	Pay Equity	Sexual Misconduct

Appendix B. Search Terms Used on Bill Corpus to Verify Topic Coding
Topic	Search Terms
Anti-Discrimination	“gender identity” “gender expression” “sexual orientation” “pregnan” “famil* status” “marital status” “famil* responsibility”
Equal Rights Amendment	“equal rights amendment” “ratification”
Harassment Training/Policies/Procedures	“sexual harassment training” “sexual harassment prevention” “complaint process” and “complaint procedure”
Leave/Accommodation	“paid family leave” “safe” “paid sick leave” “domestic violence” “accommodation” “pregnan*” “adoption” and “foster”
Mandatory Arbitration	“mandatory arbitration” “forced arbitration” “arbitration agreement”
Occupational Protections	“domestic” “hotel” “farm” “janitor” “artist” “panic” “bill of rights” “entertainment” “gratuit*”
Pay Equity	“wage discrimination” “pay dispar” “equal pay for equal work” “equal pay” “salary history” “wage history” and “pay equity”
Enforcement	“severe” “pervasive” “statute of limitations” “punitive damages” “compensatory damages” “damages” and “tort”
Regulates Government Officials/Lobbyists/State Contractors	“lobbyist” “state contract” “legislator” “members of the legislature” “legislative staff” “public officials” and “expel”
Transparency/NDAs	“disclosure” “employment history” “non-disclosure agreement” and “nondisclosure agreement”
Other	“task force” “study” “high-wage, high-demand” “law enforcement” and “boards”

98 S. Cal. L. Rev. 761

Download

Jamillah Bowman Williams*

* Professor of Law & Faculty Director, Workers’ Rights Institute, Georgetown University Law Center; B.S. 2003, Georgetown University McDonough School of Business; M.A. 2005, University of Michigan; Ph.D. 2013, Stanford University; J.D. 2011, Stanford Law School. Thank you to Angelica Sanchez Diaz, Nick Gonzales, Keniece Gray, and Lolade Akintunde for research support. Thank you to participants at the Lutie Lytle Workshop for providing valuable feedback.

Elizabeth C. Tippett†

Professor of Law & James O. & Alfred T. Goodwin Senior Fellow, University of Oregon School of Law; B.A., 2002, Harvard College; J.D. 2006, Harvard Law School.

Anu Ramdin‡

Research Fellow, Georgetown University Law Center; B.A. 2018, University of Miami; J.D. 2022, University of California, Berkeley School of Law.

Familial DNA and Due Process for Innocents

July 2025July 2025Erin Sheley*

Ever since genealogical DNA unmasked the Golden State Killer in 2018, the use of this new forensic science has been ubiquitous. Police have solved several hundred cold cases by uploading unidentified crime scene DNA samples to consumer genealogical databases and, with expert assistance, building out family trees from the resulting matches to identify suspects. This practice has raised significant concerns among privacy scholars due to the intimate nature of DNA information and the lack of consent by the parties ultimately identified through their relatives’ choice to upload data. Critics argue that the Fourth Amendment bars law enforcement from running warrantless familial DNA searches using consumer databases, even among users who have given their consent. Several states have also passed legislation banning or severely restricting such searches.

This Article argues that the current debate around familial DNA has neglected the due process rights of another person: the innocent third party who is potentially convicted in the absence of accurate, exculpatory DNA evidence. Both the Compulsory Process and Due Process Clauses give a defendant the right to obtain exculpatory evidence in their defense through judicial subpoena and from the government itself under Brady doctrine. In close cases, the respective scopes of competing constitutional rights should be defined so as to not excessively burden one another. On that basis, this Article argues against any ban on, or categorical constitutional or statutory rule requiring, probable cause for law enforcement to conduct familial DNA searches. The privacy interests raised by critics are legitimate; yet the functional inability for law enforcement to identify unknown suspects due to the lack of probable cause disproportionately burdens defendants whose Compulsory Process and Due Process rights are already limited by existing materiality requirements and inadequate resources. For many innocent defendants, the State itself is the best and only source of investigative resources and potentially exonerating evidence.

Introduction

In 2019, after Christopher Tapp had served nearly twenty years of a sentence for the rape and murder of his eighteen-year-old friend Angie Dodge, his conviction was vacated.¹ An Idaho jury convicted him in 1998 largely due to a series of interrogations during which the police attempted to get him to implicate various friends, including through the offer of an immunity deal, which was later withdrawn when it was discovered that one friend had an alibi.² During questioning, police—including a former guidance counselor of Tapp’s, whom he personally trusted—threatened him with the death penalty and told him that he could not remember the murder because he was repressing it.³ Tapp was convicted despite DNA tests that had excluded both him and the initially-targeted friend as the source of semen found at the scene of the crime.⁴

A decade after the conviction, the Idaho Innocence Project took up Tapp’s case and requested DNA testing on hairs found on the victim’s body, which had only been visually inspected previously.⁵ The FBI lab found that the DNA on the hair was consistent with that of the semen, meaning that the hair also excluded Tapp. In rejecting Tapp’s lawyers’ requests for further testing, an appeals judge found that “while DNA testing may be relevant in identifying one of the assailants, such does not make it more probable that Tapp is innocent.”⁶

Ultimately, the Idaho Falls police—now with the public support of the victim’s mother—worked with the Innocence Project and the private company Parabon Nanolabs to identify the source of the DNA sample.⁷ Parabon created a genetic profile from the sample and compared it with profiles submitted to various consumer ancestry databases.⁸ Starting with profiles in the database, Parabon was able to use other records to build a family tree that turned out to include a man named Brian Dripps, who had lived across the street from the victim and had even been questioned during the early investigation.⁹ Using a discarded cigarette butt collected by the Idaho Falls police, Parabon compared Dripps’s DNA with the sample from the crime scene and was able to identify Dripps as the source of the DNA.¹⁰ Dripps confessed to the murder and rape, saying that he had acted alone and did not even know Tapp.¹¹ Christopher Tapp’s two-decade nightmare was over.

Tapp’s case was at the vanguard of the latest development in forensic science: law enforcement use of familial DNA mapping to identify crime scene samples against the profiles voluntarily uploaded by perpetrators’ relatives to direct-to-consumer (“DTC”) genealogical sites. Starting with the apprehension of California’s infamous Golden State Killer in 2018, familial DNA has been responsible for the police solving a series of high-profile crimes in recent years, including the 1981 New Hampshire murder of Laura Kempton by Ronney James Lee; the 1984 murder of fourteen-year-old Wendy Jerome by Timothy Williams in Rochester, New York; and Sherri Papini’s false claims of kidnapping and sex trafficking by fictitious Latina women in Shasta County, California in 2016.¹²

While law enforcement use of DTC databases has brought resolution to many cases, it has also been the source of significant consternation among privacy advocates and scholars. At the start of the DTC revolution, users were not necessarily aware that law enforcement might be able to access their DNA profiles.¹³ In the current moment, even the most law-enforcement-friendly DTC companies explicitly allow users to “opt in” or “opt out” of sharing their information with police.¹⁴ Even so, many argue that such consent should not include third-party relatives eventually identified through genetic genealogy because these relatives have no control over whether their family members have uploaded their own DNA to a DTC site.¹⁵ As states have begun to regulate in this area, many have explicitly allowed for police access to DTC sites.¹⁶ However, several states have passed statutes restricting law enforcement use of familial DNA. The District of Columbia and Maryland wholly ban familial DNA searches of police databases; in Maryland’s case, the ban extends even to searches by private citizens (and, thus, criminal defendants).¹⁷ Montana, meanwhile, prohibits police from conducting familial DNA analysis using either DTC or police DNA databases without probable cause, which may be very difficult to establish in many cases involving an unidentified sample.¹⁸

The constitutionality of familial DNA searches is governed by the Supreme Court’s most recent word on the Fourth Amendment in an era of evolving technology—Carpenter v. United States.¹⁹ Carpenter invalidated the government’s use of a subpoena (rather than a warrant) to access a subject’s cell phone location data from their service provider, a practice it had previously justified under the “third-party doctrine” of the Fourth Amendment.²⁰ The Court held that a person has a reasonable expectation of privacy in the totality of their movements, and that, because carrying a cell phone is effectively involuntary in the modern age, such an expectation is not defeated by the fact that a user’s location data is shared with third-party service providers.²¹ Critics of unfettered law enforcement access to DTC DNA databases and of familial DNA searches generally point to the increasing ubiquity of both public and private databases and the large number of strangers who can be involuntarily identified through a fairly small number of users.²² They argue that a person has a reasonable expectation of privacy in their DNA against such non-consensual, indirect identification.²³

This Article is the first to examine this question with regard to a forgotten constitutional stakeholder: the potentially innocent “other” suspect who, like Christopher Tapp, familial DNA might exonerate. While critics on both sides tend to balance the privacy rights of targeted parties against the crime-solving interests of the government, the constitutional framework is more complicated than this given the due process rights of these potential defendants. In Part One, this Article explains the science of forensic DNA generally, familial DNA specifically, and their roles in investigations and exonerations. Part Two presents the Fourth Amendment background to this problem and summarizes the argument that courts should recognize a reasonable expectation of privacy that would prohibit warrantless familial DNA searches, even where DTC users give consent. Part Two also asserts that while this argument is colorable, it is weak. This is due to both standing problems and the fact that the privacy interest in not being identified as the specific depositor of a single DNA sample is not comparable to the interest in the privacy of one’s daily movements.

Part Three argues that a falsely accused suspect has a similarly colorable due process right in the availability of familial DNA testing. Such a right derives from three sources: the Compulsory Process Clause, the due process right to exculpatory evidence, and the due process right to post-conviction relief procedures. This Part argues both that defendants themselves should not be legally barred from running familial DNA searches in public or private databases and that the practical realization of their due process rights requires that law enforcement not be so barred either. Part Four concludes that, in close cases, competing constitutional rights must not excessively burden one another. On that basis, this Part argues against any categorical rule substantially preventing law enforcement from conducting familial DNA searches. While legislatures are well-advised to limit the use of this technology to serious cases in which other investigative techniques have failed, they should not adopt bans or probable cause requirements; if they do, they should at least create an explicit process for defense subpoenas. Furthermore, Part Four concludes that courts should not extend Carpenter to create a categorical warrant requirement for familial DNA searches of police databases or DTC databases where an initial user has given consent to law enforcement access.

I. The Developing Practice of Forensic DNA

A. DNA Science

Deoxyribonucleic acid (“DNA”) is a molecule found in cells that contains the genetic code for an organism’s development, function, growth, and reproduction.²⁴ DNA is hereditary, meaning it is passed from parent to child.²⁵ The information in DNA is stored as a code made up of four chemical bases: adenine (“A”), guanine (“G”), cytosine (“C”), and thymine (“T”).²⁶ Human DNA consists of about three billion bases and more than 99% of those bases are identical across the species.²⁷ Their sequence determines the information available for building and maintaining the organism.²⁸

Variations in code can occur at any point in the genome, including non-coding areas.²⁹ Scientists have discovered that these non-coding regions contain repeated units of DNA that vary in length between individual subjects; one particular type of repeat, the short tandem repeat (“STR”), is easily measured and compared for identification purposes.³⁰ The FBI has identified thirteen areas, or “loci,” on a chromosome where STRs are found, which U.S. law enforcement now uses to identify individuals—for example, to identify remains, determine paternity, or match a suspect to a forensic sample from a crime scene.³¹

DNA identification involves the use of statistics.³² The FBI has established the frequency with which each form, or “allele,” of the thirteen core STRs naturally occurs in people of different ethnicities.³³ In a specific case, a lab will determine the allele profile of the thirteen core STRs for both the crime scene sample and the suspect’s sample. If they do not match, the suspect is excluded.³⁴ If they have matching alleles at all thirteen STRs, it then becomes possible to make a statistical calculation to determine the frequency with which that genotype arises in the population.³⁵ The probability, for example, of two unrelated Caucasians having identical DNA “fingerprints” is about 1 in 575 trillion.³⁶ Because there are 5,000 trillion pairs of people out of the 100 million Caucasians in the world, roughly eight or nine pairs would be expected to match at the thirteen STR loci.³⁷

Given the exceedingly low likelihood of any given person matching the DNA profile associated with a crime scene, the probabilistic science of DNA has generated some of the most reliable evidence known to our justice system.³⁸ This does not mean that it is always perfect evidence. DNA samples may be small in quantity, improperly preserved, or highly degraded (meaning that analysts can only obtain a partial profile).³⁹ Furthermore, some crime scene samples contain DNA from multiple sources. All of these issues can confound the effectiveness of DNA fingerprinting as a means of identification. However, in cases in which all thirteen STR loci can be examined and matched, such matches are extraordinarily reliable.⁴⁰

The use of DNA science in crime solving has been facilitated by the rise of DNA databases maintained by law enforcement. The DNA profiles in these databases come primarily from two sources: (1) DNA samples taken from crime scenes where they can be obtained (for example, “rape kits”) and (2) samples taken from convicted offenders and, in some states, arrestees. All fifty states statutorily require at least some offenders, especially those convicted of sexual or violent crimes, to submit samples.⁴¹ In addition, thirty states statutorily authorize law enforcement to collect DNA samples from those arrested for certain types of crimes, usually felonies.⁴² The FBI maintains the Combined DNA Index System (“CODIS”), a program of support for law enforcement DNA databases nationwide.⁴³ It includes the National DNA Index System (“NDIS”), which contains the DNA profiles contributed by federal, state, and local participating forensic laboratories.⁴⁴ As of February 2024, CODIS had assisted in over 680,122 criminal investigations.⁴⁵ One study of Danish data found that police collection of criminal offenders’ DNA profiles increases detection probability and reduces recidivism rates over the next year by as much as 43%.⁴⁶

B. DNA and Exonerations

One of the most important developments flowing from the rise of DNA evidence since the mid-1990s has been its potential for exonerating the wrongfully accused and convicted.⁴⁷ A 1996 study found that as many as 25% of the cases sent for DNA analysis in the early days of the technology ended up excluding the primary suspect, demonstrating the significance of DNA technology in avoiding potential wrongful convictions.⁴⁸ Between 1989 and 2020, the Innocence Project reports that 375 wrongfully convicted prisoners were freed due to post-conviction DNA testing.⁴⁹ This is, unfortunately, due not only to the accuracy of DNA evidence but also to the unreliability of other forms of evidence that have been contributing to false convictions since time immemorial.⁵⁰

One significant factor in wrongful convictions has been the problem of false eyewitness identifications. Sixty-nine percent of DNA exonerations tracked by the Innocence Project involved eyewitness misidentification.⁵¹ This is perhaps unsurprising due to the role of trauma in distorting witness recollections over time⁵² and the fact that police practices around line-ups and show-ups can often—even unintentionally—prove overly suggestive and encourage bad identifications.⁵³ Cross-racial identifications are notoriously less reliable than identifications made by witnesses of the same race as the suspect.⁵⁴

False confessions, as exemplified by Christopher Tapp’s case, have been identified as a factor in over 25% of DNA exonerations.⁵⁵ On average, people who confess to crimes they did not commit have been interrogated for sixteen hours or more.⁵⁶ Other factors associated with false confessions include deceptive practices by police, such as lying about evidence, isolation, intimidation, and force.⁵⁷ Whatever the particular causes, the problem of wrongful conviction has been found to disproportionately affect Black suspects.⁵⁸ The role of DNA in exonerations thus suggests that it improves not only accuracy but also equality in the criminal justice system.

Responding to the implications of DNA science for the wrongfully convicted, all fifty states have enacted statutes dealing with defendants’ access to DNA evidence post-conviction.⁵⁹ Such statutes generally provide not only a right to post-conviction testing under certain circumstances but also an easing of traditional rules of finality in cases where the results demonstrate innocence.⁶⁰ At the federal level, the Innocence Protection Act provides for post-conviction DNA testing in federal cases, and the Justice for All Act creates financial incentives for states to provide for post-conviction DNA testing.⁶¹

Increased public attention to the problem of wrongful convictions has also prompted several jurisdictions to create Conviction Integrity Units (“CIUs”) within their prosecutor’s offices to monitor and investigate potentially wrongful convictions.⁶² The creation of the CIU in Dallas County, Texas was the result of the County’s DNA retention policy as well as the election of District Attorney Craig Watkins, who made conviction integrity a particular priority.⁶³ A nationwide study of data from the National Registry of Exonerations between 1989 and 2016 even found that Dallas County was the geographic region with the greatest likelihood of a DNA exoneration occurring.⁶⁴

C. Familial DNA

An obvious limit to the use of forensic DNA in crime-solving is the fact that the universe of potential matches for crime scene samples comprises only profiles already stored in existing databases. In cases where an unknown perpetrator has no prior record or other reason to have a profile stored, DNA has been less useful. These limitations seemingly dissolved for the first time when police apprehended the Golden State Killer (“GSK”), Joseph DeAngelo, in 2018.

The GSK was a serial rapist and murderer who terrorized California across six counties in the 1970s and 1980s.⁶⁵ Escalating from peeping Tom behavior to burglary and finally to home invasion rapes and murders, which often seemed to target couples, the GSK was responsible for at least thirteen murders and over fifty rapes.⁶⁶ While initially famous for the horrific nature of his offenses and the length of his escape from justice, DeAngelo—a former police officer—would become equally famous as the first defendant identified through familial DNA.⁶⁷

Prosecutors took genetic material preserved from the rape kits of GSK’s victims and first sent it to FamilyTreeDNA, a DTC testing company.⁶⁸ DTC companies like FamilyTreeDNA allow customers to submit their own DNA through saliva samples and receive genetic information such as countries of origin, health risks, and the names of relatives who have also submitted to the database.⁶⁹ Therefore, they contain a different and more varied range of profiles than those available in CODIS as they are unrelated to known crimes. Furthermore, the genetic profiles collected in genealogy databases, instead of consisting of STRs, are unique single nucleotide polymorphism (“SNP”) profiles.⁷⁰ These are more evenly distributed through a person’s genome than STRs and can thus carry information about a person’s physical appearance that would not be available from a profile in CODIS.⁷¹

FamilyTreeDNA created a DNA profile for the GSK’s sample which officers hoped would yield the identities of close family relations and, thus, enable them to narrow their search.⁷² That attempt yielded only distant relations and proved unhelpful in identifying the killer.⁷³ However, a civilian genealogy expert assisting the investigative team uploaded the DNA profile to another DTC company, MyHeritage, using her own personal user profile.⁷⁴ She also uploaded it to GEDmatch, a then-public site to which users could upload their DNA profiles and seek out family members for free.⁷⁵

The MyHeritage search yielded a pool of second cousins of the killer, one of whom investigators visited at her home in Orange County. She then voluntarily provided a DNA sample, which showed that the killer was related to her through other family members investigators had discovered on a third DTC site, Ancestry.com.⁷⁶ This narrowed the pool of suspects down to six men, only one of whom, Joseph DeAngelo, had blue eyes—a feature already known from the original DNA profile. After ten days of surveilling DeAngelo, investigators seized DNA-bearing evidence from his trash can, finally proving him to be the killer.⁷⁷

The GSK story raises obvious privacy concerns around the access afforded to law enforcement by the companies themselves in the first place. Immediately after DeAngelo’s arrest, an investigator confirmed only that the officers had uploaded the rape kit profile to the open-source GEDmatch site.⁷⁸ For a time, the FBI treated the three private companies as privileged confidential sources, instructing California investigators that they were not to reveal their identities as they assembled charges against DeAngelo.⁷⁹

As it turns out, the companies had widely divergent views about assisting law enforcement. FamilyTreeDNA actively assisted the FBI, giving them access to the site for investigative purposes without knowing specifically which case they were investigating.⁸⁰ Their terms of service at the time contained a warning that the company could be required to release users’ personal information in response to a “lawful request by public authorities,” and their CEO stated that he did not believe assisting the police violated this policy.⁸¹ By contrast, MyHeritage’s privacy policy contained the stronger language that information would only be released “if required by law” (suggesting the company would only release information to law enforcement in the face of a warrant or at least a subpoena).⁸² Because MyHeritage did not assist law enforcement directly but merely sold its services to a purported consumer, it arguably did not violate its own policy. Its marketing executive noted that their privacy policy “did not explicitly” address this form of access and said “[i]t is possible that the civilian geneticist thought she was not violating our terms of service.”⁸³

In the wake of the interest generated by the GSK case, most DTC companies have revised their policies to strengthen and clarify the degree of privacy to be expected vis-à-vis law enforcement. For example, MyHeritage now stipulates that “[w]e will not provide information to law enforcement unless we are required by a valid court order or subpoena for genetic information.”⁸⁴ The popular site 23andMe.com makes an even stronger statement, promising “to use all practical legal and administrative resources to resist requests from law enforcement,” though noting that under some circumstances the company “may be required by law to comply with a valid court order, subpoena, or search warrant for genetic or personal information.”⁸⁵ Furthermore, most sites now prohibit users from uploading DNA belonging to someone other than themselves or a dependent.⁸⁶ Even these strengthened policies leave somewhat open-ended the question of how hard companies will resist police requests and, specifically, whether they will take the hardline position of companies like Apple, who refuse to hand over customer data in the absence of a warrant supported by probable cause.⁸⁷ Furthermore, it is difficult to imagine how genetic companies could prevent users from uploading a third party’s DNA sample under their own name.⁸⁸

There are also two notable outliers in the privacy trend in consumer genetics. GEDmatch (now owned by the Qiagen Corporation) has taken a different approach, allowing law enforcement to upload DNA samples in cases of murder, nonnegligent manslaughter, aggravated rape, robbery, aggravated assault, or when there is a need to identify a dead body.⁸⁹ While users must actively “opt in” to have their own DNA profiles matchable by law enforcement, the site actively encourages them to do so, exhorting that users can help “provide answers to those with missing loved ones” and “enable law enforcement to solve violent crimes and exonerate the falsely accused.”⁹⁰ The GEDmatch site features profiles of crime victims identified and murders solved, along with the story of Christopher Tapp.⁹¹ FamilyTreeDNA has remained the most friendly to law enforcement, now clarifying on its site that it will allow police to create profiles to help identify bodies or perpetrators in violent crimes and requires users affirmatively to “opt out” of their data’s inclusion in such searches.⁹² The state of California now legally requires consumer genetic companies to obtain consent in this manner.⁹³

As of the end of 2022, 545 cases have been solved using familial DNA.⁹⁴ Another dimension of consumer genetic databases worth noting is their demographic differences from CODIS. Black Americans appear to be overrepresented in CODIS relative to the general population.⁹⁵ By contrast, people of Northern European ancestry are more heavily represented in consumer databases (perhaps unsurprisingly due to the uncertainty of precise ancestral origins among people of vaguely European descent).⁹⁶ One study found that at least 60% of Americans of European descent may be identifiable through a genealogical database of just 1.3 million people.⁹⁷

II. The Privacy Argument for Restricting Familial DNA

This Section will consider the privacy interests implicated by law enforcement use of familial DNA, starting with the Fourth Amendment framework protecting privacy. It will summarize the scholarly commentary arguing that warrantless familial DNA searches may violate the Fourth Amendment, as well as new state statutory protections that limit such searches. Finally, it will argue that police conducting familial searches of their own databases or of DTC databases with the consent of genetically related users raises colorable but ultimately quite weak Fourth Amendment arguments.

A. The Constitutional Dimension

The debate over familial DNA occurs at a time of high doctrinal instability around the meaning of the Fourth Amendment in a world of constantly changing technology. The Fourth Amendment provides that:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.⁹⁸

The Founders adopted the Amendment, which mirrored parallel provisions in state constitutions at the time, in response to the Crown’s use of general warrants in the Colonies.⁹⁹ Courts issued these warrants allowing Crown agents to search and seize evidence with no limitations.¹⁰⁰ In the infamous cases Entick v. Carrington (1765) and Wilkes v. Wood (1763), the Crown used these broad warrants to arrest two pamphleteers critical of the government and to seize their books and papers.¹⁰¹ Ultimately, Carrington and Wilkes won trespass suits against the relevant government.¹⁰²

With this brief history as backdrop, most early cases construing the Fourth Amendment understood it to prohibit only warrantless, trespassory interferences with property, allowing auditory surveillance by law enforcement where they did not physically intrude on the defendant’s property.¹⁰³ In the watershed 1967 case United States v. Katz, however, the Court made a change.¹⁰⁴ In Katz, FBI agents had attached a listening device to the outside of a public telephone booth in which the defendant was making a call.¹⁰⁵ Explicitly overruling prior precedent, the Court held that this constituted a warrantless search and announced a new rule: “[w]hat a person knowingly exposes to the public, even in his own home or office, is not a subject of Fourth Amendment protection. . . . But what he seeks to preserve as private, even in an area accessible to the public, may be constitutionally protected.”¹⁰⁶ This new test for the existence of a Fourth Amendment search has become known (based on language in Justice Harlan’s concurrence) as the “reasonable expectation of privacy” test.¹⁰⁷ To apply it, courts ask, first, whether the defendant had exhibited a subjective expectation of privacy and, second, whether that expectation was “one that society is prepared to recognize as reasonable.”¹⁰⁸

While Katz ostensibly renounced the property-based trespass rule of the Fourth Amendment, the subsequent half-century of Supreme Court jurisprudence has seen, nonetheless, a doctrinal struggle between property- and privacy-based conceptions of the right. Some cases have elaborated on the reasonable expectation of privacy test, recognizing, for example, a reasonable expectation against thermographic surveillance by a heat-sensing device set up across the street from one’s home.¹⁰⁹ In other cases, the Court declined to recognize such an expectation against overhead surveillance of property by police aircraft flying at FAA-approved heights¹¹⁰ or in discarded trash set out for collection.¹¹¹ Yet, in still other cases, the Court has held that the Katz reasonable expectations test has been “added to, not substituted for, the traditional property-based understanding of the Fourth Amendment” and therefore need not apply when “the government gains evidence by physically intruding on constitutionally protected areas.”¹¹²

One rule of particular relevance to the context of familial DNA is the so-called “third-party” doctrine, which predated and survived Katz.¹¹³ The third-party doctrine states that a person has no reasonable expectation of privacy in communications they share with a third party—that the law “permits the frustration of actual expectations of privacy by permitting authorities to use the testimony of those associates who for one reason or another have determined to turn to the police.”¹¹⁴ The third-party doctrine has excluded from the warrant requirement scenarios in which an associate of the defendant has voluntarily worn a wire, as well as transaction records the police obtain from the defendant’s bank and call records (so-called “pen” registers) they obtain from the defendant’s phone company.¹¹⁵

Under both the property and privacy theories of the Fourth Amendment, courts have generally recognized the presumption that if police conduct does constitute a search, a warrant is required.¹¹⁶ However, the Supreme Court has also held that some searches that would otherwise be presumed to require a warrant supported by probable cause are nonetheless constitutional because “the touchstone of the Fourth Amendment is reasonableness.”¹¹⁷ In such cases, the Court weighs the defendant’s privacy interests against competing government interests, which usually happens in cases involving either categorical exceptions to the Fourth Amendment or administrative searches.¹¹⁸ The “reasonableness balancing” model of the Fourth Amendment has created categorical exceptions such as stop-and-frisks based only on reasonable suspicion,¹¹⁹ searches incident to arrest,¹²⁰ and police uses of force in cases where the officer’s actions were reasonable based on the facts confronting them.¹²¹ Using such balancing, the Court has also created administrative exceptions for wholly suspicionless searches of parolees, airport security checkpoints, sobriety checkpoints, fire code and public safety inspections, among many others.¹²²

As investigative technology has developed at a seemingly geometric rate, however, the Court has struggled to address all seemingly invasive police behavior under the privacy and property tests—especially in light of the third-party doctrine.¹²³ In United States v. Jones, the Court unanimously held that police engaged in a warrantless search by affixing a GPS tracking device to a suspect’s car to track its movements.¹²⁴ While the majority opinion rested on trespass grounds—that “[t]he Government physically occupied private property for the purpose of obtaining information” by attaching the device,¹²⁵ Justice Sotomayor’s concurrence suggested that, in the future, the digital age might make it “necessary to reconsider the premise that an individual has no reasonable expectation of privacy in information voluntarily disclosed to third parties.”¹²⁶ The Court also distinguished between digital data and other types of searches in Riley v. California, which dealt with the search of a suspect’s cell phone incident to arrest.¹²⁷ The Court concluded that “[c]ell phones differ in both a quantitative and a qualitative sense from other objects that might be kept on an arrestee’s person,” due to the wealth of sensitive information they contain about a person’s browsing history, location, relationships, and so forth.¹²⁸ It thus held that, absent an emergency, police require a warrant to search the contents of a phone otherwise properly seized incident to arrest.¹²⁹

The future that Justice Sotomayor alluded to in Jones seemed finally to arrive in 2018 with Carpenter v. United States, in which the Court made the biggest technology-reactive shift in Fourth Amendment doctrine since Katz.¹³⁰ Carpenter involved the 1994 Stored Communications Act (“SCA”), which authorized the Government to compel private companies to disclose telecommunication records when it “offers specific and articulable facts showing that there are reasonable grounds to believe” that the records “are relevant and material to an ongoing criminal investigation.”¹³¹ While the SCA required that a magistrate make this determination in order for the government to obtain a subpoena, it was on a reasonable suspicion standard rather than the probable cause necessary for a warrant.¹³²

In Carpenter, the government obtained subpoenas under the SCA for weeks’ worth of cell-site location information (“CSLI”) from the defendant’s cell phone providers.¹³³ CSLI data consists of time-stamped records generated several times a minute when a user’s phone connects to a cell phone tower.¹³⁴ While it does not provide as precise of geographic data as GPS, CSLI data from a subject’s phone over a substantial period of time

provides government analysts with a broad picture of that person’s movements.¹³⁵

The government argued that the subpoenas were constitutional under the third-party doctrine; Carpenter had voluntarily shared his location information with his carriers in the same manner a person does their call record or as a person shares their financial transactions with a bank.¹³⁶ The Court rejected this argument, holding for the first time that an individual “maintains a legitimate expectation of privacy in the record of his physical movements as captured through CSLI.”¹³⁷ Importantly (and contrary to some pre-Jones precedent),¹³⁸ the Court held that there was no distinction for these purposes between public and private movements, even if a subject’s public movements would be visible to the naked eye of someone physically following them.¹³⁹

The Court based its rule on five factors specific to the technological context of CSLI data.¹⁴⁰ First, it noted that such data is comprehensive: it provides a record of movement that is “detailed” and “encyclopedic” and constitutes “near perfect surveillance.”¹⁴¹ Second, it is intimate in the sense that a cell phone “faithfully follows its owner beyond public thoroughfares and into private residences, doctor’s offices, political headquarters, and other potentially revealing locales.”¹⁴² Third, it is inexpensive—especially compared to “traditional investigative tools” like the in-person surveillance approved in earlier third-party doctrine cases.¹⁴³ Fourth, it is retrospective in the sense that it allows the government to go back in time and investigate anyone it wishes, without having to determine an investigative target in advance.¹⁴⁴ And, fifth, it is functionally nearly involuntary, due to the fact that cell phones are “indispensable to participation in modern society.”¹⁴⁵

Some scholars have suggested that Riley and Carpenter adopt a “mosaic” theory of the Fourth Amendment—a concept the D.C. Circuit, the lower court in Jones, had originally imported from the national security context and relied on to exclude the GPS tracking data.¹⁴⁶ Under a mosaic theory, a Fourth Amendment search can arise based on police actions taken over time, even if no individual step taken in isolation would constitute a search.¹⁴⁷ Critics such as Orin Kerr argue that the mosaic theory suffers from problems of administrability and an overreliance on a subject’s probabilistic expectations of privacy, which is ill-suited to regulate electronic surveillance.¹⁴⁸ In the wake of Carpenter, lower courts have been divided over whether to formally recognize the mosaic theory as a fundamental rule of the Fourth Amendment.¹⁴⁹

These new iterations of the reasonable expectation of privacy test have sparked broader debate about Katz itself. Many critics have advocated for replacing the Katz test entirely due to its confusing and cyclical nature.¹⁵⁰ Others have argued that modern Fourth Amendment case law is actually driven by consistent principles which can now be applied to evolving technology. Matthew Tokson, for example, argues that the expectation of privacy turns on the intimacy of the area searched, the amount of information sought, and the cost of the investigation.¹⁵¹ These factors, relevant in Carpenter, are worth bearing in mind while considering how courts apply the recent Fourth Amendment precedent to DNA evidence.

B. DNA and the Fourth Amendment

Given that traditional forensic DNA technology is nearly thirty years old, it is unsurprising that much of the Fourth Amendment case law concerning it long predates Carpenter. As discussed above, early use of forensic DNA relied heavily on databases assembled by law enforcement from perpetrators and suspects who were already lawfully in custody. Broadly speaking, “intrusions into the human body” such as blood draws constitute Fourth Amendment searches.¹⁵² Yet in Maryland v. King, the Supreme Court rejected a Fourth Amendment challenge to the Maryland DNA Collection Act (“MDCA”), which allows state and local law enforcement to collect DNA samples from individuals arrested for a crime of violence, an attempted crime of violence, burglary, or attempted burglary.¹⁵³

In King, the Court recognized that the intrusion of the DNA swabbing was a search and engaged in a balancing test to determine whether it was reasonable as an administrative exception to the Fourth Amendment (and therefore not requiring probable cause of a crime likely to yield DNA evidence). Specifically, the Court balanced the defendant’s privacy interest against “the need for law enforcement officers in a safe and accurate way to process and identify the persons and possessions they must take into custody.”¹⁵⁴ Noting that a search incident to arrest is a long-standing exception to the warrant requirement, the Court, nevertheless, did not invoke the doctrine directly (likely because it limits a search to what is necessary to protect officer safety, prevent escape, and preserve evidence).¹⁵⁵ Instead, the Court found that “[w]hen probable cause exists to remove an individual from the normal channels of society and hold him in legal custody, DNA identification plays a critical role in serving those interests.”¹⁵⁶ Finding the intrusion on the defendant’s privacy through the physical act of swabbing for DNA to be minimal, the Court upheld the MDCA. In dissent, Justice Scalia analogized the collection of DNA from arrestees for whom there was no probable cause of a crime of violence to the general warrants of the Colonial era.¹⁵⁷

King left open many questions, such as whether DNA identification analysis of a blood sample originally drawn for a non-identification purpose (such as blood-alcohol analysis) constitutes a search.¹⁵⁸ Lower courts have found some limits to the government’s Fourth Amendment authority to perform DNA analysis on samples they’ve obtained from non-arrest contexts. For example, the Fourth Circuit has held that King did “not give a law enforcement agency carte blanche to perform DNA extraction and analysis derived from clothing lawfully obtained from the victim of a crime in relation to the investigation of other crimes.”¹⁵⁹ Similarly, an Arizona court has held that a defendant’s consent to a blood draw for testing for intoxicants did not authorize the police to create a DNA profile from the blood cells to investigate other offenses.¹⁶⁰ Federal district courts have denied government requests to conduct purported administrative DNA swabs for purposes other than the identification and safety rationales stated in King.¹⁶¹ Others, however, have allowed such explicitly evidence-gathering searches on a showing of reasonable suspicion short of probable cause.¹⁶²

While most of the existing DNA-related precedent involves clear police intrusions on the person of the subject for the purpose of collecting samples, and thus clearly trigger Fourth Amendment concerns under either a privacy or trespass conception, the use of consumer DNA databases do not involve bodily invasion. Furthermore, courts generally do not recognize a property interest in one’s DNA. In Association for Molecular Pathology v. Myriad Genetics, the Supreme Court held that isolated, naturally occurring DNA sequences could not be patented because isolating a gene is not an “act of invention.”¹⁶³ More directly on point, courts have traditionally declined to recognize a property right in one’s own genetic materials that is sufficient to support a conversion action in cases where hospitals have utilized patients’ excised cells.¹⁶⁴ Scholars argue that recognizing property rights in DNA would raise a host of problems ranging from the conceptual¹⁶⁵ to the dignitary¹⁶⁶ to the policy-based.¹⁶⁷ Critics of the status quo argue that the rise of consumer genetic companies, with the accompanying privacy concerns, necessitate property-based tools such as conversion actions to ensure good behavior and privacy practices.¹⁶⁸ At least a couple of courts have recently proved receptive to privacy-based conversion claims related to genetic information.¹⁶⁹ Five states—Alaska, Colorado, Georgia, Louisiana, and Florida—have current legislation granting people ownership of their genetic information.¹⁷⁰ Nonetheless, after Carpenter, the privacy theory of the Fourth Amendment provides a stronger foundation for the argument that familial DNA database searches trigger the Fourth Amendment.

C. The Case for Genetic Privacy post-Carpenter

The proliferating concerns over genetic privacy sparked by the rise of DTC genetics companies have been particularly salient in the law enforcement context. This Section will review arguments that law enforcement use of DTC genetic databases generally and for familial DNA searches specifically raise Fourth Amendment concerns even when third-party users consent to law enforcement access.¹⁷¹ It will also survey the state legislative responses to these concerns.

1. The Scholarly Debate

The scholarly concern over privacy issues raised by forensic DNA generally and familial DNA specifically has been widespread with the weight of commentary suggesting that warrantless familial DNA searches explicitly violate Carpenter or should at least be otherwise prohibited.¹⁷² Scholars have sought to delineate the nature of the privacy harms suffered by persons who are genetically identified through familial searches.¹⁷³ Some argue that privacy violations implicate autonomy interests and result in “people’s inability to make choices in accordance with their preferences,” a consequence with additional negative emotional effects.¹⁷⁴ Furthermore, because genetic surveillance can be both large-scale and untethered to suspicion, it can also chill the exercise of civil liberties and create the risk of discrimination and selective enforcement.¹⁷⁵

Natalie Ram, one of the leading scholars on genetic privacy, argues that, based on the factors the Court identified in Carpenter, courts should recognize a Fourth Amendment reasonable expectation of privacy in one’s own DNA, even when stored in a third-party database.¹⁷⁶ On the question of intimacy, Ram notes that genetic information is “deeply revealing” as well as “highly detailed and precise about the individual information it discloses.”¹⁷⁷ As to comprehensiveness, she notes that the sudden proliferation of consumer genetic databases and the use of genetic information in medical records—while not yet as widespread as cell phone usage—suggests that genetic analysis is a substantially growing part of users’ lives.¹⁷⁸ As to expense, Ram points out that the use of genetic data in investigations is cheap, easy, and efficient, similar to CSLI data.¹⁷⁹ Finally, as to voluntariness, she points again to the sudden proliferation of genetic databases to argue that their use “may be approaching an ‘inescapable and automatic nature’ ” in the way that cell phone use once reached fairly quickly.¹⁸⁰ She concludes that these factors all urge that there be a reasonable expectation of privacy in one’s own genetic information, even when voluntarily shared with third-party databases. Furthermore, she notes that there is a much stronger property interest in one’s own genetic information than in one’s location data—a fact seemingly supported by the user agreements of genetic sites which emphasize the user’s “ownership” of their own data, as well as the statutes in the five states that recognize it.¹⁸¹

With respect to familial DNA, the argument, of course, becomes more complicated due to the fact that an individual’s genetic information could be accessed through a third party voluntarily sharing their information—not only with the database itself but also through “opting in” or choosing not to “opt out” of sharing with law enforcement. Ram argues that in such cases, the target of the investigation who is indirectly identified through a family member’s DNA cannot be said to have voluntarily consented to the search, due to the fact that: (1) genetic relatedness is intrinsically involuntary and (2) people rarely have control over the decision-making of even near relations, much less the distant relations through which investigative matches often occur.¹⁸² She thus concludes that, while a law enforcement search of a consumer database for a direct match, voluntarily shared, may be appropriate, searches for indirect matches through third-party relations is not.¹⁸³

2. Legislative Responses

In the U.S., sixteen states have explicitly allowed law enforcement to conduct familial DNA searches, though often with policies limiting the usage to serious crimes or cases in which the public safety is at risk.¹⁸⁴ Other states, however, have passed statutes greatly limiting such searches. The District of Columbia and Maryland bar familial DNA searches of the states’ own DNA databases.¹⁸⁵ As to consumer databases, Maryland also has the most detailed statute to date, requiring judicial authorization based on a sworn affidavit by law enforcement with approval of the relevant prosecutor.¹⁸⁶ Familial DNA searches will only be authorized when the crime is murder, rape, a felony sexual offense, or an act presenting “a substantial and ongoing threat to public safety or national security.”¹⁸⁷ Furthermore, law enforcement must first attempt to identify a suspect through CODIS, as well as to first pursue “reasonable investigative leads” that ultimately fail to identify them.¹⁸⁸ The law also limits law enforcement to databases that provide explicit notice to users that law enforcement may use it and that “seeks acknowledgement and express consent from its service users” regarding this possibility.¹⁸⁹

Montana’s statute, while more vaguely drafted, appears to severely limit familial DNA searches. It provides that “[a] government entity may not obtain familial DNA search results or search results from partial matching from the [state-maintained] DNA identification index or a consumer DNA database without a search warrant issued by a court on a finding of probable cause.”¹⁹⁰ It has another provision requiring the government to obtain a warrant even for direct searches of consumer database users, unless “the consumer whose information is sought previously waived the consumer’s right to privacy.”¹⁹¹ If Idaho had such a statute during Christopher Tapp’s post-conviction proceedings, he would likely remain in prison, as no probable cause would have existed at the time for the Idaho Falls Police to have assisted his family by searching for the DNA of the neighbor who was ultimately identified. In many cases, police could establish probable cause that the owner of the unidentified sample committed a crime (as, for example, when semen is found in the body of a rape victim). At other times, such as when there are large numbers of DNA samples found at a scene, this might be impossible. Furthermore, probable cause must exist for the specific area to be searched.¹⁹² It would often be a preventative burden for police to have to establish probable cause that relatives of the perpetrator have created a searchable account with a particular ancestry site, which would be necessary to establish probable cause that the database contained evidence of a crime.

D. The Fourth Amendment Argument Against Familial DNA Searches is Weak

Any attempt to evaluate the strength of the Fourth Amendment argument against law enforcement use of familial DNA is necessarily complicated by the current instability of Fourth Amendment doctrine generally. Nonetheless, this Section will show that, whichever Fourth Amendment theory one subscribes to, the argument is fairly weak.

With respect to DTC databases in which users have opted in to sharing with law enforcement, the relevant rule is that consensual searches do not require warrants or any degree of suspicion to be reasonable under the Fourth Amendment.¹⁹³ It is therefore clear that if Person A submits their DNA sample to MyFamilyTree, opts in to law enforcement searches, and is subsequently identified as a match to a crime scene sample, no constitutional issues arise. If the DNA of said Person A leads the investigators indirectly to their Grandpa B, however, Grandpa may argue—with Ram and other scholars—that his privacy rights were violated because he did not consent to the search that led to his DNA. One possible response is that, at a physical level, it was not Grandpa’s genetic sample that was searched. In Fourth Amendment law generally, a defendant does not have standing to object to the violation of a third party’s constitutional rights.¹⁹⁴ If only Person A was searched, Grandpa has no standing to object. This same argument would apply to familial searches of lawfully collected samples in police databases.

Critics argue, however, that the pervasiveness of technology should change existing Fourth Amendment principles and that there is functional identity between one person’s DNA and their family member’s, which should change our understanding of standing in these cases. As discussed above, Ram makes a good argument that genetic data obtained through familial searches meets the Carpenter factors insofar as it is intimate, comprehensive, and inexpensive, as well as involuntary from the perspective of the family member ultimately identified.¹⁹⁵ Yet there are some compelling counterarguments. As to the intimacy of the data, Jasper Monroe-Jones notes that—unlike location or cell phone data—DNA “cannot reveal anything about a person’s actual actions, conduct, or character” but only, if anything, predisposition to various traits or diseases.¹⁹⁶ This point is even stronger when coupled with the fact that law enforcement officials who access consumer DNA sites only access the same identification information as other users, which does not include intimate information such as propensity for physical or psychological diseases.¹⁹⁷ While it would be theoretically possible for law enforcement to reconstruct such intimate information from the information they can obtain, states could pass laws specifically prohibiting this practice, which the U.S. Department of Justice’s current Interim Policy on Forensic Genetic Genealogy already does.¹⁹⁸ In short, a DNA match in a DTC database only provides one significant piece of information: the fact that a person or their family member was present at the place from which a sample was recovered.

Furthermore, critics may overstate the “inexpensiveness” of familial DNA searches—while putting a DNA sample into a consumer database may be easy, the process of building out family trees based on genetic relationships requires an expert in genetic genealogy and may take months.¹⁹⁹ For now, at least, this process is not realistically comparable to looking at a map of cell tower pings. That said, due to the need for Fourth Amendment doctrine to be durable in the face of evolving technology, it should still be assumed that this technology may become less expensive in the future.

With the specific Carpenter factors only partially apposite to the problem of familial DNA, it is worth recurring to the test Carpenter attempts to apply: reasonable expectation of privacy. While it seems clear that Grandpa B cannot consent to Person A’s voluntary release of their genetic material to law enforcement, can it be said that he has a reasonable expectation of privacy in another person’s body? No authority recognizes such an expectation, which would need to be strong enough to overcome standing principles.²⁰⁰ It is a weak argument that, even if there is no reasonable expectation of privacy in personal genetic material left in another person’s bathroom, there should be a reasonable expectation of privacy in another person’s own genetic material that happens partially to match one’s own. As critics of the Katz test’s inherent circularity might note, with the increased public awareness of DNA science, such an expectation of privacy is even less reasonable than it might have been in, say, 1985.²⁰¹

In addition, the preceding sections have made it clear that trespass-based theories of the Fourth Amendment provide even less traction for constitutional objections to familial DNA searches than does the Katz test. Clearly, a search of Person A’s sample (or the data it generates) is not a search of Grandpa B’s “person,” and most existing authorities hold that neither party has a property interest in the genetic sequence itself. Indeed, Greenberg v. Miami Children’s Hospital arose in Florida—one of only five states with statutes that recognize some form of property interest in genetic test results—and the court nonetheless held that the statute did not create a genetic property interest sufficient for establishing a conversion action.²⁰² That said, as Jessica Roberts notes, at least a couple of courts have very recently shown receptiveness to conversion claims based on privacy violations involving a plaintiff’s own genetic material.²⁰³

At the end of the day, there are colorable arguments on both sides of the question. As Ram argues, there is a mismatch in consent between the DTC user and the familial relation ultimately identified. Yet that mismatch complements another one: the mismatch in the sensitivity of the information consensually revealed about the initial user (including their immediately accessible name and contact information) and the information about the ultimately identified, distant relation (the match to a specific crime scene sample, established only after extensive analysis). Given the unique accuracy value of what such a search would reveal—near-certain presence at a crime scene—one could argue for a wholesale DNA carveout to the Fourth Amendment. In such cases, one might say that the crime-solving interests of law enforcement are particularly high relative to the interests of the suspect (privacy in the specific fact of having been at a crime scene). Entertaining such an argument, however, would require engaging in higher-level debates about the nature of the Fourth Amendment generally, which is beyond the scope of this paper.

The next Part will focus instead on one specific countervailing interest: the liberty interest of the falsely accused who could be exonerated through the availability of familial DNA searches to law enforcement. For constitutional purposes, the Due Process rights of this party must be weighed against the Fourth Amendment rights of a third party, and they must be weighed particularly heavily against any sub-constitutional privacy justifications for statutory restrictions on familial DNA searching.

III. The Due Process Rights of the Innocent Suspect

This Part will argue that a falsely accused suspect has a colorable due process right in the availability of familial DNA testing. If a legislature or court were to impose a ban or probable cause requirement that could effectively preclude law enforcement from using familial DNA to identify unknown third-party suspects, it would impinge on already narrow, yet still cognizable, due process rights designed to protect innocents. The same would be true of any statute preventing private parties, including defendants, from conducting such searches. These rights derive from three sources, which this section will consider in turn: the Compulsory Process Clause, the Due Process right to exculpatory evidence, and the Due Process right to post-conviction relief procedures.

A. The Compulsory Process Clause

The Sixth Amendment guarantees a criminal defendant, among other trial rights, the right “to have compulsory process for obtaining witnesses in his favor.”²⁰⁴ Unlike other Sixth Amendment rights, the Compulsory Process Clause is undertheorized and often overlooked by litigators.²⁰⁵ The case law on the Clause has been contradictory and confusing, alternating between the rule that it stands for the “accused’s ‘right to present a defense’ ” and the idea that it should be subordinate to statutory rules of evidence, such as the rules governing the admissibility of expert testimony, which apply to all litigants alike.²⁰⁶ It is also limited by the limits of the judicial power itself, which does not extend to the “executive” functions of the police and the prosecution.²⁰⁷

The constitutional idea of compulsory process has its origins in the burgeoning trial rights Parliament enacted in the late seventeenth and early eighteenth centuries, specifically for defendants accused of treason.²⁰⁸ Trial rights in the American colonies initially paralleled and eventually surpassed their British counterparts in scope.²⁰⁹ Pennsylvania recognized the most expansive rights; its Charter of Privileges, authored by William Penn, provided that “all criminals shall have the same Privileges of Witnesses and Council as their Prosecutors.”²¹⁰ At the time of the Founding, defendants’ trial rights within an adversary system were seen as a means of offsetting the power of the government.²¹¹ Importantly, trial rights also protect truth-generation, and historical commentators emphasize their function of ensuring that innocent defendants are not convicted.²¹²

The Subpoena Power

At the time the Bill of Rights was drafted, nine state constitutions included some sort of right for criminal defendants to call witnesses but only two, Massachusetts and New Hampshire, included the subpoena power.²¹³ Nonetheless, Congress adopted Madison’s draft language for the Sixth Amendment, including the subpoena power, with no discussion.²¹⁴ While the Supreme Court would not formally consider the scope of the Compulsory Process Clause until the twentieth century, Chief Justice John Marshall, sitting as trial judge in the 1807 treason trial of Aaron Burr, would have the occasion to demonstrate his own understanding.²¹⁵ Marshall approved a pre-indictment request for a subpoena of documents in the possession of President Thomas Jefferson without requiring the defense to pre-determine which ones might be material at trial.²¹⁶

The access to subpoenas of third-party witnesses and documents formally granted to most criminal defendants by state and federal rules of criminal procedure is notably narrower than what Justice Marshall understood compulsory process to require.²¹⁷ To avail themselves of the court’s subpoena power under Federal Rule of Criminal Procedure 17(c), a defendant must show that the evidence sought has “specificity,” “relevancy,” and “admissibility.”²¹⁸ A minority of circuits have followed Justice Marshall and held that the admissibility test requires only that evidence be potentially admissible at trial under the rules of evidence.²¹⁹ A majority, however, have applied a strict admissibility test, which requires a defendant to establish that the items sought are actually admissible before a subpoena will be enforced.²²⁰ Many states have similarly strict requirements.²²¹

While proponents of the strict admissibility requirement often describe it as necessary to prevent a defendant from going on a “fishing expedition,”²²² critics point out that it foils the purpose of the Compulsory Process Clause.²²³ The defendant may never have seen the evidence sought and therefore be unable to explain how it is admissible, and in many cases, inadmissible evidence may be necessary to obtaining other admissible, exculpatory evidence.²²⁴ The rule in force in a given jurisdiction would therefore dictate whether, in a case where unidentified third-party DNA is recovered at a crime scene, the defendant could avail themself of the subpoena power to seek familial matches in a consumer database. In most cases, it would be impossible to tell in advance whether any matches would exist at all. If they did, they would be more likely to lead to relevant evidence after further testing, rather than be admissible in and of themselves. Regardless, even in a jurisdiction with a potential admissibility rule, statutes limiting familial DNA searches could fully thwart defendants’ right to compulsory process for securing evidence. For example, Maryland’s rule prohibits any “person” from conducting a familial search of the state DNA database, which would apply to defendants as well as state actors.²²⁵

The Right to Present a Defense and its State-Rule-Based Limitations

Another contested question regarding the Compulsory Process Clause is whether, beyond the right to subpoena witnesses, it gives defendants affirmative rights to put witnesses and evidence on the stand. If so, this would be a colorable basis for arguing that a defendant has a constitutional right to present familial DNA evidence—superior to any state-law prohibitions regulating DNA and at least equal to the Fourth Amendment rights of the identified parties. In 1967, the Supreme Court considered this question for the first time in Washington v. Texas, when it considered the constitutionality of two Texas statutes preventing a criminal defendant from offering the testimony of a person charged or convicted as a co-conspirator for the charged crime.²²⁶ The Court first held that the compulsory process right is so fundamental to a fair trial that it is incorporated against the states by the Due Process Clause of the Fourteenth Amendment.²²⁷ It then concluded that it includes the right to “present a defense” to the jury:

The right to offer the testimony of witnesses, and to compel their attendance, if necessary, is in plain terms the right to present a defense, the right to present the defendant’s version of the facts as well as the prosecution’s to the jury so it may decide where the truth lies. Just as an accused has the right to confront the prosecution’s witnesses for the purpose of challenging their testimony, he has the right to present his own witnesses to establish a defense. This right is a fundamental element of due process of law.²²⁸

The Washington Court’s test for whether a state evidentiary rule violates this right is whether it is “arbitrary,” holding, in this case, that Texas “arbitrarily denied [the defendant] the right to put on the stand a witness who was physically and mentally capable of testifying to events that he had personally observed, and whose testimony would have been relevant and material to the defense.”²²⁹ In so doing, the Court noted that the Texas laws had deep common law roots in a rule that once totally barred defense witnesses, which was abolished by the Sixth Amendment but lived on in the form of various restrictions on the testimony of co-defendants.²³⁰

Washington would be the first of three cases in which the Supreme Court held that the Compulsory Process Clause gave the defendant a broad right to present evidence that is “relevant and material” to telling their side of the story, over state rules that are “arbitrary or disproportionate” to their purpose.²³¹ These cases explicitly emphasize the Compulsory Process Clause’s purpose of assisting the jury in the search for truth.²³² As Janet Hoeffel notes, Washington and its progeny appear to stand for the proposition that the accused’s right to present a defense is a significant one.²³³ When there is a clash between the defendant’s right to compulsory process and some state evidentiary rule—even one, as in Washington, based in common law—and the opponent of the evidence can test its reliability through the adversarial process, these cases all hold that “the Constitution prefer[s] that the jury hear the evidence.”²³⁴

In other cases, however, the Supreme Court appears to subordinate the defendant’s right to put on a case to state laws—to “shrink,” as Hoeffel puts it, “the right to little more than a right to put on evidence, as long as it comports with the rules of evidence.”²³⁵ For example, in Chambers v. Mississippi, the Court decided for the defendant, holding that the combination of Mississippi’s hearsay rule excluding statements against penal interest and its so-called “ ‘voucher’ rule,” preventing a party from impeaching its own witness, violated Due Process.²³⁶ In doing so, however, the Court did not ask the general question it asked in Washington and its progeny—whether the Mississippi rule was “arbitrary” or “disproportionate” to its stated goal of preventing perjured or unreliable testimony. Instead, it turned to the facts of the specific case and determined that the out-of-court statements at issue had “considerable assurance of their reliability.”²³⁷

The Supreme Court further refined its compulsory process rule in cases unrelated to the reliability of evidence. United States v. Valenzuela-Bernal involved the pre-trial deportation of a possible defense witnesses—a context more analogous to statutory preclusion of defendants’ access to familial DNA searches. The Court held that to establish a violation, a defendant “must at least make some plausible showing of how [the excluded] testimony would have been both material and favorable to his defense.”²³⁸ Then, in Taylor v. Illinois, the Court—while reemphasizing that the Sixth Amendment provides a defendant not only the right to subpoena witnesses but also the right to put them on the stand—rejected the defendant’s argument that such a right to exculpatory evidence is absolute.²³⁹ It held that the accused “does not have an unfettered right to offer testimony that is incompetent, privileged, or otherwise inadmissible under standard rules of evidence” and noted that “the trial process would be a shambles if either party had an absolute right to control the time and content of his witnesses’ testimony.”²⁴⁰ On that basis, the Court held that the trial court did not violate the Compulsory Process Clause by excluding a defense witness when it found that the defendant had willfully failed to disclose the witness in an attempt to obtain a tactical advantage that would minimize the effectiveness of cross examination.²⁴¹

In practice, the circuits have routinely engaged in fact-specific analysis to uphold trial courts’ exclusion of defense evidence as not arbitrary or disproportionate, or on the grounds that the excluded evidence was not material or favorable.²⁴² Even under a narrow view of the rule, however, it seems a criminal defendant who can show the existence of material, favorable DNA evidence has a Compulsory Process right for the court to subpoena that evidence. In practice, this would require that the defendant receive a crime scene sample from the prosecution as Brady material and obtain a subpoena from the court to allow the defense to upload it to a consumer DNA database, regardless of the database’s privacy policies.

In a case involving an unknown, third-party genetic sample found at a crime scene, the defendant has an argument that any potential matches to that profile are material in the sense that they would make the identity of a particular perpetrator other than the defendant more or less likely. The challenge, of course, would be in showing that potential DNA matches in a consumer database are likely to be favorable to a defendant. In most cases, this could not be known reliably ahead of time because the third-party sample could come from someone easily explained away as innocently present at the crime scene. Furthermore, it could not even be known ahead of time whether a consumer database would provide a familial match at all. Finally, even if a database did contain a familial match, the defense would need an expensive genetic expert to build out a profile of the actual perpetrator from the data of whichever family member had made their identity public. However, the fact that a right is difficult to realize in practice does not in and of itself allow the government to violate it.

Constitutional Limitations

Another obstacle for a defendant seeking to make a compulsory process argument for the right to subpoena familial DNA evidence comes from the constitutional separation of powers. The Compulsory Process Clause governs a defendant’s rights at trial, not during the course of a police investigation. Courts have largely recognized the right to present a defense as implicating judicial, rather than executive, branch powers.²⁴³ This distinction has extinguished compulsory process clause arguments in situations in which the law limits defendants’ ability to conduct investigations for themselves.

For example, as Rebecca Wexler has noted, in cases involving extraterritorial investigations, courts have held against compulsory process challenges the asymmetries between law enforcement and defense investigators created by Mutual Legal Assistance Treaties (“MLATs”).²⁴⁴ An MLAT is a bilateral treaty that commits the signatory nations to reciprocal cooperation in criminal investigations, including waiver, where relevant, of privacy laws that would otherwise prevent the transfer of evidence across borders.²⁴⁵ MLATs do not, however, provide for defense subpoenas of material protected by such privacy laws, which puts it out of the reach of compulsory process. As one federal court put it, “the right to compulsory process . . . cannot be stretched to include compelling the invocation of treaty process powers available only to the Executive Branch.”²⁴⁶

Courts have applied similar logic in cases involving conflict between the asserted Sixth Amendment compulsory process rights of a defendant and a potential defense witness who asserts their Fifth Amendment privilege against self-incrimination. Trial courts are deemed to have a duty to protect the witness’s Fifth Amendment right, and if a trial court sustains a witness’s assertion of the privilege, the defendant lacks the right to force the witness to take the stand only to assert the privilege.²⁴⁷ If the witness has already testified for the prosecution and asserts the privilege on cross-examination, however, the trial court may strike the direct examination testimony if it finds a “substantial danger of prejudice by depriving [the defendant] of the ability to test the truth of the witness’s direct testimony.”²⁴⁸ Otherwise, the case law establishes that in such situations of competing constitutional rights courts have no authority to either compel the government to grant immunity to a defense witness or to grant such immunity themselves.²⁴⁹ This rule appears based on the idea that immunity is “pre-eminently a function of the Executive Branch.”²⁵⁰

The MLAT and Fifth Amendment examples suggest that when a statute or competing constitutional right prevents a defendant from fully realizing their compulsory process right, compulsory process is subordinate, and courts even lack the authority to force the government to take “executive” measures to allow the defendant the same end-runs around such rules that the government is allowed to take for itself. The case law is more conflicted when the question presented is whether the Compulsory Process Clause requires a court to compel a witness to give testimony protected by attorney-client privilege. Such cases involve a clash between the constitutional right of the defendant and the statutory right of the witness. The most recent Supreme Court case to touch this question, Swidler & Berlin v. United States, left it unresolved.²⁵¹ In this case, the court held that attorney-client privilege survives a client’s death, but left open the question of whether “exceptional circumstances implicating a criminal defendant’s constitutional rights might warrant breaching the privilege.”²⁵² Since then, a minority of courts have held that that a defendant’s Sixth Amendment rights (confrontation as well as compulsory process) require the court to pierce attorney-client privilege when the witness’s testimony is material.²⁵³ Other courts categorically subordinate the compulsory process right to “traditional testimonial privileges,” in a logic similar to that of the Supreme Court in Chambers.²⁵⁴ Still, other courts will engage in some sort of balancing to determine whether the compulsory process right pierces the privilege.²⁵⁵

The context of the Stored Communications Act (“SCA”) ostensibly provides an example most relevant to the context of familial DNA. Courts have interpreted the SCA to categorically bar defense counsel from subpoenaing U.S. service providers for users’ stored electronic communications regardless of how exculpatory the potential evidence might be, while expressly permitting law enforcement to subpoena such information.²⁵⁶ The government’s subpoena power under the SCA is now, as discussed above, limited by Carpenter, thus lessening the adversarial asymmetry between prosecution and defense. Even after Carpenter, however, some courts have suggested in dicta that a categorical bar on criminal defense subpoenas for electronic communications under the statute might impinge on a defendant’s right to compulsory process under certain circumstances, though none have reached the issue.²⁵⁷ A similar argument would seem to apply to a defendant being statutorily barred from subpoenaing relevant familial DNA evidence from a consumer site.²⁵⁸

Conclusions on Familial DNA and Compulsory Process

In a case like Christopher Tapp’s, an innocent defendant would benefit from the ability to run an unidentified third-party DNA sample through a database to seek matches—direct or familial—that suggest other suspects.²⁵⁹ To the extent that the companies’ own privacy policies barred individuals from uploading DNA other than their own, this would likely require a court-issued subpoena to effectuate. In jurisdictions following the “strict admissibility rule,” the defendant would likely be unable to meet the admissibility requirement because it would be impossible to prove ahead of time that such a database would in fact contain matches relevant to the case. In jurisdictions following the “potential admissibility rule,” they would likely fare better. Either way, a rule like Maryland’s—which prohibits even defendants from conducting familial DNA searches, at least in police databases—wholly extinguishes the right to compulsory process for this type of evidence.

Privacy law scholars have called for an equivalent to the SCA to protect consumers’ genetic information,²⁶⁰ yet courts have noted in the SCA context that the total unavailability of a category of evidence due to lack of subpoena power raises Compulsory Process issues. Critics note that the current admissibility requirement in the subpoena rules of most jurisdictions raises a similar problem, broadly speaking.²⁶¹ Should states pass consumer DNA statutes that wholly bar access to such evidence by defendants, these issues would be compounded, and the Sixth Amendment impairment would be even greater. A defendant has a colorable argument that such a statute would be “arbitrary or disproportionate,” under Washington, to its privacy goal.

B. The Due Process Right to Exculpatory Evidence

The Sixth Amendment right to obtain judicial assistance in securing evidence from third parties is, as we have seen, severely limited. However, the Supreme Court has also recognized a Fourteenth Amendment due process right to the government’s exculpatory investigative materials that is distinct from—and broader than—the compulsory process right.²⁶² Brady v. Maryland held that due process requires that a prosecutor turn over to the defendant all material, exculpatory evidence in their possession.²⁶³ It further held that failure to do so violates due process even when the prosecutor is not acting in bad faith.²⁶⁴ Subsequent cases confirmed that evidence tending to impeach prosecution witnesses constitutes material, exculpatory evidence for Brady purposes.²⁶⁵ Regardless of whether the defense requests the evidence, favorable evidence is material “only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.”²⁶⁶ Restated, the test is whether “the favorable evidence could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict.”²⁶⁷ However, evidence is not considered “ ‘suppressed’ by the government when the defendant has access to the evidence before trial by the exercise of reasonable diligence.”²⁶⁸

1. Brady and DNA

The Supreme Court has emphasized that the prosecutor has a duty to learn of any favorable evidence known to “the others acting on the government’s behalf,” such as the police.²⁶⁹ It is currently uncertain to what extent a defendant has a Brady right to evidence of a third-party’s DNA from a CODIS search conducted on DNA found at a crime scene.²⁷⁰ The sparse case law dealing with such claims has rejected specific arguments on their own facts, finding that the prosecution’s failure to disclose CODIS hits under the circumstances could not be “taken to put the whole case in such a different light as to undermine confidence in the verdict.”²⁷¹ Nonetheless, the reasoning in those cases indicates that if the defendant can show materiality, CODIS hits are subject to disclosure under Brady.

Brady and its progeny apply to exculpatory evidence that remains in the government’s possession, but a separate line of cases governs situations in which the government no longer possesses the evidence in question.²⁷² In California v. Trombetta, the Court held that a defendant complaining of a state’s failure to preserve evidence must show, first, that the evidence “possess[es] an exculpatory value that was apparent before the evidence was destroyed” and, second, “[is] of such a nature that the defendant would be unable to obtain comparable evidence by other reasonably available means.”²⁷³ Subsequently, the Court added a third element to this test in Arizona v. Youngblood, holding that “unless a criminal defendant can show bad faith on the part of the police, failure to preserve potentially useful evidence does not constitute a denial of due process.”²⁷⁴ Numerous states have since adopted statutes requiring law enforcement to preserve DNA evidence.²⁷⁵ The government’s failure to preserve evidence when there is a duty to do so gives rise to a due process violation under some state constitutions if the evidence is material.²⁷⁶

Youngblood also said in dicta that there is no due process right to the government using any “particular investigatory tool,” including quantitative testing, to secure exculpatory evidence.²⁷⁷ This makes it unclear whether a defendant has a Brady right to require police to upload unidentified third-party samples into CODIS to search for exculpatory hits. The argument is that, due to the cooperation between local prosecuting authorities and the federal government, which runs CODIS, any data in the database could qualify as evidence “known to” others acting on “the government’s behalf.”²⁷⁸ Some courts have adopted this reasoning despite Youngblood.²⁷⁹ Others have rejected such claims but only because of a defendant’s failure to show that the testing would have been favorable to the defendant.²⁸⁰ Yet others have construed Youngblood to mean that Brady does not include the right to CODIS searches.²⁸¹ Regardless, some states have allowed these defense-initiated searches through statute,²⁸² and individual laboratories have varying policies, some of which provide for running searches for the defense.²⁸³

In District Attorney’s Office v. Osborne, the Supreme Court made clear that there is no substantive due process right for criminal defendants to access DNA testing.²⁸⁴ In Osborne, the defendant was convicted of kidnapping and sexual assault even though restriction-fragment-length-polymorphism DNA testing had not been performed on semen found in a condom at the scene of the crime.²⁸⁵ In a 1983 action, Osborne argued that he had a substantive due process right to the state’s evidence for the purposes of applying new DNA testing technology that might prove him innocent.²⁸⁶ The Court rejected that claim on the grounds that “[e]stablishing a freestanding right to access DNA evidence for testing would force us to act as policymakers, and our substantive due process rulemaking authority would not only have to cover the right of access but a myriad of other issues.”²⁸⁷

A defendant objecting to the suppression of DNA evidence during the discovery process, thus, only has a few clear constitutional arguments. A defendant might be able to argue that a CODIS hit on a third party from samples taken at the crime scene was material and exculpatory under Brady if they can pass the rather high bar of showing a reasonable probability that the outcome would have been different had it been disclosed. Christopher Tapp, whose own DNA was not found at the crime scene despite the presence of semen from other unknown sources, might have had such an argument if the other sources had been identified but undisclosed. A more challenging case arises when the government conceals the existence of testable samples of biological material that it had not itself tested. Osborne dealt only with the post-conviction context, and thus did not foreclose a procedural due process argument, under Brady, that a defendant has a right to know about untested samples. It would simply be extraordinarily difficult for a defendant to demonstrate the materiality of as-of-yet untested biological samples. Furthermore, Youngblood implies that the argument that Brady compels the state to affirmatively test unidentified third-party samples itself may be quite weak, despite some case law to the contrary.

2. Conclusions on Brady and Familial DNA

Brady applies only to evidence known to the government, which would seem to render it irrelevant to un-searched consumer DNA databases. Even the strongest view of Brady emerging from the case law only requires the government to run searches within CODIS, which includes government-controlled databases. As a practical matter, however, the very existence of Brady doctrine renders consumer databases a powerful and promising new source of exculpatory evidence for criminal defendants. Brady requires that law enforcement turn over evidence of potentially culpable third parties generated by the familial DNA searches they conduct. Thus, preventing law enforcement from accessing consumer databases without probable cause effectively deprives innocent defendants of critical exculpatory evidence.

The reality is that police and prosecutors have independent motivations to bring the correct guilty party to justice. Some prosecutors even pursue this duty post-conviction, as evidenced by the rise of Conviction Integrity Units and examples of cooperation between the government and the defense that have led to DNA exonerations as technology has improved.²⁸⁸ Given the obstacles discussed in the prior Section that prevent defendants from subpoenaing familial matches from consumer databases, law enforcement access may be the only meaningful way for the defense ever to obtain it. This is particularly true given the exceedingly high costs of using a familial DNA hit to build out a family tree and identify the actual party who left the sample at the crime scene. Even if a defendant obtained an unnamed third-party sample from the government through a conventional Brady disclosure and also obtained a court order to input the sample into a consumer database, most defense teams would lack the resources to hire the forensic experts necessary to do anything useful with the results.

This reality also sits in significant tension with the holding of another watershed Due Process case, Ake v. Oklahoma.²⁸⁹ In Ake, the Supreme Court held that the Fourteenth Amendment requires that indigent defendants raising the insanity defense be provided with state-funded psychiatric experts to examine them and testify on their behalf.²⁹⁰ The Ake Court balanced several factors: the private interest affected by the state action, the governmental interest affected if the safeguard were provided and the probable value of the procedural safeguards sought, and the risk of an erroneous deprivation of the affected interest if they were not provided.²⁹¹ On the third point, the Court found that in the absence of a qualified psychiatric expert, “the risk of an inaccurate resolution of sanity issues is extremely high.”²⁹² Despite similar accuracy concerns raised by DNA evidence, there is no parallel right to a state-funded forensic expert.²⁹³ This means that even when the government hands over testable biological material to the defense, an indigent defendant’s counsel may still lack the means to have it tested.²⁹⁴

All considered, as a formal matter, a law or constitutional holding severely restricting law enforcement access to familial DNA searching would most likely not violate Brady as courts have construed it. Yet, by effectively blocking the defense’s access to potentially valuable exculpatory evidence and to the state’s resources for testing it, such laws would indirectly burden the due process values implicated by both Brady and Ake. In trying to protect the privacy interests of guilty parties, such a rule would inadvertently impact the liberty interests of innocent ones.

C. Due Process Rights to Post-Conviction Proceedings Under State DNA and Innocence Statutes

The Lack of Constitutional Innocence Claim

The third potential source of due process rights related to familial DNA involves post-conviction appeals on the grounds of innocence. This landscape is governed substantively by the Supreme Court’s 1993 decision in Herrera v. Collins, which narrowly held that a convicted petitioner’s claim of actual innocence does not state a ground for federal habeas relief, absent some independent constitutional violation during the state proceeding.²⁹⁵ In dicta, the Court left open the possibility that “in a capital case a truly persuasive demonstration of ‘actual innocence’ made after trial would render the execution of a defendant unconstitutional, and warrant federal habeas relief if there were no state avenue open to process such a claim.”²⁹⁶ The Court emphasized that “the threshold showing for such an assumed right would necessarily be extraordinarily high.”²⁹⁷ A few years later in Schlup v. Delo, the Court reaffirmed that no constitutional claim of innocence exists, but it also held that a showing of actual innocence may allow a habeas petitioner to avoid the procedural bar on other, defaulted constitutional claims.²⁹⁸

The Court finally addressed the issue of DNA in post-conviction appeals in the 2006 case House v. Bell.²⁹⁹ Paul House was convicted and sentenced to death in Tennessee for the 1985 murder of his neighbor, with the jury finding as an aggravating factor during the death phase that the murder was committed in the course of a rape or a kidnapping.³⁰⁰ At trial, the prosecution introduced primitive forensic evidence showing that the semen on the victim’s nightgown was consistent with House’s and that the blood on House’s pants was consistent with the victim’s, but not his own.³⁰¹ The forensic expert testified that the blood sample on the pants excluded 93% of the white population.³⁰² One of the government’s expert witnesses acknowledged that the state did not provide saliva samples from the victim’s husband, which would have helped to determine his secretor status and, thus, whether he too could have been a potential source of the collected semen.³⁰³

After his conviction, House filed a state habeas petition based on ineffective assistance and faulty jury instructions.³⁰⁴ When the trial court dismissed the petition, House appealed, but only on the jury-instructions ground.³⁰⁵ After House’s conviction was affirmed, he filed a second state habeas claim for ineffective assistance and sought investigative and expert assistance.³⁰⁶ The Tennessee Supreme Court held that these claims were barred by a statute providing that claims not raised in prior post-conviction proceedings are procedurally defaulted.³⁰⁷ On federal habeas, the Eastern District of Tennessee held an evidentiary hearing to determine whether House fell within the “actual innocence” exception to the procedural default rule recognized in Schlup. Holding that he had failed to demonstrate actual innocence, the Court denied relief, and a divided Sixth Circuit ultimately affirmed.

The Supreme Court granted certiorari to determine, as required by Schlup, whether, in light of new evidence, “it is more likely than not that no reasonable juror would have found petitioner guilty beyond a reasonable doubt” in cases where a prisoner asserts innocence to avoid a procedural default.³⁰⁸ The Court reviewed the new evidence, which reflected the dramatic evolution in DNA technology between 1985 and 2006. First, it noted that DNA testing established that the semen on the victim’s nightgown and panties came from her husband and not House.³⁰⁹ It noted that this would remove sexual assault as a potential motive for the crime and cast doubt on the jury’s finding of rape as an aggravating factor for sentencing.³¹⁰ Second, it considered the testimony of an Assistant Chief Medical Examiner for the State of Tennessee that indicated that the blood found on House’s pants was chemically too degraded and too similar to blood collected during the autopsy to have come from the victim’s body on the night of the crime.³¹¹ He concluded that the stains were more likely to have been spilled from vials containing autopsy samples.³¹² On the basis of all of this evidence, the Court concluded that “although the issue is close [due to other, non-forensic evidence] . . . this is the rare case where—had the jury heard all the conflicting testimony—it is more likely than not that no reasonable juror viewing the record as a whole would lack reasonable doubt.”³¹³ The Court then remanded for consideration of the merits of House’s ineffective assistance claims.³¹⁴ While the prosecutor initially vowed to retry House if he successfully overturned his conviction, the prosecutor later dropped the charges after subsequent developments in DNA technology revealed an unknown perpetrator.³¹⁵ House had spent over twenty-two years on death row.³¹⁶

While House presents a rare case in which evidence of actual innocence had constitutional relevance in a federal habeas appeal, it is important to note how narrow a gateway the case represents. House’s substantive appeal was premised not on innocence but on ineffective assistance under the Sixth Amendment—a distinct constitutional error as required by Herrera. As in many similar cases, House was attempting to argue that his counsel had been ineffective in failing to discover and present exculpatory evidence at trial.³¹⁷ Under Strickland v. Washington, a petitioner can make out an ineffective assistance claim if they can show a reasonable probability that counsel’s poor performance prejudiced the outcome.³¹⁸ However, the Strickland test is explicitly deferential to the judgment of counsel, and courts will evaluate the reasonableness of attorney decisions ex ante, from the perspective of counsel at the time the decisions were made.³¹⁹ It is often the case that—particularly given evolving technology—the absence of exculpatory DNA evidence at trial cannot be attributed to counsel’s ineffective assistance. In such cases, given Herrera, a petitioner has no constitutional grounds for an appeal.

State Post-Conviction Statutes

As discussed in Part I, in response to political demand and the lack of existing remedies, all U.S. states and the federal government have now enacted statutes providing some sort of right to post-conviction DNA testing and for vacatur of sentence on demonstration of innocence.³²⁰ Rules of finality have likewise been displaced by statute in order to permit motions based on new evidence of innocence.³²¹ As Brandon Garrett shows in a taxonomic study of such statutes, however, most statutes create various procedural hurdles that a petitioner must surmount before obtaining testing.³²²

Some states focus on the outcome and require a showing that the DNA evidence is at least probative of innocence or, more strenuously, “material,” meaning that “a reasonable probability exists that the petitioner would not have been convicted if exculpatory results had been obtained through DNA testing.”³²³ Garrett notes that such tests would not be insurmountable if interpreted “to simply require that the DNA testing could be probative of innocence” but points out that some courts will “construct flimsy hypothetical scenarios and then hold that, if there exists even a possibility that DNA testing might not exculpate, it should not be granted.”³²⁴

Some states limit post-conviction innocence relief to DNA testing alone, precluding other sorts of scientific evidence of innocence, and others limit the relief only to technology unavailable at trial.³²⁵ A majority of states limit DNA testing to cases involving serious or violent crimes.³²⁶ Many states require that the petitioner be in custody to seek relief.³²⁷ Some exclude those defendants who pled guilty, and some specify that identity must have been at issue in the underlying trial (which, as Garrett notes, precludes relief in cases of guilty pleas).³²⁸ Four states hold that attorney error or failure to exercise due diligence at trial to preclude post-conviction DNA testing (though such defendants might be eligible for a new trial on a Strickland claim).³²⁹

The accessibility of relief under these state statutes raises distinct constitutional due process issues. In District Attorney’s Office v. Osborne, the Supreme Court recognized that a convicted individual has “a liberty interest in demonstrating [their] innocence with new evidence under state law.”³³⁰ The Court affirmed that a state-created right “can, in some circumstances, beget yet other rights to procedures essential to the realization of the parent right.”³³¹ The Sixth Circuit had held below that Alaska had violated Osborne’s due process rights under Brady by failing to provide the defendant access to the DNA evidence used at trial, because the Alaska statute provides a substantive right to be released on a sufficiently compelling showing of new evidence that establishes innocence.³³² The Supreme Court, however, held that Brady did not apply post-conviction and that a state “has more flexibility in deciding what procedures are needed in the context of post-conviction relief.”³³³ The appropriate test, the Court said, is whether Alaska’s post-conviction relief procedures “are fundamentally inadequate to vindicate the substantive rights provided” (in this case, the liberty interest in being released under Alaska’s innocence statute).³³⁴

The Osborne Court examined Alaska’s post-conviction procedures, which provide for discovery in post-conviction proceedings and specified that such discovery is available for those seeking access to DNA evidence.³³⁵ The Alaska post-conviction statute limits discovery to “newly available” evidence and imposes due diligence and materiality limitations.³³⁶ Furthermore, the Court found that Alaska courts had suggested in dicta that in DNA cases, the state constitution may even provide an additional right of access to those who cannot meet the usual post-conviction discovery requirements, though the issue remains unsettled. The Court held that these procedures, which Osborne had not yet exhausted, were “adequate on their face” to the realization of his liberty interest under the post-conviction DNA statute.³³⁷

Federal courts of appeals have been relying on Osborne to find state post-convictions procedures fundamentally adequate and, thus, reject Section 1983 claims based on refusal of post-conviction DNA testing.³³⁸ However, the Second Circuit came to the opposite conclusion about the New York City Police Department’s evidence management system in Newton v. City of New York.³³⁹ In 1985, Alan Newton was convicted of rape, robbery, and assault based on eyewitness testimony.³⁴⁰ The rape kit was not tested at trial, but in 1988, Newton moved for an order authorizing his expert to conduct forensic tests on the rape kit pursuant to New York’s post-conviction relief statute. The statute authorizes vacatur based on the discovery of new evidence that “could not have been produced by the defendant at the trial even with due diligence on his part and which is of such character as to create a probability that had such evidence been received at the trial the verdict would have been more favorable to the defendant.”³⁴¹ The New York Supreme Court granted the motion, but the officer of the Chief Medical Examiner Office reported that the rape kit contained no testable sperm.³⁴²

In 1994, New York passed a post-conviction DNA statute that allows for testing upon the court’s determination that “if a DNA test had been conducted on such evidence, and if the results had been admitted in the trial . . . there exists a reasonable probability that the verdict would have been more favorable to the defendant.”³⁴³ When Newton made a motion for testing under this statute on the grounds that technological advances since 1988 had rendered smaller samples testable, the court denied his motion after the District Attorney’s Office revealed that it had lost the rape kit entirely after the 1988 analysis.³⁴⁴ Ten years later, upon request from defense counsel, an Assistant District Attorney who had not been involved in Newton’s case searched for the rape kit again and found it in a barrel in an NYPD Property Clerk’s warehouse in Queens.³⁴⁵ After Newton had served more than twenty years in prison, new DNA testing of the rape kit exonerated him.³⁴⁶

In reviewing the district court’s decision to set aside the jury verdict for Newton in a subsequent Section 1983 case, the Second Circuit applied Osborne to the facts of the case. It held that, first, like Osborne, Newton had a liberty interest in demonstrating his innocence on new evidence.³⁴⁷ It then turned to the question of whether New York’s procedural rules were fundamentally adequate to the realization of this interest. Newton argued that, like Alaska’s in Osborne, New York State’s similar procedures were fundamentally adequate, and, unlike Osborne, Newton had diligently availed himself of all of them. However, Newton argued that New York City’s ineffective evidence management system effectively nullified those procedures.³⁴⁸

The Second Circuit agreed. Noting that Newton’s action focused not on the State law itself but on the execution of the law, the court did “not decide what specific City procedure is necessary to manage and track evidence,” but the court reinstated the jury’s verdict for Newton.³⁴⁹ It found that New York’s post-conviction DNA statute “is consistent with requiring the NYPD’s evidence management system to provide an adequate means to determine if evidence is available for testing and, if so, where the evidence is located.”³⁵⁰

Post-Conviction Statutes and Familial DNA

Newton suggests courts might be open to the argument that state procedures rendering DNA testing of old evidence effectively impossible could violate a defendant’s limited due process right to vindicate substantive rights under post-conviction relief statutes. Mismanaged storage procedures are distinctly different, however, from a formal legislative decision to restrict familial DNA searches through consumer sites. If Osborne stands for anything, it is the proposition that “[f]ederal courts should not presume that state criminal procedures will be inadequate to deal with technological change.”³⁵¹ The Court made it quite clear that “the development of rules and procedures” in the area of DNA science should be left in the “hands of legislatures and state courts shaping policy in a focused manner.”³⁵² A convicted prisoner might try to argue that a state statute effectively barring law enforcement from even attempting to re-open a case through familial DNA searches renders its procedures “fundamentally inadequate” under Osborne to vindicate their right to state-created post-conviction process. However, given the Court’s strong deference to legislatures, this argument is unlikely to prevail. It should be noted from this discussion of post-conviction process, however, that the lack of a constitutional innocence claim and the inapplicability of Brady post-conviction emphasize the importance of accuracy in initial criminal adjudications.

Conclusion: Adjudicating Due Process in Conflict

Part Two of this Article considered the argument that a subject has a Fourth Amendment right not to be subject to warrantless searches of their family member’s DNA within databases, even with the database user’s consent. It agreed with commentators who argue that DNA contains intimate information, and the consent of an actual database user is not logically equivalent to the consent of that user’s relatives who may be identified through a familial search. Nonetheless, Part Two concluded that the argument that such searches require a warrant after Carpenter is fairly weak. In the first place, the specific privacy interest recognized by Carpenter in the totality of one’s movements implicates detailed information about one’s actual life. By contrast, with the proper limitations on use, a DNA match reveals only one significant piece of information: that a subject left a particular sample in a particular place. More importantly, no authority recognizes a reasonable expectation of privacy in someone else’s bodily materials. Familial DNA works because a subject has left their own sample at a crime scene, and the police match it to the profile of a physical sample offered voluntarily by a relative. A suspect does not have a right to bar their spouse from allowing the police to take a sperm sample from their own body after marital intimacy in order to compare it to the sample in a rape kit. In the case of consumer databases, the suspect’s connection to the consenting party is even more attenuated and less voluntary. Thus, the idea that the Fourth Amendment allows them to override the third party’s use of their own body sample might be somewhat more colorable, but it is still, ultimately, weak.

On the other hand, familial DNA searches implicate the Sixth and Fourteenth Amendment rights of other criminal suspects in cases where unidentified third-party samples are found at crime scenes. Part Three of this Article argued that a criminal defendant has a colorable due process right not to have consumer DNA evidence rendered effectively unattainable by law enforcement. First, any statute preventing criminal defendants from exercising their Sixth Amendment right to subpoena that information would raise compulsory process issues. Even if legislatures or courts only barred law enforcement from conducting such searches, the defendant’s access to the subpoena power is heavily limited in many jurisdictions by the materiality and favorability requirements, and any useful hits resulting from such a search would in any case require expensive forensic testing to lead to a specific subject. Thus, the reality is that a defendant is usually only going to access important exculpatory evidence through Brady, which would put familial DNA largely out of reach if the police were precluded from conducting searches at all or only with probable cause. In any given case, this problem is also compounded by limited options post-conviction, at which point the Supreme Court has held that Brady does not apply and no independent right to DNA testing exists. Law enforcement’s ability to access the most accurate evidence on the front end affects a defendant’s right to obtain the most important kind of exculpatory evidence under Brady.

In Osborne, the Supreme Court declined to “constitutionalize” post-conviction DNA testing as a freestanding due process right to protect the liberty interest of a potentially wrongfully convicted party.³⁵³ Innocent defendants would then be especially harmed by an asymmetrical attempt to “constitutionalize” DNA on the front end—during investigations. Should courts recognize, or legislatures adopt, a Fourth Amendment rule that keeps familial DNA away from law enforcement absent probable cause, such defendants would be effectively shut off from the most accurate and valuable evidence at all stages of the criminal process. Like the Fourth Amendment argument, this Sixth Amendment argument is also somewhat weak since most courts do not even recognize a Brady right to law enforcement searches of public databases. Nonetheless, because a categorical ban on warrantless searches would essentially render all familial DNA evidence inaccessible, it would inherently block a defendant’s access to exculpatory evidence.

The admissibility of familial DNA is, thus, not a due process question in which it makes sense to think of balancing one defendant’s rights against purely “government interests” (although the accuracy of DNA evidence certainly makes the government’s crime-solving interest particularly high). Rather, there are two potential defendants’ liberty interests at stake—one of whom is most likely innocent. Constitutional rights necessarily have limits when they impact other parties’ constitutional rights; scholars debate whether these constitute limits on the scope of rights or simply their realization.³⁵⁴ Many nations’ constitutions have general limitation clauses, which set out the specific conditions under which constitutional rights can be limited.³⁵⁵ However, the United States does not have a general limitation clause and defines certain rights without any limitation.³⁵⁶ Courts have treated some rights as having implicit limitations, such as when they engage in interest balancing to determine “unreasonableness” under the Fourth Amendment. For other situations, the Supreme Court has adopted its own, much criticized tiered system of scrutiny.³⁵⁷ Sometimes, though inconsistently, courts limit constitutional rights through recourse to some version of John Stuart Mill’s harm principle, which holds that the only acceptable limitation on liberty is when it is necessary to prevent harm to others.³⁵⁸

Generally, it is true that a constitutional right’s lack of limitation clause does not render the right absolute but instead subject to interpretation in harmony with the Constitution’s other provisions and other legitimate public interests.³⁵⁹ This inherently entails courts prioritizing certain rights over others based on the facts of a particular case.³⁶⁰ In the context of criminal justice, for example, claims of Sixth Amendment deprivations “are subject to the general rule that remedies should be tailored to the injury suffered from the constitutional violation and should not unnecessarily infringe on competing interests.”³⁶¹ Thus, trial courts have discretion to limit a defendant’s Sixth Amendment confrontation right in a particular case through limiting the scope of cross-examination based on potential harms, such as harassment of the witness, prejudice, confusion of the issues, the witness’s safety, or questioning that is repetitive or only marginally relevant.³⁶² In United States v. Stein, the Second Circuit found a violation of the Sixth Amendment right to counsel where a Department of Justice charging policy caused a private employer to terminate attorneys’ fees that they would otherwise have paid for according to the terms of their employees’ contracts.³⁶³ While the employees did not have a Sixth Amendment right to employer-funded legal counsel—in the same way that a suspect may not have a right for the police to conduct a familial DNA search—it nonetheless constituted an interference with the employees’ exercise of their Sixth Amendment right to mount a defense with their own assets that could not be justified by the state interest in encouraging cooperation by the corporate defendant.³⁶⁴

Cases of competing constitutional interests are more delicate. In Nebraska Press Association v. Stuart, the Court rejected an absolute balancing test between the First and Sixth Amendments where, in order to protect the accused’s right to an impartial jury, a trial judge had issued an order restraining members of the press from publishing or broadcasting accounts of the defendant’s admissions.³⁶⁵ The Court noted that “[t]he authors of the Bill of Rights did not undertake to assign priorities as between First Amendment and Sixth Amendment rights, ranking one as superior to the other”³⁶⁶ and proceeded to balance the nature and extent of pretrial news coverage, potential mitigating measures, and the effectiveness of a restraining order at preventing the danger; ultimately, the Court held for the petitioner journalists.³⁶⁷

When courts must adjudicate clashes between the constitutional rights of two criminal defendants, their work is complicated by the fact that the meaning of criminal due process is itself contested. Due process has a specific doctrinal meaning related to the procedures required by the Fifth and Fourteenth Amendments, the latter of which incorporates most rights in the Bill of Rights against the states.³⁶⁸ It also has a broader meaning that represents the “principles of civility or fairness that are supposed to underlie procedural and often substantive legal rules.”³⁶⁹ Because of its prominent role in the Bill of Rights, due process has become a distinctly American concept, with jurists in other nations resorting to other concepts such as “procedural justice,” “the rule of law,” “natural justice,” and “fundamental justice.”³⁷⁰

In The Limits of the Criminal Sanction, Herbert Packer famously divides criminal justice thinking into two camps.³⁷¹ One, which he refers to as the “Due Process Model,” values the observation of fair procedures, even at the expense of letting the guilty go free, while the other, the “Crime Control Model,” values accurate determinations of guilt and innocence.³⁷² Accuracy has, in and of itself, been a value central to constitutionalized criminal due process since the start of the criminal procedure revolution in the mid-twentieth century.³⁷³ Rules about impartiality and reliability, the right to counsel for indigent defendants, the right to cross-examination, and the rule against involuntary confessions all exist to prevent the conviction of innocents due to malicious, lazy, hasty, or overzealous police and prosecutors.³⁷⁴ In In re Winship, the Supreme Court held that the standard of proof beyond a reasonable doubt is a due process requirement in criminal trials on accuracy grounds.³⁷⁵

As David Resnick argues, due process serves an important justificatory purpose, providing reasons for subjecting a person to criminal punishment.³⁷⁶ He notes that “our concern with the subjective probability of correct outcomes reflects a requirement of morality and not simply efficiency or rationality.”³⁷⁷ Thus, while critics tend to pit accuracy values against due process values in cases where fair process requires guilty defendants to go free, the protection of innocents is a critical moral component of due process. The availability of DNA evidence inherently implicates accuracy values—not only the broader societal goal of identifying the guilty but also an individual’s right not to be punished without justification.

That said, Packer’s “Due Process Model” fits the argument that unwarranted familial DNA searches violate the Fourth Amendment rights of the third-party family members identified. Under this view of due process, the absence of “fair play” for the suspect should override the state’s interest in crime solving.³⁷⁸ Some of the Supreme Court decisions that fit this model have been described as premised on “fundamental fairness”—most notably Rochin v. California, which excluded evidence found after law enforcement pumped a suspect’s stomach against his will and without judicial authorization.³⁷⁹ In an opinion by Justice Frankfurter, the Court held that the stomach pumping—while producing accurate evidence—violated due process because it “shock[ed] the conscience.”³⁸⁰ The Fourth and Fifth Amendment exclusionary rules in general represent the primacy of fair play over accuracy since they necessarily better protect the procedural rights of the guilty than they do the innocent. However, an unjustly accused party could also argue that principles of fair play and fundamental fairness should prevent them from being criminally convicted without access to a whole body of highly reliable evidence.

It should be noted that there is a third set of due process values beyond these two that is related to the expressive function of criminal procedure and its relationship to public confidence in the justice system. The public will be more likely to support and participate in criminal justice when it believes that the process operates fairly.³⁸¹ On that basis, some critics point out that “[a] focus on the individual rights of the defendant to the exclusion of the interests of other participants belies the public’s critical role.”³⁸² In this view, a trial is not just a fact-finding mission but serves an “important symbolic function as public expressions of the affected parties’ rights to demand that official acts be explained and justified.”³⁸³ This concept of due process is particularly important in the case of familial DNA. High-profile cases like the Golden State Killer have increased public awareness of this science, and exonerations like Christopher Tapp’s, which was widely covered in true crime media, have created demand for it as a means of recourse for the wrongfully accused and convicted.

Ultimately, the case of familial DNA does not present a simple, bilateral competition between fair Fourth Amendment process for a criminal defendant and public interest in crime-detection; it also implicates the Sixth and Fourteenth Amendment rights of an innocent suspect whose risk of wrongful conviction would be higher in a world with law enforcement access to familial DNA substantially restricted. Should courts consider Fourth Amendment claims challenging familial DNA evidence, they should bear in mind these other defendants when determining whether such searches are reasonable without a warrant. Furthermore, should states seek to regulate law enforcement use of DNA via statute, they should not adopt restrictions as broad as Montana’s. Ideally, even if states do restrict law enforcement access, they should at least create a formal subpoena process to clarify how criminal defendants can access this evidence. However, given the limitations on defense testing resources, this mechanism would likely benefit only a minority of defendants and should be supplemented by some degree of law enforcement access.

98 S. Cal. L. Rev. 761

Download

Erin Sheley*

* Professor, California Western School of Law. Many thanks to Daniel Yeager and Donald Dripps for their valuable comments on drafts of this Article. I am also grateful to the helpful feedback I received at workshops from the faculties of California Western School of Law and the University of San Diego School of Law.

Punishment as Placebo

May 2025May 2025Sheldon A. Evans*

The modern criminal punishment regime has failed to deliver on its promise of public safety. For all of the resources expended and all of the human costs incurred, the ever-growing carceral state does not make us safer. Scholars across the social sciences have studied these shortcomings for decades using various methodologies. The burgeoning prison population has little relation to the crime rate, which advocates have argued supports decarceration efforts to remedy the inefficiency, inequality, and subordinating effects of harsh sentencing policies and institutional design choices that have supported the mass incarceration crisis. What then justifies the continued propagation of this failed regime?

This Article proposes a new balancing of considerations to answer this question that innovates punishment theory through the medical and experimental lens of placebos. The efficacy of policy must be balanced with its public perception; thus, a policy may not be effective at fixing a problem, but this is often less important than the perception of the policy. A medical placebo treatment may be physiologically inert, but it still can have a positive psychological and therapeutic impact by making the patient feel better because they think they are receiving effective treatment. In the same way, the cultural value of mass incarceration extends beyond its failed effectiveness by providing psychological and therapeutic value to help the public cope with their fear of crime, their moral commitments to justice, and their socialized feelings towards offenders.

This placebo methodology offers several unique insights about punishment theory and practice that learn from the extensive medical research and ethics literatures. By viewing punishment as a placebo, the modern punishment regime must confront unique questions concerning the social harm of punishment, the justifications for public deception, the ethics of experimentation, and the perverse economics of inefficacy. These theoretical synergies also provide practical insights into how to reverse and regulate such placebo punishments while focusing on a more humane and ethical punishment future.

Introduction

“Freedom from fear is a basic right of every American. We must restore it.”¹ Richard Nixon’s bold message to America illustrates one of the most underappreciated aspects of modern criminal punishment policies: the fear and perception of crime is just as important as the actual presence of crime. Nixon understood these political winds and social yearnings, and he ushered in a new vision of law-and-order national leadership through his War on Crime.² This “fear of crime” was the overriding concern of Ronald Reagan’s administration that passed some of the most sweeping crime bills and policies in modern history.³ George H.W. Bush picked up this thread when he declared that “freedom from crime and the fear that stalks our cities” was a “fundamental civil right” that he was committed to protecting through crime control legislation.⁴ And not to be outdone by Republicans, Bill Clinton continued this trend when supporting further crime legislation by arguing that if “the American people do not feel safe on their streets, . . . then it is difficult to say that the American people are free.”⁵

These words from past presidents highlight an important duality in criminal law that has shaped an entire generation of policy yet is curiously understudied in the legal academy: the relationship between efficacy and perception. In other words, there are two values that criminal law and punishment seek to deliver as a public policy. First is the value of efficacy, which measures how effective the law is in actually achieving public safety goals by lowering crime rates and making us safer. Second is the value of public perception, which measures how the law assuages public fear and controls how we perceive or feel about our safety and the prevalence of crime. Both of these values have explanatory weight in how we study and think about criminal law and punishment. While there has been robust consensus that harsh, tough-on-crime policies have not delivered on their promises of efficacy in making Americans any safer, these discussions miss perhaps the more important value at play in our system. As we can see from former chief executives,⁶ crime policy is just as (or even more) concerned with assuaging public fear and controlling public perception as it is with the law’s efficacy.

This Article takes the path less travelled by focusing on the value of perception in criminal legal policy. Even if we accept arguendo that criminal law enforcement and policies have objectively failed to lower crime rates, do these policies nevertheless hold some value in making Americans feel better about crime? This Article tackles that question through a theoretical methodology that brings to bear medical, social, and ethical considerations in a framework that portrays modern punishment as a placebo: something that is not efficacious at fixing the underlying problems of crime yet might have psychological and therapeutic value in impacting society’s perceptions of crime.⁷

Placebos have been a part of medical research for generations, but they have also become a respected tool for studying social behavior.⁸ For example, a person’s lucky hat might make them feel more confident; as a result, social scientists can conceptualize and measure a positive placebo effect on this person’s perception and expectations when wearing this lucky hat.⁹

At a societal level, the same can be true of legal rules. Anup Malani has considered how legal regulations in healthcare and even tort law shape people’s perceptions and expectations similar to placebo effects.¹⁰ Amitai Aviram has also studied the law’s role in manipulating perceptions when a law is designed to signal an increased or decreased risk of certain activity.¹¹ Regardless of a law’s efficacy, Aviram recognized the law’s role in impacting social perceptions of safety.¹² In this way, the law may have created a safer environment, but its primary value in increasing social welfare was a psychological placebo.¹³

In the criminal legal context, this Article argues that modern punishment policies—and specifically, mass incarceration as a case study—would benefit from being conceived as a placebo. These policies promise to address the social problem of crime, but they are no longer designed or implemented efficaciously to fix that underlying problem. Instead, the primary value of these policies to the public is their ability to control, assuage, and in some cases manipulate feelings, perceptions, and expectations toward crime.

To build out the analogy a bit further, let’s liken punishment to a form of medical treatment. A patient may go to see their doctor with a particular physical health problem; the doctor may choose to prescribe the patient a placebo treatment that is not meant to efficaciously treat the patient’s physical problem but instead is meant to make the patient think they are receiving an effective treatment. The placebo is only delivering a psychological or therapeutic benefit to the patient. And because of the interesting ways that placebos work, the patient does indeed feel better. Applying this logic to punishment requires a bit of imagination, but nevertheless has conceptual value. The patient is the public, who is dealing with the social ill of crime. Thus, the public/patient seeks solutions from social decisionmakers such as politicians, police officers, prosecutors, and other actors who diagnose the problem and purport to give an effective solution; however, like a placebo, these policies that have increased the surveillance state, police power, and mass incarceration have not delivered on the efficacy front but are instead meant to control public perception and make the public feel better about crime. I understand this is not a perfect analogy and there are nuances to be explored below, but using this methodology of placebos is an important turning point in how we should think about criminal policy and how the efficacy and perception values interact. The objective study of efficacy must be balanced with these more subjective studies of feelings and perceptions of crime to understand the social currency of punishment.¹⁴

Using interdisciplinary tools of medicine, sociology, and ethics, this Article explores more than a mere metaphor by developing a fuller methodology that reframes criminal punishment in ways that hold valuable insights and complexities for the criminal theory literature. First, the placebo methodology studies and illuminates the types of social cues that impact public perception in the criminal space, such as the emotion of fear and the social and cultural meanings of punishment.¹⁵ Second, it refocuses the punishment literature on the oft-forgotten history that the medical field shares with criminal law.¹⁶ The so-called disease of the criminal mind,¹⁷ the dangerousness of the disabled,¹⁸ and the criminalization of illness¹⁹ are just some examples. Just as individual people could be perceived as dangerous due to an illness or impairment, so too have we used the criminal law to excise people from society that threatened our fragile moral and cultural foundations. Third, the socio-medical ethics of placebo treatments frame unique questions specific to that field that provide an opportunity to discuss the larger ethical questions of punishment. For example, the crossover between the ethics of harm, benevolent deception of the public/patient, using placebos and punishments as tools of experimentation, and the economic incentives at play are important considerations of both the theoretical and practical stakes.

Finally, the above theoretical framings are foundational to the practical interventions that the placebo methodology might inspire.²⁰ As with all punishment, underlying theory heavily impacts the types of practical punishments we impose on people who have violated criminal laws.²¹ Throughout the analysis of the placebo methodology, this Article explores unique policy solutions gleaned from medical and scientific literature to consider how to fix the problems that overreliance on placebos can bring. This includes unique interventions rarely trodden in the punishment literature, such as new approaches to government regulation, ethical boundaries, policy experimentation, and cultural messaging. Thus, after a generation of failing to solve the problems of mass incarceration and the ballooning punishment regime, the placebo methodology hopes to elucidate innovative practical interventions.²²

The placebo methodology and all its theoretical and practical draping must ultimately answer the most important question of punishment scholarship: Is punishment justified? If punishment does indeed operate as a placebo and provides meaningful value to public perception, is this enough to justify the status quo? Many people would agree that in general, we police and punish too much in this country. And even if it does not make us any safer, we should not give short shrift to the law’s power to make us feel better. These are questions that address the ultimate payoff of the criminal legal system and the degrees of punishment severity to which we are willing to condemn our fellow man for the utilitarian benefit of our collective feelings.

This Article humbly contributes to these discussions in four parts. Part I builds out the placebo methodology by diving deeper into the shared logic between placebos and punishment as treatments for physical and social ills, respectively. Part II applies the placebo methodology to the mass incarceration crisis as the most salient case study that dominates modern conversations about punishment. Part III further explores the placebo methodology as it applies to mass incarceration under a host of different socio-medical considerations, including the harm principle, justified deceptive practices, the ethics of experimentation, and economic incentives. Further, this Part also confronts the all-important question of whether punishment is justified under the placebo methodology. Part IV closes the Article by broadening the framework to other worthwhile legal disciplines, as well as considering tangential socio-medical framings of punishment that might also be worthy of future exploration. This Article hopes to advocate for the placebo methodology as a malleable toolset that should be utilized in future legal research exploring social modeling of behavior and policy.

Exploring punishment as a placebo policy is more than just a clever alliterative analogy. The placebo methodology gives punishment theory and policy interventions a new set of tools that fully appreciate the cultural and social impacts of criminal punishment. And with such an outlook that values both actual and perceived goals and benefits, we can establish a more careful and nuanced critique of the status quo that calls for a rejection of placebos in favor of panaceas.

I. The Placebo of Punishment Theory

When boiled down to its purest form, the placebo methodology seeks to add a set of social and cultural tools to understanding modern punishment based on how placebos work in the medical field.²³ Such analogies are not new in legal scholarship and often add value by discovering latent assumptions or highlighting underappreciated aspects of legal analysis through a new theoretical framing.²⁴ Following this tradition, this Part pushes deeper by explaining why framing punishment as a placebo is more than a metaphor, but a more fulsome methodology. Section I.A fleshes out the theoretical and literal connections between punishments and placebo treatments by focusing on the actors in each system. Section I.B explores the placebo effect of punishment and whether it succeeds in making us feel safer. These Sections offer important insights into the socialization of punishment and how we should think about the causes of crime, as well as introducing the ethical and practical dilemmas of power, politics, and subjective perceptions to be explored in later Parts.

A. The Placebo Methodology

The theoretical and logical connections between placebo treatments and criminal punishment is a simple idea: Like a medical treatment, punishment has been prescribed to our modern society to solve the underlying problems of crime; but like a placebo, our modern punishment system does not deliver on its promises of efficacy by lowering crime rates but rather is meant to control and mitigate public perception of crime.

Although the exact definition of a placebo treatment is contested in the medical literature,²⁵ placebos, for purposes of this Article, can be defined as treatment that is administered by a person in power (usually a doctor or medical researcher) to provide psychological or otherwise therapeutic benefits to make patients feel better, with little or no active ingredients that would be efficacious to treat the physical medical problem.²⁶

In this methodology, the individual patient being treated is the body politic. This is based on a larger relationship between the study of the physical body and its relation to society as a whole.²⁷ Since ancient times, the human body has served as a structural archetype of society and its institutions.²⁸ The Greek and Latin traditions had a double meaning for the term “diet.” It meant both a mode of ordering one’s life, as well as a political assembly of princes that performed legislative and administrative duties.²⁹ The term “regimen” also had a dual meaning, referring both to a medical means of therapy and also a system of government that we refer to as a “regime.”³⁰ This is part of the larger Western tradition that has found a double and connected linguistic and cultural meaning between caring for the individual body and upholding the larger social body. Political tradition borrows from human and natural traits to name its institutions, such as the “head” of organizations and governments, “branches” and “arms” of the same, and using the “body politic” to identify an entire society as one social organism.³¹ Religious tradition is rife with such symbols, often using the body as a metaphor for spiritual insights: the body of Christ was a symbol for the institutional church; Adam’s rib was a symbol of patriarchy; the flesh versus spirit dichotomy that represented moral weakness and strength respectively; and the list can go on.³² These examples do not do justice to the sheer number of body-as-society analogies that abound to illustrate the body’s enduring importance into how we think about society and its institutions. The human body is a foundational lens of Western thought and philosophy on social order.³³ Society is personified because our political and philosophical tradition recognizes that institutions mirror the very humans they seek to regulate.³⁴

The body’s dysfunction and disease has also been an important characterization of social disorder in Western thought. Physical diseases were frequently interpreted as manifestations of a “deeper malaise in the social structure.”³⁵ Consequently, the concept of disease was used as a symbol of social problems that could break down and destroy the body politic just as actual disease could break down the physical body. Only a few hundred years ago, obesity was considered to be an individual “manifestation of the flabbiness of the social system,” especially as it exposed the sedentary lifestyle of the rich.³⁶ Even cancer has a history of being described as a “disease of civilization” that manifested during certain social failings.³⁷ This is an interesting cause-and-effect relationship because in many ways, social hierarchies have always exposed some classes of society to various diseases. For example, the social hierarchy of the rich allowed those in agrarian societies to live less active lifestyles.³⁸ Thus, social ordering can certainly impact the development of individual diseases, which in turn communicates a broader identity of a society in accordance with its social priorities. Consequently, the history of associating the presence of physical disease as a larger sign of social disorder, malaise, and even moral decay is well documented.³⁹

The individual human body is more than biological muck; it is our tool for interacting with and understanding the natural and social world. The body serves as a mirror image to help us understand social institutions, and in turn, those social institutions help shape the roles of our bodies in social structure. The intersection between our bodies and our institutions govern issues of “health, gender, childhood, ethnicity[,] and disability[,] and many more substantive areas[,]”⁴⁰ including crime and punishment.

If the patient is the public, then the patient’s disease can be understood as crime itself that threatens the health of social order just as physical disease causes strain and breakdown of the physical body. To let such a social disease fester and ferment would be tantamount to allowing a physical plague to spread without proper intervention of cultural sanitation.⁴¹

The disease/crime comparison is a longstanding principle in our social understanding and moral disapproval of deviant and antisocial behavior. Academics, commentators, and public officials have often used this crime/disease analogy to display their disgust and rally public and intellectual opinion. Emile Durkheim and contemporary scholar Martha Grace Duncan have separately written about how society treats crime and criminals as socially unclean things that must be resolved through institutional intervention.⁴² Academics have built on this concept by discussing a “quarantine” model of criminal justice that evokes the need to separate people with the contagious disease of criminality from the rest of society;⁴³ there are frequent comparisons to crime as a cancer of society that

suggest it will metastasize unless properly dealt with.⁴⁴ Even our most liberal politicians have fallen prey to such logic, comparing the spread of violent crime to “a national plague.”⁴⁵

Given these rhetorical similarities, it comes as no surprise that the criminal law has often found an ally in the medical sciences. First, we can consider how disease and crime were understood together throughout our history, which captures how physical diseases have been criminalized to various extents as a method of preserving social order and public health. Criminal laws have often been used to identify medically “dangerous” and “contagious” people who suffer from a host of diseases and ailments and excises them from the community through the same logic of medical quarantine. As medicine was used to diagnose a host of illnesses that illustrated social anxieties toward what dominant groups deemed unacceptable behavior,⁴⁶ crime policy followed close behind to regulate and punish such behavior.⁴⁷ Religious laws that determined cleanliness go back for thousands of years, and were used to regulate and criminalize behavior of people suffering from leprosy and bloodborne diseases.⁴⁸ In the era of the HIV/AIDS epidemic, intentional spread of the communicable disease was at one point criminalized as first-degree murder; only in the 1990s were these doctrines struck down in favor of lesser criminal offenses.⁴⁹ Thus, the medical authority of the era always determines which individuals are dangerous to public health, and criminal law has been a part of the infrastructure used to prevent spread of these diseases. It has also been used to assuage the public that they would be safe from such “dangerous” individuals because the law would keep them in their appropriate place.

Mental illness and disability have a jaded past that continues in criminal law to this day. Homosexuality was treated like a mental illness up until only a few decades ago,⁵⁰ and was criminalized accordingly.⁵¹ Alcoholism, substance abuse, malingering, and even political dissent have been subsumed under various definitions of mental or physical disease to be regulated by the criminal law.⁵² People dealing with various mental health issues have often suffered from stereotypes that they possess characteristics of criminality.⁵³ These people are often mischaracterized by police and bystanders as being more dangerous, which then justifies legal use of force and coerced treatment through criminal law.⁵⁴ Consequently, the prison as an institution has turned into the neoliberal mental health facility,⁵⁵ often responsible for distributing more prescription mental health treatments than any other institution.⁵⁶ This should come as no surprise because over one-third of incarcerated persons have been diagnosed with a mental illness.⁵⁷

The criminality of physical ailments and the treatment of mental illness change according to social constructions of who the dominant social group deems deserving of criminal punishment. As Jamelia Morgan has studied, mental and physical disability is just one area that has been “conceived of as a social contagion or pathology to be contained through policing and carceral control.”⁵⁸ All of these criminal laws are socio-medical reactions that were purportedly meant to keep society safe based on the best science of the time; but they often functioned by subjugating the sick, disabled, and the most vulnerable. Fear was central in these criminal responses; it was fear of the unknown, fear of the unclean, and fear of the transmission of such conditions in the general public.⁵⁹

This interaction between the patient/public and disease/crime raises even more interesting questions. For one, the entire public may not be suffering from the disease of crime, just as a medical malady might not be spread throughout the entire body. Just as diseases are often concentrated in certain areas of the body, crime too is unfortunately concentrated in certain pockets of our body politic. Crime tends to be concentrated in poor areas with high population density.⁶⁰ Perhaps there is a further lesson here in the methodology; neglect can often lead to problems. If an individual neglects to care for their body through nutrition, hygiene, or general maintenance, we might not be surprised that such neglect correlates with medical problems down the road. So too should we think about our body politic neglecting our inner cities, our less financially fortunate citizens, and how such social and institutional neglect leads to problems of crime.

Also, conceptualizing crime as a social ill or disease can lead to some problematic conclusions unless properly contextualized. Framing crime as a disease is a tool of dehumanization that seeks to justify treating criminal offenders as a cancer: something to be removed or excised from society, or even killed.⁶¹ Humans have a pattern of dehumanizing others by comparing them to animals, nonconscious life forms, or inanimate objects to justify treating them as nonhumans. This is one of the darkest corners of our collective psyche; labelling people as rats, roaches, diseases to be cured, mud and filth, and a whole host of other analogies that form the basis for treating human beings like those things being analogized.⁶² Cages and steel bars are suitable for animals; a dustpan and trashcan are suitable for dirt and grime; excising and killing unhealthy cells is quite appropriate for cancer and biological diseases. We must resist the urge from our ancestors to go down this dehumanizing path because history bears witness to where it often leads. Labelling people that violate criminal laws as animals, dirt, or cancerous cells has the danger of repeating the punishment mistakes of the past.

Instead of using the cancer analogy and the broader disease/crime framing to dehumanize and justify excising or segregating cancerous cells, we should instead address the underlying problems to prevent the disease in the first place and to make conscious interventions for the social body to heal cells to prevent their criminal or antisocial growth.⁶³

The disease/crime framing also challenges us to consider the possibility that a treatment itself has become harmful to the social body. Just as chemotherapy introduces a carefully regulated level of poison to the body to kill cancerous cells, such a treatment can also wreak havoc on healthy cells and the body as a whole.⁶⁴ Unfortunately, chemotherapy also has the potential to lead to adverse health outcomes, including death, as part of the natural course of the treatment.⁶⁵ The considerations of punishment and mass incarceration must ask these same questions. How much has punishment had external and downstream impacts on the rest of society, and might it too also lead to negative outcomes for the body as a whole? And as will be discussed below, has the placebo treatment of mass incarceration itself become harmful to the body? The final part of the placebo analogy connects the role of doctors to various actors in the criminal justice system who prescribe and apply the treatment of punishment. Doctors play an important role in the healthcare system as care providers, and they deal directly with patients to give medical advice and prescribe treatments. Even with all of the medical advancements and growth in scientific knowledge, doctors continue to this very day to prescribe placebo treatments to their patients in clinical settings.⁶⁶ In this way, public officials in our executive departments and legislatures have built similar relationships of trust with their constituencies who have decried the social disease of crime; these decisionmakers have prescribed a solution of harsh criminal punishment as a way of treating this disease and returning society back to a state of health and harmony.⁶⁷ High-level politicians help facilitate and work with legislators and state actors such as police and prosecutors to deliver the treatment of punishment at the street level.⁶⁸

However, the incentives of politicians are not congruent with those of doctors. Consequently, we see politicians wield their prescribed treatment of punishment in a perverse loop meant to keep themselves in power. Instead of using criminal policies to make Americans feel safer, they use these policies to control public perception of crime in a way that ensures it will be politically advantageous. This is accomplished through the somewhat oversimplified political process that follows: Politicians and media whip up and create fear of crime in the populace because fear drives votes⁶⁹ and views⁷⁰ respectively, and voters give power to these politicians who promise to fix the problem.⁷¹ While crime is indeed real and its impacts often terrible on an individual and societal level, politicians perversely operationalize fear of wrongdoing to bolster their political power.⁷² Therefore, there is an incentive to maintain an optimal level of fear and to push the “fear” button every few years to ensure that voters continue to trust in the politician and their policies. In 2022, for example, Fox News ramped up its coverage of crime to prime the public to vote for law-and-order Republican candidates before the congressional midterm elections; after the elections, their coverage of crime dropped by a stunning 50%.⁷³ Studies have also shown that Americans’ perception of crime changes favorably when a member of their party occupies the White House.⁷⁴ Thus, while Americans do not feel safer from crime, they do feel that they are in safe hands when voting their preferred politician into office.⁷⁵ Putting these principles together, the placebo methodology takes shape by explaining the operation and incentives of using punishment as a treatment that turns out to be a mere placebo as a consequentialist project.⁷⁶ Most people, even in high-crime areas, aren’t often witnesses to actual crimes that unfold before them.⁷⁷ They hear about crime from their friends, they see it on the news, and they feel the anxiety of their neighbors.⁷⁸ Thus, lowering these perceptions of crime would have multiple social benefits. Businesses are more likely to invest in areas that they perceive as low-crime, individuals are more likely to civically engage in these communities, and even property values can be impacted by market perceptions of crime.⁷⁹ Thus, if a placebo policy of punishment made society feel better, the law would be justified as yielding positive future outcomes. This explains in part why politicians and decisionmakers have been so adamant to prioritize public perception of crime.⁸⁰ It is indeed a social lever they can turn to yield the social and political outcomes they desire.

The social and political aspects of crime policy are all tied together in a muddled mess that prioritizes perceptions of crime over the efficacy of crime policy. Thus, the placebo of crime and punishment in the modern era is indeed meant to control perception, but in a way that benefits the very people who prescribe the treatment.

Punishment operates on a placebo treatment cycle. The placebo methodology shows how punishment policies utilize similar social and cultural dynamics to control public perception and feeling toward the diagnosed problem, as opposed to treating the actual problem through effective means. Ultimately, this operates according to power, which is a similar motivation in most policy, medical and criminal alike.

B. The Placebo Effect

Conceptualizing punishment as a placebo gets even more nuanced if we theorize the placebo effect, which refers to the physical benefit that placebos can produce for patients. So far, this Article has focused on the dichotomy between efficacy and perception, arguing that punishment acts like a placebo because these criminal policies have little efficacy but instead operate to control public perception in ways that benefit politicians, law enforcement, and even the broader Prison Industrial Complex (“PIC”).⁸¹ But placebos provide more than mere psychological benefits. One of the curious miracles of placebo treatments is that if a doctor prescribes such a treatment to a patient, the patient will certainly feel better; but in some circumstances, the placebo will actually deliver a measurable physical benefit in which the patient’s body will actually improve from a biological and medical perspective. The medical literature has extensively studied this phenomenon called the “placebo effect,” in which a treatment that has little physical benefit can somewhat trick the patient’s body into activating its own powerful disease-fighting and recovery systems.⁸² This Section theorizes what this means for thinking about punishment as a placebo.

Modern medicine and sociology have come together to understand the mysteries of how the positive placebo effect works. Scientists have hypothesized that placebos release endorphins and dopamine in the brain, which have been shown to increase pleasure and decrease the feeling of pain;⁸³ these natural chemicals in the brain also impact a patient’s positive motivation and psychological outlook on their medical condition.⁸⁴ Sociologists add to this understanding by explaining the cultural and social meanings that give placebos such power.⁸⁵ First, the disease must be socially conceptualized since the diagnosis dictates the treatment. In some cultures, a headache might be the affliction of demonic forces, or it might be chemical activity in the brain.⁸⁶ Thus, the meaning and definition of the disease is the starting point to understand the cultural and social power of the placebo. Second, the doctor’s relationship to the patient also carries weight; Anne Harrington—a prominent leader in placebo studies—once characterized doctors as walking placebos because of their powerful role in shaping patient expectations.⁸⁷ If the doctor, healer, or shaman in the community is somebody whom a patient respects, trusts, and recognizes as a person of high social standing, these feelings will increase the patient’s own subjective belief that the doctor is prescribing them a treatment that is actually efficacious.⁸⁸ Third, patients’ own experiences shape their subjective social understanding and response to placebos. Research shows that patients who have had positive experiences with placebos in the past are good candidates for future placebo treatments.⁸⁹ Thus, hypothetically, if a patient had a good experience with eye of newt to treat their previous malady, this could subjectively impact their belief that eye of newt will perform well with a new ailment, or they would just as soon trust their doctor’s prescription of a crocodile tooth.

There is something about these medical and sociological factors that trigger the body’s own immune system towards positive physical results. Placebos can deliver positive medical results when treating things like the common cold and cardiovascular disease.⁹⁰ Placebos are also incredibly effective at treating more subjective medical ailments such as pain and depression.⁹¹ As it turns out, placebos can deliver positive psychological and physical results.

The placebo effect offers an interesting insight into crime and punishment as a social institution that also prioritizes and capitalizes on the subjective perceptions of the public in ironic ways that might indirectly produce public safety benefits. The crime and punishment literature has long had to grapple with two seemingly contradictory truths. On the one hand, crime rates have steadily declined in this country since the 1990s, which means we are actually getting safer.⁹² On the other hand, there is near universal agreement among criminologists, sociologists, and economists that modern incarceration policies have contributed very little, if it all, to this public safety outcome.⁹³

Scholars have tried to answer the all-important question of the factors that cause the rise and fall of crime. This Article offers a unique contribution to this literature by conceptualizing the placebo effect of punishment, which describes the indirect ways that punishing offenders over the past generation has triggered other social and institutional movements that are much more responsible for the actual drop in crime than the actual placebo of punishment itself. In other words, punishment acts like a placebo because it manipulates how the public perceives crime in ways that trigger other social and institutional behaviors that actually are responsible for the lower crime rate. While punishment may have a very small direct effect on crime, it has much more indirect impact on crime.

For instance, more prisons and the growth of the PIC might mean more police officers on the streets to fill these prisons. While the police and prisons are two separate institutions, there is tremendous practical overlap in how they complement each other and work together. Consequently, as the number of prisons across the nation skyrocketed during the 1990s prison-building boom, so too did politicians dedicate tremendous funds to hiring thousands of police officers to enforce the rising drug war.⁹⁴ In this way, punishment policies of incarceration indirectly supported the growth of police forces. And there is evidence that the presence of additional police officers on the street deters would-be offenders because it increases the likelihood that they will be caught in the act or effectively tracked down soon after committing a crime.⁹⁵ Similar to a placebo, punishment policies did not contribute to actually treating the disease of crime; instead, they triggered other institutional behavior to increase police presence in a way that more effectively lowered the crime rate.

In some places, there might also be a self-fulfilling prophecy for communities that feel safer and employ social behavior that then makes them safer. Pedestrian traffic is a well-documented feedback loop in this regard, because people who feel that their neighborhood is a safe place to walk will choose to walk more places during day and night hours; in turn, this increased pedestrian traffic means that there are more eyes, ears, and potential witnesses which deter criminal behavior in these neighborhoods.⁹⁶ It also might be interesting to theorize how punishment policies indirectly impact other social behaviors that have been linked to lower crime rates, such as increasing employment and high school graduation rates,⁹⁷ among others.

These few paragraphs are not meant to be exhaustive and represent a short foray into digging into the empirics of crime policy to understand how punishment policies might be impacting social and institutional behavior to indirectly have positive impacts on the social ill of crime. Just as placebos operate according to biological functions that are triggered through a patient’s social understandings and expectations, so too can punishment be theorized as a treatment to the social ill of crime that triggers beneficial behaviors that produce lower crime rates.

* * *

As a closing thought to this Part, I implore readers to not lose the forest for the trees. This Article has thus far argued that mass incarceration policies can be conceptualized as a placebo treatment that while not efficacious, may nevertheless deliver benefits to public perception. But admittedly, analogizing criminal punishment to medical placebo treatments is not a 1:1 comparison; nor is it a perfect analogy that is seamlessly intuitive or foolproof to the avid intellectual. For example, comparing politicians, police, and others to doctors leaves a bit to be desired given the unique doctor/patient relationship that may not be fully captured in the punishment context. Some might find that analogizing crime to a social disease is also problematic and might struggle to make the connection between an individual’s reaction to a placebo treatment and society’s reaction to the institution of punishment. But the frustration that comes from trying to perfectly fit an octagon peg into a round hole misses the fact that the shapes are similar enough to make observations that might be shared by both. Sure, it doesn’t fit perfectly, but they are indeed from the same family and the same logic of design.⁹⁸ Like all analogies, theorizing punishment as a placebo helps us conceptualize and understand similar principles in seemingly unrelated subjects.

The placebo methodology that this Article builds must be judged according to its ability to highlight latent truths about punishment policies. Consequently, this punishment as placebo analogy should be justified because its theoretical and explanatory contributions outweigh its imperfections around the edges.

II. Mass Incarceration as Placebo

The theoretical grounding of placebo punishment policies is perhaps most impactful in how we should reconceptualize the study and discussion of mass incarceration. Although the placebo methodology could be applied to many case studies of different punishment tools,⁹⁹ mass incarceration is the “beating heart”¹⁰⁰ of the criminal punishment system. It is the prison, as an institution, that is at the center of the wider penal imagination and has been imbedded with so much social and cultural meaning as to how we separate society from dangerous “others.”¹⁰¹ While the national population of incarcerated persons is declining,¹⁰² the carceral system still boasts physical control over 1.9 million people.¹⁰³ Nearly 7 million are surveilled under a growing e-carceration and parole population.¹⁰⁴ And the carceral state extends even further to the experiences of the 113 million individuals who have had close relatives incarcerated at some point in their lives.¹⁰⁵ In effect, one-third of Americans have been impacted by the wide and unforgiving reach of mass incarceration.

The placebo methodology as applied to mass incarceration argues that it is best understood as a set of laws and policies meant to control public perception of crime, playing on the subjective emotions of fear, safety, and the ever-elusive ideal of justice. Much has already been written about the failed efficacy of mass incarceration policies to improve public safety,¹⁰⁶ but this misses one of the foundational principles of the policy’s placebo: regardless of its actual effects, it still delivers costs and benefits based on how it controls public perception of crime.

This Part builds on the existing mass incarceration literature by discussing whether the benefits of perception are worth the costs of mass incarceration. Section II.A develops the history of mass incarceration through the lens of policies that would assuage public fears, and Section II.B continues by discussing the policies’ ineffectual impacts on public safety. In light of these historical and modern conclusions, Section II.C argues that the placebo methodology more accurately captures the intent and impacts of mass incarceration policies.

A. Research and Development

The history of fear and mass incarceration in the modern era begins with the unprecedented crime wave in the 1960s and 1970s. During these two decades, violent crime rose by 271%, burglary and robbery rose by 231% and 318%, respectively, and similar increases were reported for sex crimes and property crimes.¹⁰⁷ This fear of the “other” was further cemented by studies during the 1970s showing that a small number of offenders were responsible for a high proportion of offenses committed.¹⁰⁸ Further, criminological studies began persuading lawmakers that rehabilitation did not work consistently¹⁰⁹ and that high levels of recidivism would continue without more incapacitation efforts.¹¹⁰

The Nixon administration, for its part, was driven by its own fear of political opposition, namely from recently empowered African Americans after the Civil Rights movement and anti-war “Hippies,” two groups that threatened Nixon in national politics.¹¹¹ Thus, Nixonian criminal politics was defined by convincing the nation that “the whole problem is really the blacks . . . The key is to devise a system that recognizes this while not appearing to.”¹¹² While there was already a racialized fear of the dangerous “other” imbedded in American culture,¹¹³ the rise in crime during the same era as the Civil Rights and anti-war movements prompted an unprecedented fear of crime that was directed towards certain communities.¹¹⁴

By targeting these “others”—groups that were outside of the mainstream social and cultural understandings of American white middle-class identity—Nixon captured the populism of his era. Social scientists have long recognized that the fear of crime became a defining “cultural theme” during this era that called for policies that “aim[ed] to reduce fear levels.”¹¹⁵ Empirical criminologists of the time were baffled by these changes. The mass incarceration policies developing before their very eyes had little to do with actual crime control or preventing victimization but were instead driven by fear and myth that traded on racial and cultural stereotypes emerging from the anxieties of suburbanite imagination and political rhetoric.¹¹⁶ And this penal populism of fear has been carefully nurtured as a winning political strategy ever since.¹¹⁷ National and local actors alike adopted this tough-on-crime strategy, promising more punitive punishment as a way of promising to reduce fear and anxiety related to crime.¹¹⁸ Consequently, fear was increasingly used as a justification and catalyst for harsher punishments, and politicians were all too willing to stoke, control, and benefit from these public fears.¹¹⁹

B. Treatment Efficacy

Although controlling public perception and fear of crime was one of the primary drivers of developing mass incarceration policies, their efficacy in reducing crime should not be ignored. The mass incarceration policies of the 1980s and onward have enjoyed association with lower crime rates because crime has generally gone down over the past thirty years.¹²⁰ But there is debate among criminologists and economists about the actual correlation between incarceration rates and crime rates.

The weight of analysis tends to agree that there is some correlation between the population of incarcerated persons and the crime rate, but the connection is minimal.¹²¹ In just twenty years, between 1980 and 2000, the incarceration rate exploded by over 300%.¹²² And yet, this treatment has yielded marginal utility. For example, Steven Levitt found that increased rates of incarceration in the 1990s were only responsible for 25% of the drop in crime during that time.¹²³ Don Stemen found that between 1980 and 2000, every 10% increase in the prison population was only responsible for a 2% to 4% decrease in the crime rate.¹²⁴ Additionally, many others have come to argue that there is a much lower correlation; they posit that the exploding incarceration rate was only responsible for a 2% to 15% decrease in crime rates.¹²⁵ But even the high watermark of a 25% impact should make us question whether such a use of resources is justified when these resources could be reinvested in programs that have been proven to be far more efficient at lowering crime rates.¹²⁶

Another group of scholars refutes the above by arguing that there is no statistically significant correlation between the increase in the population of incarcerated persons and the decrease in crime.¹²⁷ Starting in the 1990s, the National Criminal Justice Commission found that there is “little or no correlation between rates of crime and the number of people in prison.”¹²⁸ Twenty years later, during a time when the incarcerated population reached its height,¹²⁹ Jeffrey Fagan and Tracey Mears found that increasing the number of people incarcerated resulted in “stable if not higher levels of crime.” They decried the lack of commensurate public safety benefit for the ever-increasing rates of incarceration.¹³⁰

A smaller group of scholars have come to even starker conclusions, arguing that mass incarceration has had negative impacts on public safety, causing a criminogenic effect. Don Stemen found that when incarceration rates increase past 325 to 492 inmates per 100,000 people (of which many states have far surpassed), crime rates begin to rise.¹³¹ Francis Cullen, Cheryl Jonson, and Daniel Nagin performed a systematic review of multiple meta-analyses of the impacts that incarceration has on crime rates and reported that in the most compelling studies, incarceration was not associated with any deterrent effects but instead increased recidivism rates anywhere from 7 to 14%.¹³² These authors have linked this data to a number of potential explanations since incarcerating large numbers of people can depress economies, threaten social network formation, and politically disenfranchise communities, which can then lead to criminal behavior.¹³³

On the other end of the spectrum, there are scholars that have defended the status quo, arguing that mass incarceration policies in the 1990s were causally related to lower crime rates. This literature is most associated with the harshest criminal sentencing policies, such as the three strikes laws that went into effect during that time.¹³⁴ But even among these scholars, there was growing unease about the continued growth and marginal impacts of the carceral state. The intellectual and empirical journey of John DiIulio, Jr. illustrates this point. In 1996, he argued that prisons were a bargain since locking up a person for “about $25,000” per year in the prison system saved society from the potential murder and mayhem he or she would otherwise be committing on the streets.¹³⁵ As a Democrat, he channeled bipartisan fears in the 1990s, stating that “a wave of young, violent, mostly minority, remorseless juvenile delinquents, known as super-predators would plague the country and increase crime rates” without harsh incapacitation measures.¹³⁶ Yet only a few years later in 1999, he realized that mass incarceration had reached the limits of its utility. He argued that locking up more people would not have any further public safety benefit and that it was time to shift the penal focus to drug treatment and effective parole programs.¹³⁷ Unfortunately, politicians benefiting from mass incarceration were not so keen to heed his advice.

These seemingly conflicting streams of data can be best harmonized with a general model of marginal utility across four stages.¹³⁸ At the first stage, studies have shown that incarceration is most effective as a deterrent and incapacitation tool when the crime rate is high.¹³⁹ For every unit of punishment, you get a higher payoff in crime reduction. At the second stage, punishment suffers from the law of diminishing returns, in which every unit of punishment still achieves a measure of crime reduction but much less than the previous stage.¹⁴⁰ At the third stage, the crime rate levels off, during which every unit of punishment does not achieve any payoff in crime reduction.¹⁴¹ At the fourth stage, there is an inflection point at which punishing offenders actually increases the crime rate.¹⁴²

Like many medical treatments, law and policy can have similar diminishing returns over time as social problems and disease adapt in the same way that all organisms adapt and evolve in the physical and social body. As a result, a treatment program must be adaptive to long-term illnesses and should be administered differently in different phases.

These statistical studies, cost-benefit analyses, and other tools of administrative expertise have been abandoned in the criminal punishment regime.¹⁴³ There are decades of data to which this Article cannot do justice that question and prove the inefficacy of mass incarceration policies to improve public safety. Thus, a policy that fails to track and account for actual impacts on public safety should be understood as prioritizing the placebo methodology’s focus on tracking and accounting for public perception of crime, which is all too vulnerable to political power, abuse, and perverse incentives.

C. Placebo Responses

To summarize, mass incarceration policies do not deliver on their public safety promises and were never solely designed to do so. Out of a culture of fear, myth, and superstition, our society decided to pivot from careful, rehabilitative carceral efforts to a mass incapacitation effort to satisfy collective retributive cultural morality and expressive condemnation of the “others” who we feared.¹⁴⁴ We now benefit from thirty years of data that shows that mass incarceration policies are no longer (and may never have been) effective at lowering crime rates. Yet we still choose to punish harshly and ineffectively. This Article concludes that modern punishment can no longer be justified under the veneer of public safety. Rather, it is motivated by controlling public perception and fear of crime in a way that subjugates entire communities to the benefit of those in power. This Section further cements the data connecting medical and social placebos in the punishment context. First, placebos enjoy the largest impacts on patient outcomes when the patient is experiencing the highest amount of symptoms. For example, placebos deliver the highest amount of pain relief when the patient is experiencing the most amount of pain.¹⁴⁵ This makes intuitive sense; when somebody is experiencing pain, their main concern is finding something to subside that pain, so they are more susceptible to and even desperate for any treatment that a trusted doctor tells them will provide relief. The subjective aspect of pain means that the placebo’s effects are judged in relation to the level of pain the patient is suffering. The higher the pain, the more the placebo relieves.¹⁴⁶ The very nature of placebos means that they do not deliver an objective and consistent level of psychological and therapeutic benefit; it is all relative to the patient’s perception of their problem.¹⁴⁷

This finding also makes intuitive sense when considering the placebo methodology approach to punishment. Public support for mass incarceration policies was highest when the “disease” of crime had the most crippling impacts on public fear. Data shows that public support of harsh tough-on-crime policies was highest from the 1970s through the 1990s but has declined as of the late 2000s.¹⁴⁸ Criminologist Lisa Miller has documented this by studying public perception of crime and concluded that politicians’ ability to manipulate public perception of crime for political gain is highest when violent crime rates are high; this power of perception sharply declines when violent crime rates are lower.¹⁴⁹ To simplify, the public enjoys a higher positive reaction to the placebo treatment of mass incarceration when its fear of crime is at its highest. As the disease subsides, so does the effectiveness of the placebo.

Second, placebos also share in the law of diminishing returns. This is described as a “placebo sag” in the medical literature to indicate that there is a danger of treatment complacency.¹⁵⁰ Although patients may receive tremendous therapeutic benefit from Placebo A, the benefit wanes over time as patient perceptions of what they believe to be an efficacious treatment continues to decline. It is an understandable psychological cycle if a patient perceives Placebo A as not having the same effect as it did a month ago; thus, their future expectations based on their own experience contribute to the sag. Continuing to prescribe Placebo A will not deceive patients forever—or society for that matter.

The support for mass incarceration has also started to sag. Indeed, this goes a long way in explaining recent movements for criminal justice reform;¹⁵¹ after over thirty years of mass incarceration policies, the public is starting to realize that the placebo is losing some of its luster. There has been a steady drumbeat toward reform, but this must be tempered with continued fluctuation in public perception of crime. When the public is confronted with political and media messaging that suggests a spike in crime, public support for harsh criminal punishments returns in certain segments of the social body.¹⁵² This ever-changing landscape of public perception—and the powerful harbingers of doom that continue to press these “fear” buttons—is best understood as a placebo method of controlling public perception to maximize political gain.

Finally, perhaps the most consequential aspect of placebo treatments is that placebos can indirectly have a negative impact on a patient’s health by giving them a false sense of healing. When a patient takes a placebo treatment, three things can happen regarding the patient’s actual physical health. First, the placebo makes the patient feel better, which has a positive impact on the body’s ability to physiologically fight the disease and the patient actually gets better. Second, the placebo has some effect on the patient’s feelings but has no impact on the body; in this scenario, the disease is allowed to run its natural course and may result in the patient’s natural immune system fighting off the disease and the patient improving. Third, take the same scenario, but the disease runs its natural course, resulting in the patient getting worse.¹⁵³ As a result, placebos can indeed be dangerous depending on the seriousness of the affliction. All the while, the patient believes they are receiving an efficacious treatment and forgoes seeking other treatments that might be more effective.

This third scenario has important ramifications for the status quo and how we chart future treatments for the social problem of crime. If we are indeed in an era in which mass incarceration has a criminogenic effect,¹⁵⁴ then the placebo is truly lulling society into a false sense of feeling better. Because mass incarceration policies operate in the modern era primarily as a placebo, this comes at the expense of the disease of crime being permitted to run its natural course and make us all worse off than we would otherwise be under a different punishment regime.

III. Socio-Medical Ethics and Justifications

The study, theory, and practice of punishment can benefit from the same reflection that placebos have sparked in medicine, but the synergies between these fields still spark questions of ethics and justifications. This Part builds on this theme by applying considerations of socio-medical ethics to frame and answer whether placebos in the punishment sphere can be justified. Like medicine, these answers are nuanced and context specific.¹⁵⁵ There are indeed a number of different ethical justifications to consider, but this Part focuses on the issues that carry the most insight for future discussion inside the punishment literature. This Part also focuses on perhaps the most important question of punishment theory and answers in the negative: Is punishment justified under the placebo methodology?

Section III.A considers perhaps the most important question in the placebo methodology, which is the harm principle of punishment when considered against the “do no harm” principle of medicine. These different approaches are important when considering the harm and social costs that punishment wreaks across entire communities. Section III.B explores the important ethical differences between the doctor-patient and politician-constituent power dynamics. Benevolent deception and medical expertise are at the heart of the doctor-patient placebo treatment schedule, but such benevolence and policy expertise are missing from the politician-constituent power dynamic. Section III.C discusses the ethics of experimentation with placebo research toward finding efficacious treatments. This contributes to how we might view the dark past of experimenting with harsh punishments and informs us towards a more ethical future. Section III.D covers the economics of placebos and their continued popularity and use in the medical field. Just as economic markets have a way of shaping how placebos are prescribed by doctors and perceived by patients, so too can the economics of punishment shape our social and cultural understandings in ways that can lead toward a different punishment regime. In turn, each vignette also considers policy solutions that seek to remedy the ethical and practical problems of punishment placebos and their overreliance on controlling and even manipulating public perception.

The discussion and solutions proposed throughout each Section are not magical or mystical cures that can solve all of the many problems and social harms of mass incarceration. These realities and strategies of punishment reform—which are unique to placebo understandings of punishment—are meant to mitigate and not propagate. The most important concern is whether these discussions and solutions themselves become placebos, proposed and advocated by a school of scholars as a way of assuring and placating more revolutionary thinkers away from pursuing transformational work.¹⁵⁶ Getting closer to equitable justice is worthwhile, and trading among lesser evils is often the difficult choice scholars and policymakers must pursue to achieve incremental progress that is transformative in scope.¹⁵⁷

A. Placebos as Harm

Although this Article focuses on the synergies between the medical field of placebos and the policy design of mass incarceration, there are important differences that shed light on the theoretical and practical framing of punishment. Perhaps the biggest shortcoming of the placebo methodology of punishment is the difference between how these fields view harm.¹⁵⁸ By recognizing these shortcomings, this Section seeks to further refine and tailor the placebo methodology to the specific attributes of the punishment context.

The medical field’s approach to “do no harm” is an important foundational ethic upon which the entire institution is built. This principle traces back to the Hippocratic Oath¹⁵⁹ and undergirds the purpose of the entire medical field. This is of prime importance when applied to placebo treatments. Placebos, by definition, must be either inert or have active ingredients that will not harm the patient.¹⁶⁰ At best, they may help the patient; at worst, they will do no harm.¹⁶¹

The opposite is true of the punishment field, which is specifically designed to impose harm on the criminal offender.¹⁶² This harm has come in many forms throughout history,¹⁶³ and modern criminal punishment continues to take away some aspect of the offender’s liberty by exerting government control over the offender’s body, including incarceration, e-carceration, and parole.¹⁶⁴ Capital punishment, of course, imposes the ultimate harm upon the offender by ending their life. This is not merely an accidental byproduct; it is the intended purpose of punishment.

This requires at least two analyses that can be bifurcated into theoretical harm and practical harm, both of which are important to answer whether these harms of punishment might be justified under the placebo methodology. On the theoretical side, we must remember the conceptual framework outlined in Part I. If society is like the body, and crime is like a disease, then treating the disease benefits the body at the expense of the disease. If punishment were effective, it would serve the same purpose as medicine by treating and getting rid of the disease of crime to the benefit of society. This becomes more problematic when this social disease is being directly caused by humans in society. As discussed above, this need not require inhumane treatment to punish and/or rehabilitate them. Thus, punishment is intended to impose some level of accountability on the offender just as medicine is intended to treat infirmities and the body of society benefits.

Under this theoretical treatment, the placebo policies of punishment become problematic because punishment is used to deliver harm to offenders regardless of any actual benefit to society. The punishment placebo would not merely be inert as in the medical context but would be delivering harm to a large segment of society—primarily poor people of color¹⁶⁵—for the mere psychological or therapeutic benefit of everybody else and for the political benefit of the powerful. Delivering this level of theoretical harm would not be justified under utilitarian or deontological justifications. The utilitarian might be persuaded, but ultimately must consider the costs and benefits of the policy to understand the greater good. Here, placebo policies are pitting the minimal benefits of manipulating public perception against the social costs of imposing harm on millions of people and tens of millions of families.¹⁶⁶ The level of harm hardly seems justified according to the minimal benefits and the perverse political incentives behind it. Kantian deontologists would also oppose a placebo justification, since this school of thought rejects using a person as some means to achieve a broader social goal.¹⁶⁷ But modern retributivists may still justify a placebo punishment if it indeed makes society feel that moral justice has been paid to the offender.¹⁶⁸

On the practical side, the placebo of punishment policies results in actual harm being visited upon entire communities that face the brunt of mass incarceration, which adds further consideration to the utilitarian balancing above. It is not enough to recognize the tremendous harm done to the individual offender during and after their time incarcerated, but these harms extend to their communities as underappreciated externalities.

Mass incarceration policies have devastated economic growth in these communities.¹⁶⁹ Working-age men and boys are often the most targeted due to their involvement with criminal behavior; when these communities are hollowed by the incarceration of these men, small and large businesses alike are less likely to invest and survive in these communities.¹⁷⁰ And even when incarcerated persons are on parole or get out of prison, they face an uphill battle when finding a job and are often paid less than their non-offender competition.¹⁷¹

Social networks and family formation also suffer objectively negative outcomes from mass incarceration. In heterosexual relationships, it becomes harder for women to find suitable mates if many potential options are incarcerated or have difficulty assimilating back into society after being incarcerated.¹⁷² Incarceration also leads to the breakdown of strong family dynamics, since so many incarcerated persons are held in places that are geographically far away and thus prevent in-person visits from their family. In addition, the exploitative prison-phone industry creates cost prohibitive conditions for many poor families to stay in touch over the phone.¹⁷³ Also, many incarcerated persons are parents, which has negative behavioral and educational impacts on their children who grow up without them.¹⁷⁴

There are also tremendous negative political impacts on these communities. Voting-age incarcerated persons are not able to vote during their incarceration. In some jurisdictions, these persons are permanently disenfranchised even after their incarceration.¹⁷⁵ Not only does this have measurable impact on elections, but it also contributes to the individual offender’s sense of self as a nonpolitical actor. Political scientists and sociologists have shown how affected communities have developed a sense of isolationism from civic discourse.¹⁷⁶ This impacts the politician-constituent relationship and has served to undermine the legitimacy of the criminal legal system as a whole in these communities.¹⁷⁷ Further, as explained above, mass incarceration can have a criminogenic effect if it contributes to community perceptions of illegitimacy that following the rule of law has little benefit.¹⁷⁸

This short summary of actual harm caused by the placebo of mass incarceration policies cannot be justified under any view other than those that recognize mass incarceration’s intent to maintain political power and use that power to subjugate communities.¹⁷⁹ If the goal is to make the streets safer, that goal has failed. If the goal is to maintain existing power dynamics, then placebos in medical and punishment contexts are quite effective at controlling perceptions of crime and disease and cementing the doctor and politician as the conduit through which things will be perceived as getting “better.”

Perhaps Vincent Chiao said it best: “To rely on criminal punishment solely . . . to make other people feel safer, even when doing so does not make them actually safer” is not justified because it is hard to see how these people’s subjective feelings “could be sufficiently important to justify imposing such serious costs on others.”¹⁸⁰

B. Deception Versus Consent

Placebos also introduce the ethical dilemma in medical treatment of utilitarian notions of deception, which is all too relevant to the punishment policies of the modern era. In the broader social context, lying, fraud, and deception are viewed as unacceptable social behaviors that can emotionally hurt and financially injure a victim.¹⁸¹ But like most social mores, there are exceptions to this general rule that might be applicable to the placebo punishment context. For instance, intentionally lying might be justified if it produces beneficial results and emotions or even serves as a lesser evil to avoid more harm.¹⁸² Indeed, we paternalistically deceive our children;¹⁸³ we might sugarcoat the truth to shield the feelings of a spouse or loved one; and even in the cold world of legal ethics, it is ethical to intentionally withhold information from a client for their benefit.¹⁸⁴

There are at least three ethical considerations that, when considered together, might be used to justify placebo treatments in the medical field that are found lacking in the context of punishment policies. The first considers the expertise of the person in power. In the medical field, placebo practitioners are medical experts who can correctly dose and administer placebos as a treatment specialized for each individual patient.¹⁸⁵ Doctors prescribe placebos because they often follow the science showing how effective they are in assuaging subjective patient ailments such as pain, depression, and other complaints.¹⁸⁶ Indeed, it is this very expertise, built on years of experience and advanced studies, that supports any potential deception. The only reason the doctor knows such a treatment will work is because they have likely used these placebos in the past to provide therapeutic treatment and have tracked the progress of past patients in conjunction with studying the scientific literature.

To illustrate the importance of expertise and how it implicates trust, contrast the following situations: first, a doctor prescribes a placebo to a patient; second, a person pretending to be a doctor prescribes a placebo to a patient. Deception is taking place in both scenarios, but most would admit to being more comfortable with the former. If a doctor prescribes a placebo, there is a certain level of trust on the part of the patient toward the doctor that the doctor knows what they are doing; a patient may not like the doctor’s decision to prescribe a placebo, but many would feel comfortable trusting the doctor’s experience. In the latter situation, the ultimate medical result might be the same, but many would still be uncomfortable since the fake doctor has no expertise whatsoever. The patient received a placebo in both situations but would no longer have the same trust and understanding of the treatment in the hands of a fraud. Even if the patient experienced a positive placebo effect in both situations, the expertise and trust in the doctor in the first situation seems more justified than the deception and lack of expertise in the second situation.

In the punishment context, there is a much looser sense of expertise from our legislators and executive actors who are prescribing and delivering the treatment of punishment and mass incarceration to society.¹⁸⁷ Politicians are not experts and many are not even former practitioners of criminal policy and punishment.¹⁸⁸ They are lawyers, businesspeople, career public servants, and the like that bring generalist viewpoints to a host of different issues. Rachel Barkow opens her book Prisoners of Politics with the stark statement that “[w]e do not rely on experts or use studies and rational assessment to minimize crime” but instead base criminal policy “on emotions and the gut reactions of laypeople.”¹⁸⁹ This is a far cry from nearly every other consequential policy area, including public health, the environment, financial institutions, and other complex social issues.¹⁹⁰ Instead, political incentives, penal populism, and fear of the “other” all rule supreme in crime policy.¹⁹¹

Practitioners like police and prosecutors are a bit more mixed. There are still many who subscribe to outdated and disproved methods such as the broken-windows theory,¹⁹² predictive policing,¹⁹³ overcharging and stacking,¹⁹⁴ and pushing for harsh sentences for incapacitation purposes. However, there are several more progressive practitioners who are challenging the old guard and trying to use emerging data to facilitate more effective policies.¹⁹⁵ This is the double-edged sword of expertise, because both of these camps can be considered to be “experts” due to their level of practical experience. Just like doctors who prescribe a large number of placebos to patients with the hope that it will have a placebo effect, these practitioners on the ground are prescribing punishment policies that they hope will have some effect on public safety, whether directly or indirectly. And we the public have some sense of trust in these practitioners, in part, because of their expertise.¹⁹⁶ The takeaway from this analysis is that there can be no placebo-like justification for non-expert legislators who enact criminal statutes, but there may be some argument that the expertise of police and prosecutors should be given some weight.

The second ethical consideration that might justify deception is the intent of the person in power, which also highlights stark differences between the medical and punishment fields. Many doctors prescribe placebos out of benevolent care only when they view it to be in the best interests of the patient.¹⁹⁷ This gestures toward the “classical paternalistic paradigm that dominated traditional medicine.”¹⁹⁸ Like parents who benevolently guide, deceive, and withhold information from their children because of large gaps in information, intellectual ability, and emotional stability, the traditional practice of medicine justified treatments under a paternalistic view of patients.¹⁹⁹ Not only does this ethical justification substantiate the earlier requirements of expertise, but it also highlights the necessary selfless intent of doctors to justify deception. Parents deceive their children most often for the sole benefit of the child, and while there are indeed some that might use such deception for their own parental gain, this is often not accepted as best practice. Similarly, doctors may be justified in using benevolent deception if the primary purpose is the sole benefit of the patient. As argued above, this is a complex question in which a doctor (or politician) must consider the full range of costs and benefits of deception. But the ultimate takeaway is that deception is partly justified by benevolent and selfless intent.

Using this methodology to analyze politicians and their role in punishment leaves much to be desired. In a perfect world, politicians would be public servants who seek to serve their constituents’ best interests, but political theory has long disproved this romanticized ideal.²⁰⁰ Lawmakers are not bound by a comparable code of medical ethics where constituent interests are placed at the forefront.²⁰¹ They do indeed seek to serve their constituents but are also incentivized into doing what is necessary to maintain their power.²⁰² There may be an argument that politicians believe they are serving society by assuaging fears with placebo punishment policies if they believe that fixing the actual crime rate is not as important to public health as controlling perceptions of crime. However, as argued above, there are simply too many self-interested incentives involved in the tough-on-crime movement from those wishing to maintain their power. The lies that politicians maintain about mass incarceration, the manipulation of public fear, and the withholding of the truth from the public cannot be justified under the medical ethics of benevolence,²⁰³ but should be characterized as maleficent and at least partially self-interested deception that wreaks havoc in the form of underappreciated social costs.

Yet there is some nuance to consider regarding intent among lawmakers, police, prosecutors, and others who prescribe and deliver the placebo treatment of punishment. As in the substantive criminal law, we assign blameworthiness according to a person’s level of intent for their wrongdoing. The highest level of culpability is committing an act with purpose, meaning that it was the person’s conscious desire to commit the act or to procure the desired result of the crime.²⁰⁴ Next comes knowledge, which describes a person who has near certainty that their action will procure the result of the crime.²⁰⁵ Then there is recklessness and negligence, which capture situations in which a person knows or should have known that their actions will procure a certain result.²⁰⁶ There are indeed actors in the criminal legal system who are purposive or at least have sufficient knowledge that these punishment policies are ineffective but nevertheless have some placebo effects towards public safety. These are the most culpable actors in the mass incarceration crisis. But there are others who are less culpable but still responsible for contributing to the crisis. These decisionmakers might honestly believe the status quo is an effective system that they seek to propagate and expand. At best, these lawmakers, law enforcement officials, and prosecutors can be said to be reckless or even negligent because they are consciously disregarding the danger of these policies or really do not perceive any dangerousness but nevertheless should have known the dangerousness of these policies. Given the amount of scholarship, the accessibility of experts, and the continued deterioration of the criminal legal system, these decisionmakers are at best negligent, and at worst purposive, in their culpability of the mass incarceration crisis.

The third ethical consideration that might justify deception in the use of placebos is the issue of consent. Modern medicine has largely departed from the benevolent paternalism of the past and has been sharply redefined towards a progressive ideal that fosters partnership between doctor and patient.²⁰⁷ This type of dynamic has restructured the entire culture of medical treatment, in which doctors are resources of information that a patient can consult; by explaining complex medical information and treatments to their patients, doctors can partner with their patients and ultimately allow the patient to make the final decision on what treatment they receive. Thus, the modern ethics of prescribing placebos is complicated by the issue of consent.

This ethical dynamic is resolved through a fascinating aspect of placebo treatments, namely the consent of self-deception. Patients may desire to gain the benefits of placebos or other unverified alternative medical treatments for the sake of achieving a larger health goal.²⁰⁸ Perhaps an active medication is known to have side effects, whereas a placebo might deliver similar therapeutic relief without the side effects. Or perhaps active medication is cost-prohibitive to patients whereas placebos are more cost-effective for the relative benefit the patient receives.²⁰⁹ Medical ethicists have considered these possibilities and ultimately come out in favor of patient discretion; in other words, patients should have the power and control over their treatment to give doctors the green light to deceive them.²¹⁰ Under such treatment plans, the patient will never know which treatment is a placebo or when the deception is coming. That way, they can still enjoy the ignorant bliss of placebo benefits. This adds yet another wrinkle in an already complicated ethical space. As opposed to the principle of informed consent, in which the patient must consent to each separate treatment, a doctor-patient relationship may also be formed on the basis of general consent under which the patient consents to the doctor or the overall treatment plan.²¹¹ This obviates the need to consent to each individual treatment and instead gives the patient the power to consent to the doctor and all of their subsequent methods.²¹² If patients want to be deceived, and give consent to be deceived,²¹³ placebos might also be justified under a consensual doctor-patient relationship.

The self-deception model of consent is an interesting consideration of collective cognitive dissonance and might also have some explanatory value in placebo punishment policies. Let us first consider that most Americans want criminal justice reform.²¹⁴ There may indeed be much disagreement among these constituents about what such reform would look like, but there is a growing dissatisfaction. But like most political opinions, these Americans are looking for a type of general consent as opposed to informed consent. These constituents simply want the problem to get better and are not necessarily invested in each reform proposal that might come to the table. This is our system of republican government: we vote in politicians, thereby giving some form of majoritarian consent, and we expect them to fix the problems we have elected them to fix. Whether it be sentencing reform, decriminalization of drug crimes, or defunding the carceral state and reinvesting in other social programs, the average American voter is not concerned with the specifics of how politicians address these problems.

The public has accepted the diagnosis of the social problem of crime, and they are giving general consent to the overall treatment plan of mass incarceration with little care or concern about the overall placebo of the treatment. This is quite different from saying that the public wants to be deceived, but the political system allows for some consent for elected officials to deceive us, even after the public collectively understands that the criminal legal system is broken.²¹⁵ A political theory of self-deception, as realized through the placebo methodology, might be the strongest case that justifies politicians intentionally deceiving the public through mass incarceration and punishment placebos. If the public implicitly agrees to deception, politicians should be allowed to deceive.

Most considerations in this Section have failed to justify the use of placebo policies in the punishment context, and these medical ethical concerns can also inform some future policy safeguards. Whatever the next tectonic shift in carceral policies may bring, we must be aware of the placebo nature and politics of these policies. Politicians will support what makes voters feel better; for the Left, this may look like progressive policies that make them feel like criminal justice is no longer targeting certain communities; for the Right, this may look like less costly and more efficient policies that make them feel safer and more fiscally responsible.²¹⁶ There is indeed political will to make a change, as Red and Blue states across the country have experimented with new punishment policies,²¹⁷ presidential administrations from both parties have pushed through reformist legislation,²¹⁸ and grassroots organizations have become part of the mainstream decarceration movement.²¹⁹ But even with this bipartisan support to change something about the system, will the resulting change be yet another placebo? The next generation must not make the same mistakes of our past by settling for yet another placebo punishment policy.

There are no easy answers to what appears to be an insurmountable social and cultural problem based on political incentives and unethical behavior. An informed consent punishment model would be an interesting thought experiment, requiring partnership and active participation between politicians and constituents before any new criminal law, punishment, or reform was adopted. However, such an informed consent model would be practically impossible and would also theoretically fail to address the problems of punishment placebos. Putting so much power in the hands of uninformed constituents carries the risk of even more criminal legal policy falling prey to fear-based penal populism.²²⁰ Providing the public with more information, facts, studies, statistics, and so forth in an effort to educate the masses through public service announcements is unlikely to decouple the social and cultural feelings of punishment and its connection to safety. Using direct voter ballot initiatives is another interesting proposal because they have had success in legalizing marijuana and banning forced prison labor,²²¹ but have also maintained capital punishment²²² and a number of harsh sentencing laws.²²³ There is simply no safeguard for people’s fear and none for politicians’ penchant for exploiting it.

This Section has leaned heavily into diagnoses and has been light on effectual treatments. The ethics of social deception of placebo punishment policies cannot be justified through the same traditional medical lens of paternalism nor the benevolent power that doctors use to maintain public health in the community. Neither can placebo punishment policies be fully justified through modern medical ethics of informed consent, but the concept of general consent and self-deception paints a dismal picture of the American public’s apathy toward specific involvement in criminal justice. This leaves hope that further examination of placebo ethics can lead to some policy interventions, especially when considering the infrastructure of expertise below.

C. Experimentation Ethics

The placebo ethics of scientific experimentation introduces another tool that shifts the discussion of punishment theory and mass incarceration. The medical field often uses placebo trials as a methodology for finding treatments that are more effective than their predecessors, which justifies the use of the placebos toward the gain of medical knowledge. This Section uses this ethical justification of placebos to explore the unjustified experimentation of punishment policies and mass incarceration.

Starting with the medical context, double-blind placebo studies are the gold standard in the experimental setting to determine the efficacy of pharmaceutical drugs.²²⁴ There are at least two groups²²⁵ of consenting patients,²²⁶ both of whom are suffering from a disease or ailment that serves as the subject of the treatment. The first blind describes the patients who are unaware of what treatment they are receiving. The first group is given the drug being tested by the trial, while the second group is given a placebo.²²⁷ The second blind describes the researchers themselves. To avoid the power, bias, and favoritism of researchers who are administering the treatments,²²⁸ the researchers are also blind to which group is getting the drug versus placebo treatments. Thus, the efficacy of the drug being tested is the positive difference between the medical outcomes of the drug group and the placebo group. If both the drug group and placebo group enjoy physical improvements, isolating the positive impact of the placebo is paramount to determining the actual effectiveness of the drug apart from any placebo effects.²²⁹

The utilitarian outcomes of experimenting to develop more effective medicines justify the necessary deception of the placebo groups along the way. If the placebo group knew they were getting the placebo and the drug group knew they were getting the active drug, the positive placebo effects would be lost and would jeopardize the utility of the study.²³⁰

Medical ethicists have raised concerns based on deontological grounds of the duty of doctors to always provide the best available care.²³¹ This goes past the Hippocratic Oath to do no harm and requires doctors to provide the best care possible based on the inherent worth of a human life and its connection to individual health. These ethicists have criticized placebo trials as exploitative and have argued that it is inherently unethical to evaluate new treatments when there are already proven treatments available.²³²

For example, if a research trial is trying to find the next generation of treatments for a non-deadly disease, such as depression, a placebo trial may be organized. The new drug to be tested is X, even though Y is already on the market and available to patients. Y has an efficacy rating of one, which is quite effective at helping depression patients. However, the placebo group will not receive any effective treatment for their depression. Thus, their depression may get better or worse based on the placebo effect or the natural course of the disease. The drug group receiving X, on the other hand, sees a great improvement, measured at an efficacy rating of two. This result is traditionally seen as a success in medical science because X will benefit countless future depression patients, but it came at the expense of the placebo group who all the while could have gotten better if they had been prescribed Y.

In the punishment placebo context, there are many lessons to learn from medicine’s more developed and sophisticated rules of ethics. Placebo methodologies have long been used in sociological and policy studies to determine the efficacy of criminal law and policy when compared to the status quo or an inert placebo policy.²³³ These types of experiments and analyses have also been a hallmark of Our Federalism.²³⁴ The Brandeisian ideal of state laboratories has yielded new innovations in punishment practice but has often yielded harsher and more ineffective punishments as a result of the fear and penal populism that has taken hold since the tough-on-crime era.²³⁵ Three strikes laws, ending parole for certain offenses, and even the very development of prisons as an institution, were all policies started and championed by various states until they were more widely adopted.²³⁶ And while states have been experimenting with many decarceration policies,²³⁷ the norm still rests on the harsh side of the scale.

Given the placebo nature of mass incarceration policies, they are difficult to justify unless the punishment system is becoming more efficacious, like the ethics in the medical field. States will continue to experiment with various progressive and harsh punishment policies, and they do so in part to align these policies with the preferences of their constituency.²³⁸ Thus, if Jurisdiction A continues with the status quo of mass incarceration policies and Jurisdiction B decides to experiment with bail reform, lower sentences, or drug courts, there would be a value to criminology and punishment theory if the delta between these similarly situated jurisdictions could be measured. Jurisdiction B might prove that these new policies are ineffective, or they might create support for a new paradigm in punishment, as has been the case throughout federalism’s state experimentation in criminal law.

For the reform-minded utilitarian, mass incarceration is a necessary part of the experiment to prove to politicians and bureaucrats around the country that progressive punishment reform is more efficacious than the current placebo. The delta between the placebo of mass incarceration policies and the growing number of programs that have been shown to reduce first-time offenses and recidivism²³⁹ is a powerful policy tool that justifies change. This is true even though it comes at the cost measured in years of human lives. But while such reformers use metrics to prove their efficacy points, mass incarceration as a placebo is allowed to remain across the country, and the disease of crime is allowed to fester.

For the abolitionist-minded deontologist, there are two groups of people to consider. The first is the communities in Jurisdiction A who are not benefiting from the placebo of mass incarceration policies, although they may indeed enjoy psychological or therapeutic benefit.²⁴⁰ The second group is the millions of people who are suffering because of the harm and social costs of mass incarceration. These people did not consent to be a part of this grand thought experiment but have been caught in the middle of this ethical dilemma and human rights crisis.²⁴¹

The moral supremacy of the deontological and abolitionist arguments cannot be denied, but the uncomfortable truth is that utilitarians hold the pragmatic high ground.²⁴² Experimentation is at the heart of placebo and punishment policy alike. It is a double-edged sword that can bring about tremendous positive change, but it can also be captured by political incentives and penal populism. Consequently, the only utilitarian justification for continuing to use mass incarceration as a placebo would be to wean ourselves off from relying on placebos and begin rigorous analysis that will lead to improved public safety.

The ethics of using placebos in experimentation provides a unique starting point to discuss potential ways to mitigate the harm and justify the practices of placebos in the punishment context. Just as the government plays a key regulatory role in the administration of new drugs, this process might have some value in approving the next generation of punishment policies. Drugs often undergo a rigorous regulatory process that can last years before they are approved for doctors to prescribe, and placebo human trials are merely a step along that long journey.²⁴³

What this experimentation ethic requires is an FDA-like project that gives experts the space, the time, and the resources to approve, reject, or reform new and existing punishment policies. This FDA model would hand over full regulatory power to expert agencies and commissions. The legislature would no longer be in control because they simply do not have the expertise required and the stakes are too high to trust generalists.²⁴⁴ Just as Congress does not have much say in whether Drug A is both efficacious and safe for the market, legislatures would not have much say in whether getting rid of mandatory minimums, for example, would be efficacious and safe for the public.

As Barkow has suggested, there is already an infrastructure of expertise that can be readily tapped for such a transition. Sentencing commissions exist across the country that serve advisory roles to federal and state legislatures; although quite different from the FDA, they are comprised of expert scholars, advocates, and judges.²⁴⁵ While existing commissions have succeeded in recommending reforms and pushing back against penal populist policies, most have little teeth and their policy recommendations can often be ignored if they do not line up with the political incentives ruling their respective legislatures.²⁴⁶ Consequently, there would only need to be a formal legislative delegation and other adjustments to ensure these existing commissions have the necessary power to assume regulatory control over punishment policies.

One of the flaws of this FDA punishment model, however, is that the marketplace of regulating food and drugs is vastly different from regulating criminal acts in material ways. Whereas the FDA is a government agency that regulates private companies, a government agency that regulates other government actors usually takes the form of checks and balances. Since any proposed expert agency of punishment policies would be the progenitor of new policies, there would need to be sufficient checks against these new policies. Thus, if the expert agency adopts a new punishment regulation, the courts would still have the ability to strike it down as a check against the administrative state. Executives would still have the discretion to fire appointees, and the legislature would still be able to impeach such appointees.

This author has also contemplated that such punishment agencies could work in the federalism context, whereby the federal government agency serves as a watchdog over state punishment practices.²⁴⁷ Since 94% of criminal convictions²⁴⁸ and around 83% of incarcerated persons are handled by the states,²⁴⁹ most of the harm of placebo punishment policies are being felt at the state level. This federalism approach would borrow less from the FDA and more from the DOJ’s Civil Rights Division and the preclearance process of the Voting Rights Act.²⁵⁰ The preclearance process gave the federal government power to approve or reject new voting laws before they went into effect in certain states and counties to mitigate voting discrimination. The DOJ could indeed perform similar watchdog functions over various state punishment policies, and it might also have the benefit of creating some uniform accountability for both carceral and non-carceral punishments.²⁵¹ Admittedly, this structure would lack the power of the Voting Rights Act to prevent policies from going into effect. However, the DOJ and the federal government could more easily tie federal funding to this punishment approval process. For instance, if Louisiana wanted to increase penalties for drug offenses, they would maintain their sovereignty to do so. However, if the DOJ disapproved of this policy, it would be well within its purview to rescind federal funding for criminal legal projects,²⁵² and it could partner with other federal agencies to discuss their funding options as well.²⁵³

Handing over so much power to agencies filled with experts is not without its own concerns. Many scholars have expressed reservations about shifting power to experts in ways that might not be productive. Megan Stevenson’s study of more than two dozen meta-analyses led her to conclude that there is very little evidence from randomized control trials that many of the rehabilitation reformist interventions are effective at reducing recidivism or first-time offenses.²⁵⁴ The reliance on “evidence-based reform” might also give people a false sense of efficacy, reliability, and neutrality when discretion is still being used to draw lines, make judgment calls, and enshrine perverse incentives.²⁵⁵ Even experts can be influenced by their own moral and social views on punishment, and these views will often impact which experts are selected by the government to serve on these commissions. Therefore, there is potential for political capture of such agencies, as we see in the administrative state quite commonly.²⁵⁶ Agencies and commissions can be vulnerable to corporate capture, especially if lobbyists or other interests seek to influence commission members by employing them as consultants outside of their government duties. These are important considerations that require careful institutional design choices to mitigate such influences.

The takeaway of this Section is that a placebo punishment methodology enlightens the discussion of ethical experimentation and requires careful consideration of the utilitarian benefits and social costs. Using the mass incarceration placebo is unfortunately a necessary evil to pave a way forward, as justified by creating better consequentialist punishment outcomes. And further, allowing experts into the experimentation process is key to mitigate the past generation of perverse political incentives and penal populism. Allowing experts to conduct these experiments and approve the experiments of others would begin to yield the type of results to turn back the tide of mass incarceration. And while the data, studies, and policies of these experts may not alone be enough to change the social and cultural sentiments towards punishment,²⁵⁷ they can get much closer to punishment policies that effectively make us safer.

D. Placebo Economics

The placebo ethics of economic incentives holds further insight for the marketplace of placebo punishment policies. Ethicists criticize the use of placebos on economic grounds because it requires a patient to pay for a product that is physiologically inert. Placebos are tantamount to snake oil, even if sold or prescribed benevolently by the doctor acting as a salesman. The same ethical criticism should be made of placebo punishment policies and the larger economic debates that have the potential of influencing the very social and cultural meanings necessary to change public views on punishment.

The economic aspect of placebos may be important to shaping the patient’s subjective expectations and perceptions of the treatment. When a doctor prescribes a treatment or drug, this usually necessitates some pecuniary commitment of the patient to go the pharmacy and buy the drug or over-the-counter product.²⁵⁸ Imposing this economic cost may be necessary to the placebo’s impact because a patient would perceive an effective drug to require such out-of-pocket expense. This imposes costs on patients and insurance companies for products that carry little physical benefit but nevertheless provide some psychological and therapeutic benefit. Thus, patients and insurance companies are required to pay for products that are designed to deceive the patient into feeling better, as opposed to a physiologically efficacious treatment.²⁵⁹

In addition to the doctor-patient relationship, there is also a robust, direct-to-consumer placebo market of supplements, vitamins, and other remedies that claim to contribute to health benefits without any scientific proof.²⁶⁰ In fact, many of these so-called health supplements have been scientifically debunked, and yet consumers still flock to this marketplace—which sells substances that are not regulated by the FDA—to the tune of $40 billion every year.²⁶¹ Thus, these alternative treatments that have no scientific backing still make up a “significant segment of the healthcare industry.”²⁶²

Economic incentives also complicate the ethics of the medical industry. Doctors can be incentivized by personal economic gain, receiving stipends, cash bonuses, expensive trips and a whole host of other extra benefits based on their prescriptions. The underworld of pharmaceutical companies paying doctors to increase prescriptions has been uncovered to great public outcry.²⁶³ Even the government could not turn a blind eye to these perverse medical incentives and has been active in investigating and increasing regulations to mitigate these harms.²⁶⁴ This is relevant to the current discussion because, according to some experts, these personal incentives exist for doctors to prescribe drugs they know to be placebos.²⁶⁵

Applying these economic ethical concerns to the placebo punishment sphere, taxpayers should not be expected to pay the expensive costs for a punishment regime that is both inefficient and not efficacious. The prison system itself costs taxpayers approximately $80 billion annually according to conservative estimates and as much as $182 billion annually according to more liberal estimates.²⁶⁶ During the height of the mass incarceration era, state and local expenditures for the incarceration system rose by 946%, outpacing the very social programs better equipped to handle rising crime rates such as education, healthcare, and public welfare.²⁶⁷ And this is just a fraction of the $305 billion that conservatively estimates the total cost of the criminal legal system.²⁶⁸ This is quite the sum for government services that do not effectively deliver public safety benefits.

Politicians are also incentivized to support the punishment marketplace through the PIC. The broader PIC supports a multi-billion-dollar industry that includes a host of private and non-profit companies outside of the prisons’ walls.²⁶⁹ Private prisons alone are estimated to bring in $4 billion in profit,²⁷⁰ and incarcerated people who are forced to work at slave wages produce $11 billion in goods and services.²⁷¹ This is a small sample of the larger PIC, which supports the carceral industry by providing soap, clothes, food, surveillance, weapons, construction, and other goods and services that uphold the status quo. Although not nearly as scandalous as private companies making direct payments to doctors, lobby organizations and unions for the PIC contribute to political campaigns in ways that incentivize politicians, benefit private companies, and incentivize locking up and surveilling more and more people.²⁷²

Placebo ethics questions the economic practice of requiring people to pay for a product that is known to be ineffective and scrutinizes the perverse economic incentives at play. But looking forward, a focus on inefficient or ineffective economic practices can also play a role in changing the very social behavior and cultural understandings necessary for a new punishment paradigm.

The last generational shift in punishment was based on this very placebo ethic, but in slightly different terms. The dominant rehabilitative model was believed by scholars and politicians of the time to be far too ineffective for the relative economic costs. This logic took hold in the 1960s and 1970s, when the federal government spent billions of dollars to improve existing prisons and jails to lower recidivism,²⁷³ all just to see the crime rate rise.²⁷⁴ Politicians and taxpayers alike lost their logical patience to see the long-term positive impacts that rehabilitative models were having²⁷⁵ and instead turned to fear and emotion by favoring harsher sentencing. This paved the way for the implementation of mass incarceration policies during Reagan’s 1980s²⁷⁶ and Clinton’s 1990s.²⁷⁷ It took nearly twenty years of frustrations (1960s and 70s) to boil into a generational shift in policies over the next twenty years (1980s and 1990s).

We have started to see this same multi-decade cycle reflected in the current twenty year frustration period in the 2000s and 2010s. Fiscal conservatives and libertarians—who might have traditionally been the staunchest supporters of law-and-order politics—have started to ring the alarm bells that mass incarceration policies are simply too expensive for the relatively small social benefits.²⁷⁸ Perhaps the surest bet in decarceration over the past fifteen years is that economic downturn has yielded the most significant reforms. The smart-on-crime movement, which boasted more efficient and effective punishment policies, started in the early 2010s as it was spurred by state budgetary cost concerns coming out of the Great Recession of 2008.²⁷⁹ The country’s recent battle with COVID-19 also sparked increased debate about penal spending when states struggled to balance their budgets given the tremendous negative economic impacts of the pandemic.²⁸⁰

Economics were not the entire story, but they were part of the social and cultural shift of these times. Social and cultural expectations shape a patient’s positive and negative experience with a placebo and so too are we seeing the social and cultural expectations of economics impact how Americans are starting to negatively view the placebo of mass incarceration. There are few things more “American” from a cultural standpoint than demanding to get your money’s worth. And as perceptions of the costs of mass incarceration change,²⁸¹ so too will economics impact our cultural rejection of placebo punishment policies.

The placebo punishment framework and the broader health literature it incorporates not only help us think about these ethical economic problems but also gesture towards unique health-related solutions that can inform punishment reform. One approach would be to regulate the market with economic tools like a “sin tax” of some type. This author and others have previously proposed various methods for increasing the economic costs of incarceration on government actors to disincentivize them from over relying on carceral punishment tools,²⁸² such as imposing a tax on local prosecutors when charging and prosecuting a person and suggesting prison time.²⁸³ These types of sin taxes have been applied across the public health sphere from cigarettes to sugary drinks.²⁸⁴ Passing along these costs onto constituents, such as raising taxes for the express purpose of maintaining the prison population, would also likely change the social and cultural expectations of the mass incarceration placebo and could result in further public outcry for reform.²⁸⁵

Another economic insight from the placebo punishment framework is its focus on effective treatments and diagnoses alike. The crime-as-disease framing diagnoses crime as our social and cultural norms have defined it: as the thing to be focused upon for economic investment in the punishment remedy. However, drug addiction, gun violence, mental health and educational problems, and disparities in economic opportunity are the more accurate diagnoses. It is understandable that if a doctor diagnoses cancer, the medical priority is to immediately address that life-threatening problem; but the secondary treatment should be to do everything possible to prevent the cancer from coming back as part of a preventative medical treatment regimen. While this does not condone the use of placebo punishment policies in the past, we must now focus on economically investing in the right treatments for the right diagnoses. This calls for a multivariate investment in punishment policy into education, health policy, job training, and the “all of the above” approach that reformists have advocated for since the beginning of the mass incarceration era.²⁸⁶ Further, the broader we understand this preventative medicine approach so as to avoid future need for placebos, the more we can rely on a multi-agency funding campaign involving the likes of HHS, HUD, DOE, and a host of other governmental grant and funding partners that can tackle the crime problem.²⁸⁷

These economic interventions are not without critics, but dismantling the placebo punishment policies of mass incarceration is designed to take into account more than just dollars and cents. For instance, Marie Gottschalk has levied a series of criticisms against exclusively focusing on economic cost-benefit analysis to solve the mass incarceration crisis in part because it fails to capture the human rights deprivations of incarceration and only serves to reinforce the neoliberal zeal to further cut costs of the welfare state.²⁸⁸ Just like in the 1970s and 1980s, economics will not be the only factor that leads to a generational shift in punishment policy.²⁸⁹ Neither do these tools of economic accountability and reinvestment seek to hollow out the welfare state; rather, they propose the opposite. By focusing on the right diagnoses, we transform the social and cultural tradition from viewing human offenders as a cancer to viewing the institutional failures of the social welfare state as the disease to be remedied with robust reinvestment.

This Section’s focus on highlighting the ethical concerns of placebo economics adds another analytical tool to the placebo methodology of punishment. It is hard to justify mass incarcerations’ placebo focus on controlling public perception of crime based on the actual price tag for such ineffective policies. But this Section also highlights that economics play an important role in shaping public perception and expectation of punishment policies. The more something costs, the more we expect as a society and the more we resent the waste when the policies do not work. Economics can indeed shape the social and cultural perceptions of our punishment future by imposing interventions that increase costs or that refocus policy on the underlying causes of the crime/disease we seek to prevent.

IV. Beyond Punishment and Placebos

The comparisons, synergies, and differences between the medical use of placebos and legal punishment policies offer numerous insights into how we might think about and approach the biggest problems of mass incarceration, but these placebo analyses also offer several insights across criminal legal policy and society as a whole. This Part zooms out to briefly consider these broader questions and applications. Section IV.A illustrates that a placebo methodology can be valuable in legal analysis whenever there is need to compare the effectiveness of a policy with its value on changing the public’s perception of the problem it was purported to address. Section IV.B considers a few alternatives to the placebo methodology that utilize some of the same truths about punishment but according to different, but tangential, socio-medical frameworks.

A. Beyond Punishment

The placebo methodology need not be limited to punishment. Learning from medical ethics and public health literatures, placebo analyses offer an underutilized methodology that impacts theoretical framing and practical problem solving. This Section briefly considers this bigger picture by developing the placebo methodology around a few case studies to spark further scholarly inquiry.

First, it is interesting to consider how much the future use of placebos in the legal literature would contribute to other well-established methodologies. For example, political economists have developed an entire field studying the symbolic value of law and policy.²⁹⁰ Analyzing the law as a symbol implicates some of the same themes as the placebo methodology because they both consider how law impacts both perceptions of society and perceptions of the lawmakers enacting such symbols. Thus, scholars have studied the theoretical symbolisms of tax policy, the environment, health law, and numerous other fields.²⁹¹ One of the key differences is that this symbolic methodology generally focuses on how lawmakers avoid political accountability through passing symbolic laws; in other words, they can gain the benefit of public perception by appearing to do something meaningful through symbolic legislation, and need not be answerable for the law’s actual effects.²⁹² The focus on political accountability and shirking can certainly add to the placebo methodology, but the methodology focuses more on the effectiveness versus perception value proposition along with the ethical considerations of policy justifications.

Behavioral economics walks a parallel path by studying how social contexts and heuristics impact decision-making.²⁹³ This discipline adds tremendous explanatory value to a host of legal and policy decisions by studying the extent to which social norms can contribute to irrational decisions.²⁹⁴ Feel-enomics—or how people feel about the economy—is often more important than the actual economy.²⁹⁵ For example, negotiators will seek to punish their adversary when they perceive unfair treatment, which can actually lead to a negative result for both parties.²⁹⁶ Thus, a person’s perception of what is socially right and wrong behavior can impact their decision-making, which can lead to net-negative results for all.²⁹⁷ Heuristics have also been an important contribution of behavioral economics, which study the mental shortcuts we use when making snap decisions about people, the natural world, and social conventions that can lead to erroneous assumptions.²⁹⁸ This spans the gamut of making snap judgements about a person’s profession, their guilt in committing some offense, or other attributes based on stereotypes that function as judgmental heuristics.²⁹⁹ Heuristics are often informed by learning social cues and stereotypes about different classes of people, and this social context impacts a person’s perception of others at the danger of making a bad and inaccurate judgment call.³⁰⁰

The similarities between behavioral economics and the placebo methodology of criminal punishment should not go unnoticed. Behavioral economics focuses on social contexts and operationalizes the importance of a decisionmaker’s social perception. The placebo methodology benefits from this literature, but it prioritizes public perception as opposed to that of the decisionmaker. Heuristic judgements are often a theme in criminal legal scholarship addressing racial profiling and sentencing disparities, and this relates to how the crime-as-disease framing characterizes certain people and communities as being part of the crime problem.

Within this broader context of using these themes about perception and accountability in legal scholarship, placebos have found some limited value. Aviram has already used slightly different placebo methods to analyze travel regulations.³⁰¹ Malani has also convincingly applied placebo methods to analyze pharmaceutical regulations, health law, consumer protection, and torts.³⁰² Others have invoked the language and terminology of placebos but often as a hollow comparison meant to denote an ineffective legal policy in fields such as financial regulation,³⁰³ labor and employment,³⁰⁴ and environmental policy.³⁰⁵

Narrowing this to the criminal legal context, there are myriad values of the placebo methodology outside of studying mass incarceration. This could extend to other punishment policies such as whether placebo policies might factor into the state of e-carceration and bail reform. In the same way that mass incarceration policies have been used to control public fear and perception, it might be possible that similar goals are imbedded in the pre-conviction bail context and even the post-conviction surveillance context.³⁰⁶ Courts are willing and ready to add to this literature, often looking for unique ways to characterize problems as placebos in the punishment regime.³⁰⁷

Policing is also a ripe field that might benefit from placebo methodologies. Deterrence advocates are often at odds with critical race theorists and abolitionists in this debate, disagreeing about the effectiveness of police presence on catching offenders and deterring future crime. While there are convincing criminological studies suggesting that more police officers can deter crime if it increases an offender’s likelihood of getting caught,³⁰⁸ many Critical Race Theory (“CRT”) scholars argue that policing comes at too heavy a cost for poor and minority communities subjected to surveillance, excessive use of force, and police killings.³⁰⁹ A placebo analysis of policing has some precedent, since the American Public Health Association adopted a policy recognizing policing as a public health issue.³¹⁰ This recharacterization paves a smoother path for a placebo analysis that may focus on how policing impacts perceptions of public safety versus its actual efficacy on public safety; this would, of course, include the nuances of racial politics such as considerations of who the police are meant to keep safe, who benefits from feeling more safe, and who suffers the social costs and consequences of police presence.

This Section is not meant to be exhaustive, but it illustrates the broader appeal and applicability of deeper examinations and use of placebo methodologies in legal scholarship. The ongoing discussions that pit the objective effectiveness of policies against how the police impact public perception of the problem will continue to shape future decision-making. By adding socio-medical ethics and policy solutions to these discussions, multiple literatures can benefit from the placebo methodology as deeper considerations of the law’s value and continued role in society.

B. Beyond Placebos

As this Article has highlighted, comparing punishment policies to a medical placebo treatment is a worthwhile methodology but one that is not a perfect analogy. This Section considers a number of alternative socio-medical framings that also have value for future consideration. Each of these alternative framings expand on different aspects of our social understanding of punishment in ways that highlight latent truths that might impact how we should think about future policy.

Perhaps the most analogous framing is thinking about punishment as a penicillin treatment. The core value of this alternative is that it highlights the changing efficacy of punishment policies across time. Penicillin is an antibacterial treatment that has significant diminishing returns. As penicillin continues to be administered to a patient and a broader population, bacteria will begin forming new strains that are resistant to this treatment.³¹¹ These diminishing returns are shared among many antibacterial treatments, which is why many doctors do not prescribe the treatment long enough for any leftover bacteria to develop immunity. These so-called “superbugs” are incredibly dangerous because they can withstand antibacterial treatments, leaving very few options to treat them otherwise.³¹²

As discussed in Section II.B, there are indeed diminishing returns in the mass incarceration context. Many scholars believe that harsh incarceration policies may have been necessary in a time when violent crime was high but would now argue that these same policies are ineffective and potentially damaging to society in the modern age.³¹³ Perhaps, like penicillin, crime and those who commit it have adapted. People who are sent to prison have an incredibly high recidivism rate in part because prison has been known to increase offenders’ access to criminal networks.³¹⁴ Therefore, the very people we are trying to quarantine from society in such a medical model of punishment are the very people the criminal justice system is responsible for turning into so-called “super predators” as many in the 1990s believed.³¹⁵

If this penicillin model has any practical value, the most important thing we can learn is that we must transition away from the penicillin of prisons. Perhaps prisons served their role in the 1980s and 1990s when violent crime was high. But like penicillin, this “treatment” is not effective over long periods of time; in fact, it is actually dangerous if used long-term. Instead, using prisons effectively to mitigate the growth of crime is important, and now we are in an era of transitioning to second-order treatments meant to ensure the infection or social disease of crime can be managed long-term.

Another alternative framing in the same vein is to consider punishment as an addictive drug. This controversial concept results in quite a different set of considerations in which there is very little benefit and almost entirely unjustifiable costs. For many abolitionists and progressive reformers, mass incarceration might fit within this description. It is a social institution to which society seems to be addicted. Every time we attempt to make reforms, the same system reemerges, and we go back to our carceral fix. Worst yet, just like a drug, many in our society realize how bad mass incarceration is.³¹⁶ Yet we cannot help ourselves. We are addicted to the retributive ideas, the incapacitation logic, and the dopamine feelings of safety. If we honor the alliterative patterns of this Article, we might portray this framing of punishment as a type of poison. The cost of continuing to ingest such a substance is a slow assured death given the harm it is causing the body politic.

This framing is a bit more radical, but it may nevertheless have some kernel of truth to it. How then does this make us think about punishment and mass incarceration? Like an illicit substance, we must break our addiction from these policies. But not even an abolitionist would support such a cold-turkey approach in abolishing prisons. Prison abolition is a generational project that will take years and even decades to accomplish,³¹⁷ which is comparable to a long-term weaning off from the substance. Perhaps we can make annual benchmarks and create a five-, ten-, and twenty-year plan to gradually lower the incarceration rate in ways that will not result in millions of offenders being released back into society all at once.

But even the most stalwart survivor who has dedicated themselves to sobriety knows that relapse is only one moment of weakness away. Whenever crime ticks up, even infinitesimally, we see people rallying for a return to tough-on-crime principles.³¹⁸ It only takes the slightest trigger of crime—whether it be a summer crime wave or even just one particularly egregious crime that gets news coverage—to make society crave retributivism, incapacitation, and the feeling of safety and justice that comes with locking people up and throwing away the key.

Another fascinating socio-medical framework to consider is malingering by proxy, which describes a psychological condition in which a person tells others that their own child or loved one is sick in order to garner sympathy, attention, and financial gain.³¹⁹ This is most prevalent among parents, who will tell others in the community that their child is sick.³²⁰ The parent will make their otherwise healthy child appear sick by mimicking symptoms, such as shaving the child’s head to mimic hair loss suffered by chemotherapy, or by forcing the child to wear bandages or a face mask to mimic injuries or an immune-compromised disorder.³²¹ By mistreating their child in this way, parents often hope to garner sympathy from the community, to be seen as a hero in their own right for taking care of their sick child, and even to ask for donations to help them pay for fake medical treatments.

Malingering is an interesting framing when conceptualizing the role of politicians, law enforcement, prosecutors, and others who are delivering mass incarceration services to the public. Under the placebo framework, these actors were likened to doctors prescribing treatment; but unlike doctors, these actors were deceiving the public for political gain. Under a malingering framing, these actors would similarly be deceiving the public for their own gain by trying to convince the public of the fake malady of criminal behavior in a way that garners public support for their cause. Like the placebo methodology, this framing is rough around the edges and requires the reader not to lose the forest for the trees. Instead of looking for exact 1:1 comparisons, the big picture takeaway is that this framing highlights the perverse incentives of politicians and decisionmakers in advocating for and convincing the public about the merits of harsh punishment policies.

The medical literature is still exploring the level and types of treatment for malingerers, having to confront whether this condition should be considered a psychological disorder.³²² Regardless of this medical question, malingering by proxy is an incredibly harmful antisocial behavior that harms both the communities subject to the false information and the proxy, a usually-vulnerable party who must pretend to be sick. The takeaway from such a framing would be to hold our lawmakers, police officers, prosecutors, and all others who are using deceptive means (whether purposely, knowledgeably, recklessly, or negligently)³²³ accountable for their mischaracterizations that have damaged the community and the public trust.

This Section is meant to be creative, thought-provoking, and controversial. In the same way that the placebo methodology challenges the status quo and our assumptions about mass incarceration, the framings of penicillin, illicit drugs, and malingering offer brief vignettes of alternative ways to highlight different aspects of the modern punishment crisis and how to resolve it. This Section illustrates the value of socio-medical tools to break free of stale approaches to crime and punishment, and I hope it sparks new discussion and theories of the criminal legal system to push the boundaries of what is possible toward reformist interventions.

Conclusion

The medical study of placebos offers a unique approach to theorizing criminal punishment. Because crime is often considered a social disease and punishment is seen as its treatment, mass incarceration policies truly deliver a placebo impact. The placebo methodology captures the importance of the social and cultural context required for mass incarceration policies to control the public’s fear and perception of crime, without having commensurate value on impacting public safety.

This approach offers multiple innovations that raise interesting questions about the future of mass incarceration policies. The interdisciplinary insights from medical ethics provide a stark picture of the sheer lack of justifications for using placebo policy in the punishment context. The negative social costs and harm, malevolent use of deception, the pervasive use of experimentation toward harsher punishment, and the perverse economic incentives require us to abandon placebo policies and move toward punishment practices that can actually make us safer by focusing on what causes the crime/cancer problem through preventative medical and social means.

This is the double-edged sword of the placebo methodology. It reveals the latent problems of punishment policies, but it also provides insight into avoiding the same placebos in the coming generation. This Article hopes to contribute to the larger conversation, ensuring that the sword is slicing in the right direction toward a safer, fairer, and more justified punishment future.

98 S. Cal. L. Rev. 513

Download

Sheldon A. Evans*

*Professor of Law, Washington University School of Law. I thank Erin Blondel, Vincent Chiao, Chad Flanders, Daniel Fryer, Trevor Gardner II, Fareed Nassor Hayat, Ben Levin, Angelo Petrigh, Sandra Mason, Ion Meyn, Mark Pickering, Ji Seon Song, and Kit Wellman for helpful thoughts and comments. In addition, I extend my thanks to the community of scholars that provided valuable comments and feedback at the University of North Carolina School of Law, the John Mercer Langston Writing Workshop, CrimFest!, the Decarceration Works-in-Progress Workshop, and the St. Louis Criminal Law Theory Workshop. I also thank Sydney Everett for her excellent research assistance.

Decriminalizing Condemnable Conduct: A Miscalculation of Societal Costs and Benefits

May 2025May 2025Paul H. Robinson* & Jeffrey Seaman†

Recent developments have seen a trend toward de facto decriminalization of conduct that the community continues to see as criminally condemnable. This includes effectively decriminalizing certain kinds of conduct, such as lower-level theft, immigration offenses, illicit drug use, or domestic violence without serious physical injury, as well as criminal conduct by certain groups, such as rioters or statue vandals motivated by a cause with which officials sympathize. Such de facto decriminalization can come about in a variety of ways, including policy decisions by the local prosecutor or city council not to arrest or prosecute or by state legislation or referendum that (often unintentionally) decriminalizes the conduct.

The supporters of such decriminalization are typically acting in good faith to produce what they see as a better society, commonly driven by one or more of four motivations: (1) a belief that medical, social services, or mental health professionals are a better response to crime than the criminal justice system—what might be called the anti-criminal justice motivation; (2) a desire to reduce sanctions that would otherwise be imposed upon a group seen as oppressed—what might be called the social justice motivation; (3) a belief that a moral elite are in a better position than the rest of society to decide what is and is not criminally condemnable—what might be called the moral superiority motivation; and (4) a belief that locales rather than the broader jurisdiction should decide what is criminalized, even though the state or federal constitution provides otherwise—what might be called the local superiority motivation.

This Article argues that these justifications for effectively decriminalizing condemnable conduct are questionable. Even if they did offer some societal benefit, any such benefit would be dramatically outweighed by their societal costs: the immediate costs of lost deterrence and lost ability to incapacitate repeat offenders, as well as the even more damaging long-term costs of lost moral credibility and legitimacy, a loss which can only serve to reduce compliance, cooperation, and acquiescence and instead increases resistance, subversion, and vigilantism in the community.

We point out, however, that the same principles apply in reverse situations. Just as conduct seen by the community as criminally condemnable ought not be decriminalized, so too conduct seen as blameless or of reduced blameworthiness ought to be entitled to a defense or mitigation, which current criminal law sometimes fails to do. Similarly, just as conduct seen as condemnable ought to be criminalized, so too conduct that is no longer seen as condemnable ought to be formally decriminalized, which, again, current criminal law sometimes fails to do. Ensuring that the criminal law tracks society’s criminalization-decriminalization judgements should not be controversial in a democratic society, and this Article argues for a fair and consistent application of that principle to all areas of criminal law.

Introduction

Increasingly, what the public views as criminal does not coincide with what receives punishment. This Article examines the understudied phenomenon of decriminalizing conduct still viewed as criminally condemnable by society and argues for bringing all areas of criminal law back in line with the community’s criminalization judgements.

Criminal law marks itself off from other areas of law by dealing with conduct that is seen as sufficiently condemnable to deserve the punishment and stigma of criminal conviction. Society’s criminalization standard is nuanced and leaves a great deal of morally condemnable conduct non-criminalized. Cutting in line to buy concert tickets is likely to be seen as condemnable, especially by those who camped overnight, but probably falls short of the condemnation required for criminalization. Some behavior may be seen as quite immoral—betraying a confidence, cheating in a relationship, etc.—but it is not considered criminal since it falls within a perceived realm of personal liberty to make even wrong choices.¹

Society’s criminalization judgements can change over time, downgrading criminal behavior to immoral-but-legal or even fully acceptable conduct. For example, adultery was once seen as sufficiently condemnable to deserve criminalization but is now widely seen as immoral behavior falling within a sphere of personal liberty, and thus a matter for purely social disapproval instead of criminal condemnation.² Some changes can be even more drastic and quick, as with same-sex intercourse going from criminalized to same-sex marriage being widely supported.³

The point along the continuum of condemnation where social disapproval turns into criminalization will have its close cases, especially since people’s understanding of personal liberty differs, but in the vast majority of cases the difference will be clear—and important. Criminally condemnable conduct calls for stigmatization, denunciation, and punishment that serves to announce and reinforce the societal norm, with the hope that people will internalize the norm. Marking out conduct as noncriminal is equally important. Decriminalizing birth control or same-sex intercourse sent the important message that such conduct was a matter for personal liberty, protected from government interference,⁴ and any residual societal norm against the behavior was thus weakened and left open to replacement. Criminalization and decriminalization are two equally important tools of regulating societal behavior.

Conduct may be de jure decriminalized by the legislature, but it may also be de facto decriminalized by nonlegislative state or local officials who adopt and announce a policy that it will no longer be treated as criminal and subject to prosecution. Once such a de facto decriminalization policy is announced, it logically signals police to consider not making arrests for such conduct. As a legal matter—and as a constitutional and a democratic matter—the de jure and de facto forms of decriminalization are importantly different, but in the real world, they can have similar effects on victimizers, victims, and the rest of society.⁵

Part I of this Article documents the increasing instances in which conduct seen as criminally condemnable by the community is nonetheless decriminalized by official action or inaction. Part II details the variety of motivations and justifications given for such decriminalization, many of which readers may find reasonable or sympathetic. However, as Part III argues, we believe that those motivations and justifications, commonly aimed at creating a better society, do not fully account for the significant societal costs of decriminalizing conduct viewed as criminally condemnable by the community. The supporters of these policies no doubt believe that theirs is a path to a better society, but we argue that they fail to take account of both the immediate costs of lost deterrence and lost ability to incapacitate repeat offenders and the longer-term societal costs of lost moral credibility and legitimacy of the criminal justice system.

This Article is not an attack on the proponents of such decriminalization policies or an attempt to prevent all decriminalization. We argue that the principles developed in Part III need to be applied equally to the full range of reverse situations, in which improper criminalization and undeserved punishment may occur. Part IV argues that, just as criminal law ought not violate shared community views by failing to criminalize, it also has an obligation to recognize defenses and mitigations sufficiently broad to reflect community views, which it currently does not do. (Of course, there are limits to the appropriate breadth of excuse defenses, and those limits exclude recognizing ideologically driven defenses like a “rotten social background” defense,⁶ which has been supported by the same decriminalization motivations detailed in Part II, and which clearly conflicts with community views.) Additionally, there is always a need to make sure criminal punishments align with the community’s view of justice, but a public desire to reduce punishment for crime should not be mistaken as support for decriminalization. However, when community views on criminalization do change, the same cost-benefit analysis from Part III demands that criminal law decriminalize conduct that falls below the community’s condemnation threshold required for criminalization.

I. Decriminalizing Condemnable Conduct

Conduct seen as criminally condemnable by a clear majority of the community can be de facto decriminalized in a variety of ways. This Part examines those various mechanisms and gives a series of case examples. Section I.A discusses how local prosecutors can announce non-prosecution policies for conduct that is clearly criminal under state law, thus effectively decriminalizing that conduct by eliminating the chance of criminal liability. Similarly, local officials (such as a mayor or city council) may direct police to not make arrests for certain criminal conduct, thus achieving the same result. Section I.B examines how local authorities may de facto decriminalize conduct when committed by specific groups in what appears to be a selective decriminalization for ideological or political reasons. Because this kind of decriminalization only applies to certain kinds of offenders, it appears that local officials are essentially providing an ideological defense for crime. Section I.C discusses how a formal change in law by the legislature can have the practical effect of decriminalizing condemnable conduct, such as by reducing the conduct to an offense category for which arrest is not authorized and for which the only sanctions available are those used for civil or administrative violations (e.g., a ticket or citation). Section I.D answers an obvious question: How can decriminalization enacted by duly elected officials, or even voters, conflict with community judgements of what is criminal conduct?

A. Non-Prosecution Policies Rejecting a Legislative Criminalization Decision

De facto decriminalization occurs when local prosecutors create policies of non-prosecution for certain crimes. In addition to creating failures of justice in the eyes of the community, such policies also raise serious separation-of-powers concerns as such prosecutors are essentially arrogating legislative criminalization authority to themselves, in conflict with the state constitution’s grant of that authority to the state legislature. Local prosecutors obviously need to exercise discretion in individual cases because not all crimes and offenders are the same, and statutes cannot comprehensively cover all variations in circumstances. But to enact a policy of non-prosecution of a general category of conduct criminalized by the state legislature is to usurp the criminalization authority of the legislature in deciding what conduct is criminal in the first place.⁷

While there have been dozens of “progressive prosecutors” who have engaged in such policies,⁸ consider the examples of Larry Krasner in Philadelphia and Biba Biberaj in Loudoun County.

Larry Krasner: Decriminalizing Illegal Gun Carrying in Philadelphia.

In 2017, Larry Krasner was elected to the Philadelphia District Attorney’s office after winning the Democratic primary with 38% of the vote.⁹ He immediately set about pursuing a policy of non-prosecution whereby he used his discretion to not prosecute criminals guilty of certain crimes—including felonies. Perhaps his most controversial decision, from the community perspective, was his lax approach to illegal gun possession or carrying. With his overall goal being keeping offenders out of prison, Krasner chose to essentially decriminalize most felony gun possession.¹⁰ The extent of this policy was striking: “The Krasner office has withdrawn or dismissed 65 percent of gun charges this year [2021], up from 17 percent in 2015.”¹¹ After Krasner enacted these policies, guns flooded Philadelphia, and the number of arrests for gun crimes in Philadelphia tripled, but, predictably, the arrests failed to deter as the prosecution rate dropped by 85%.¹² In 2020, murders in Philadelphia increased by 40%, totaling 499—a higher number of homicides than in New York City, which has a population five times the size of Philadelphia.¹³ In 2021, Philadelphia recorded its highest homicide count ever, with 562 people murdered and another 1831 non-fatal shootings.¹⁴ But Krasner achieved his goal of keeping offenders out of jail by logging the lowest number of felony prosecutions in modern history, even as a controlled study found his non-prosecution policies were responsible for seventy-five extra murders a year.¹⁵

Krasner’s non-prosecution policies are so clearly damaging that he has drawn sharp criticism from his own party. After Krasner stated that Philadelphia did not have a “crisis of crime” or a “crisis of violence”¹⁶—statements that he ultimately walked back due to their obvious falsehood¹⁷—former Philadelphia mayor, Michael Nutter, who is Black, wrote that if Krasner “actually cared about [Black and Latino communities], he’d understand that the homicide crisis is what is plaguing us the most.”¹⁸ The crisis of violence escalated in 2022 to the point where the head of the Philadelphia City Council suggested reinstating stop-and-frisk policies to deter illegal gun carrying as Krasner’s non-prosecution policies failed to do so.¹⁹ Despite public concerns leading the Philadelphia Democratic Party to refuse to endorse him,²⁰ Krasner received large out-of-state campaign contributions helping him win the 2021 Democratic primary.²¹ After winning the Democratic primary, Krasner easily secured reelection in 2021,²² given that registered Democrats outnumber Republicans by seven to one in Philadelphia.²³ But Krasner’s de facto decriminalization of gun crimes led to his unprecedented impeachment by the Pennsylvania House of Representatives in 2022, though he is unlikely to ever face trial in the state senate, which requires a two-thirds vote of the chamber to convict.²⁴ Additionally, the Pennsylvania legislature passed a bill in 2023, signed by the state’s Democratic governor, to appoint a special prosecutor for crimes that occur on or near Philadelphia’s public transit system—a move made necessary by Krasner’s refusal to enforce the state criminal code.²⁵ Krasner’s unpopular decriminalization decisions may have also contributed to the victory of Cherelle Parker in the Philadelphia Democratic mayoral primary in May 2023. Parker promised to hire hundreds of extra police officers and crack down on the crime that flourished on Krasner’s watch.²⁶ Crime was the top concern among Democratic voters, with the progressive Krasner-aligned candidate, Helen Gym, finishing third with 21.5% of the primary vote.²⁷ While he might argue otherwise, it seems clear that Krasner has de facto decriminalized conduct that the community sees as criminally condemnable.

Buta Biberaj: Decriminalizing Domestic Violence in Loudoun County.

In 2019, Buta Biberaj was elected as the Commonwealth Attorney of Virginia’s Loudoun County.²⁸ In an effort to reduce prison populations and explore alternative approaches to crime, she pursued a de facto decriminalization policy for a large portion of domestic violence cases.²⁹ Biberaj explained: “[It is a] waste [of] resources investigating incidents that don’t involve serious injuries or risk of recurrence.”³⁰ Instead of supporting criminal justice intervention, Biberaj urged the community to treat domestic violence “as the health crisis it is.”³¹ Her decision contradicted research suggesting direct justice system intervention is more effective than alternative methods of dealing with domestic violence.³²

Biberaj’s policy soon produced tragic results. When a woman came to the emergency room after an assault by her husband, police arrested him for domestic assault, strangulation, and abduction.³³ However, the DA’s office simply released the man with a small unsecured bond.³⁴ The man returned home and later beat his wife to death with a hammer.³⁵ After the murder, Biberaj attempted to shift blame by claiming that the Sheriff’s office failed to offer enough details for the DA’s office to judge the seriousness of the situation—a claim vigorously denied by the Sheriff.³⁶ Biberaj’s record does not seem to suggest that victims’ safety was a priority in the DA’s office; if anything, she had a history of releasing dangerous offenders. When a high school girl was sexually assaulted by a classmate who had a history of sexual misconduct beginning in elementary school, Biberaj publicly announced that the offender would be kept at home on an ankle monitor, but instead privately worked to have the student transferred to a different high school.³⁷ Just weeks after school began again, the released offender pulled a fellow student into a bathroom, then choked and sexually assaulted her.³⁸

Of 735 domestic violence cases referred for prosecution, Biberaj dismissed 491, or 67%.³⁹ Kristen Umstattd, a Democratic member on the Loudoun County Board of Supervisors, sounded the alarm on Biberaj’s policies:

We’ve had two discussions as a board with her about our concerns. We had hoped she would take those seriously. She [Biberaj] has paid lip service to victim protection but if you talk to her privately, what she says about victims is they can be very untrustworthy and she needs to get them back—she needs to get the abuser back into the home so the victim doesn’t lose any family income. That is considered the worst possible approach to victim protection by people who work with victims all the time.⁴⁰

Umstattd and other local officials pointed to the dangerous consequences of Biberaj’s decriminalization policy: “[Biberaj] let go a man who had attempted to murder his wife. She let him go and he went back and he did murder his wife.”⁴¹ The Democratic chair of the county board, Phyllis Randall, also lamented: “Since [Biberaj] came in, the number of victims, and I mean domestic violence victims, who have reached out to us has been bloodcurdling.”⁴²

In 2023, Biberaj was challenged by Republican Robert Anderson, who attacked her record on domestic violence.⁴³ The charge resonated with residents who still viewed domestic violence as serious, criminally condemnable conduct. While Biberaj should have easily swept to reelection given that 2023 turned out to be an excellent year for Democrats in Virginia, there was one exception to the Democratic sweep: Buta Biberaj was ousted from office by Loudoun County voters angry with her non-prosecution policies.⁴⁴

The Progressive Prosecutor Movement and De Facto Decriminalization of Low-Level Crimes.

It is important to note that the above cases of de facto decriminalization are not rare, quirky exceptions, but representative examples of a broad national movement.⁴⁵ The de facto decriminalization of a variety of low-level crimes has been widespread with the success of the progressive prosecutor movement.⁴⁶ Since 2016, a wave of “progressive prosecutors” have taken power in many of America’s largest cities.⁴⁷ In 2022, at least half of America’s largest prosecutorial districts, covering some 72 million Americans, were run by progressive prosecutors, who presided over cities such as Los Angeles, Seattle, Chicago, Philadelphia, and New York City.⁴⁸ While some of those prosecutors won elections based on grassroots efforts, many others were boosted to victory in Democratic primaries by enormous campaign contributions from progressive megadonors like George Soros—helping to explain why so many progressive prosecutors can win elections despite advocating decriminalization policies that may contradict community views.⁴⁹ Many progressive prosecutors have little prosecutorial experience, but they share a commitment to instituting non-prosecution policies⁵⁰ that effectively decriminalize a range of criminal conduct that was democratically criminalized by the legislature and is still likely viewed as criminally condemnable by most of the community.⁵¹ Of course, not every self-described “progressive prosecutor” implements the same policies, and not all run counter to community views, but a commitment to non-prosecution policies, such as those enacted by Larry Krasner and Buta Biberaj, is one of the defining features of the movement.⁵²

B. Non-Prosecution of Selected Political or Ideological Groups

Decriminalization can sometimes be tied to offenders who belong to select political or ideological groups rather than specific offenses. For example, choosing not to prosecute rioters where the authorities share the perceived political or ideological beliefs of the rioters effectively decriminalizes conduct committed with that political or ideological motivation. Consider as an example the treatment of rioters during the protests following George Floyd’s death in 2020.

Non-Prosecution of Rioters in Portland.

After the death of George Floyd in May 2020, America saw waves of mostly peaceful protests by day turn into mostly violent riots by night. Portland, Oregon suffered more than 100 nights of riots starting in May 2020. The riots caused tens of millions of dollars in damage and terrorized the residents of Portland, who suffered an enormous decline in quality of life.⁵³ Mayor Ted Wheeler surveyed the damage after a night of rioting and reported: “I’ve never seen anything of this magnitude. The damage here is widespread. It’s not localized, it’s all throughout the downtown area.”⁵⁴

But the nightly destruction had just begun. On June 25, the fifty-seventh consecutive night of riots, Portland’s North Police Precinct building became a target of rioters. As the people inside the building tried to escape, attempts were made to keep them trapped in the building as efforts were made to burn down the structure. Authorities tried to extinguish the flames, but a crowd of around 300 continued to advance, throwing rocks and bottles.⁵⁵ Rioters came prepared with gas masks, shields, and weapons.⁵⁶ On April 14, 2021, almost a year after the riots began, the Portland Police Association building’s main entrance was set ablaze.⁵⁷ Rioters threw rocks and bottles at the officers. Other rioters stayed busy looting local businesses.⁵⁸ The unrest was unprecedented in the city, with nothing off-limits to the rioters. Private businesses were looted and burned; even public parks were vandalized. In the mass disturbances, Portland police only arrested one thousand or so individuals with enough specific evidence of a specific offense to support prosecution. Many of these were arrested on more than one occasion.⁵⁹

In response to this, District Attorney Mike Schmidt used his authority to decline to prosecute most (91%) of these especially egregious offenses.⁶⁰ He issued a written statement explaining his decision, in which he essentially applauded the arrestees for “tak[ing] action to bring about immediate change”:

Since late May, we have seen nightly demonstrations where people take to the streets to express their collective grief, anger and frustration over the senseless murder of George Floyd, and the countless other abuses People of Color have endured throughout history. “As prosecutors, we acknowledge the depth of emotion that motivates these demonstrations and support those who are civically engaged through peaceful protesting. We will undermine public safety, not promote it, if we do not take action to bring about immediate change.”⁶¹

Schmidt’s decision not to prosecute almost any rioters who did such damage to downtown Portland for months would likely have been unfathomable had the riots been perceived as supporting a different political cause. Schmidt’s political decriminalization of the rioting likely contributed to a steep drop in his approval ratings. A 2023 poll found that just 11% of respondents thought Schmidt was “doing a ‘good job.’ ”⁶² In May 2024, Schmidt lost reelection to one of his own deputies, Nathan Vasquez, who campaigned on a promise to take “lawless behavior seriously and enforce our laws.”⁶³ It seems clear that Schmidt’s decriminalization policies conflicted with the criminalization judgments of the community.

The National Picture.

Unfortunately, Schmidt was not alone in his decision to effectively decriminalize rioting during 2020. Despite the nationwide rioting costing up to $2 billion,⁶⁴ and leading to many deaths,⁶⁵ progressive prosecutors across the country chose not to prosecute the vast majority of arrested rioters for political reasons. In most large cities (with progressive prosecutors), 90–95% of riot-related cases were dropped.⁶⁶ Around the country, while buildings burned and stores were looted in front of rolling cameras, thousands of violent protestors who were known to authorities were not prosecuted. The justifications offered varied by jurisdiction, but political or ideological support for the rioters’ perceived cause was invoked by a number of prosecutors.⁶⁷ For example, in Chicago, Cook County State’s Attorney Kim Foxx stated that she would not prosecute the majority of the cases stemming from more than one thousand arrests during the George Floyd protests because “[o]ver the past month we have seen righteous anger, collective grief, action, and demands for justice.”⁶⁸ Foxx instructed her staff to act with a presumption of dismissal,⁶⁹ despite the fact that the lawless chaos that descended on Chicago had severely hurt minority communities.⁷⁰ The increasing dissatisfaction with Foxx’s handling of crime in Chicago likely contributed to her deciding in 2023 not to seek reelection.⁷¹ Unsurprisingly, many community members feel strongly that rioters and looters should face consequences no matter what triggered the initial unrest. The de facto decriminalization of looting in New York City in 2020 led to outrage among many storeowners, with one Bronx business leader describing her progressive DA’s decision to drop most looting charges: “Those numbers, to be honest with you, is [sic] disgusting. . . . I was in total shock that everything is being brushed off to the side.”⁷²

The de facto decriminalization of rioting has continued in some jurisdictions beyond 2020, with the commonality being expressions of sympathy for the rioters. In September 2023, looters and rioters took to the streets in Philadelphia after a judge dismissed murder charges against a police officer involved in a fatal shooting. Philadelphia’s DA, Larry Krasner, responded to the widespread looting by announcing a policy of prosecutorial restraint and urging sympathy with some of the offenders: “We’ll look carefully to see whether this is a one-off situation and they’re fundamentally law-abiding people.”⁷³ This statement may have puzzled listeners who might wonder how “fundamentally law-abiding people” could find themselves robbing an Apple Store, Rite Aid, and Dunkin’ Donuts, among the other Philadelphia stores looted in the unrest.⁷⁴

The de facto decriminalization of rioting in 2020 clearly ran counter to community views: a poll taken after Floyd’s death showed 72% of Americans viewed “violent protests and unrest” as an inappropriate response to police killings even as a clear majority also supported non-violent protests.⁷⁵ When it comes to politically motivated violence more broadly, less than 7% of Americans in one 2022 survey expressed support for such conduct, and “nearly all respondents support criminally charging suspects who commit acts of political violence.”⁷⁶

The de facto decriminalization of rioting also seems likely to have tarnished the justice system’s reputation, making those not on the political left more likely to view future prosecution decisions as having a political bias.⁷⁷ For example, many conservatives saw the aggressive prosecution of January 6 Capitol Hill rioters as constituting a clear double standard because of the previous one-sided decriminalization of rioting in 2020.⁷⁸ The events at the U.S. Capitol on January 6, 2021 led to federal prosecutors pursuing charges against more than 1,200 individuals.⁷⁹ A large number of these

were ordered held without bond. D.C. jail officials later determined that all Capitol detainees would be placed in so-called restrictive housing—a move billed as necessary to keep the defendants safe, as well as guards and other inmates. But that mean[t] 23-hour-a-day isolation for the accused, even before their trials beg[a]n.⁸⁰

Those who attacked federal buildings in Portland, Oregon were treated very differently. Of the 96 federal Portland cases which have been brought forward, 47 have been dropped, no one has faced trial, and those who have pled guilty have been largely sentenced to community service.⁸¹ No Portland protestor has been held in solitary confinement.⁸² While clearly an attack on the U.S. Capitol is more serious than attacks on federal buildings in Portland, the disparity in punishment is still sufficiently striking as to raise even non-partisan eyebrows. Most convicted January 6 defendants received prison or jail sentences, with some being sentenced up to 20 years,⁸³ while most of those who burned federal buildings and attacked federal officers in Portland had their cases quietly dropped.⁸⁴ Indeed, this disparity was noted by Judge Trevor McFadden in sentencing January 6 rioters when he openly questioned why prosecutors were demanding stringent sentences for January 6 defendants when they had seemed unconcerned with addressing widespread rioting in D.C. in 2020: “I think the U.S. attorney would have more credibility if it was even-handed in its concern about riots and mobs in this city.”⁸⁵ To be clear, prosecuting rioters who broke into the U.S. Capitol on January 6, 2021 is perfectly appropriate, but the strictness of the response highlights how little effort was made in many jurisdictions to punish the widespread rioting in 2020. Unfortunately, it appears rioting is only a crime in some jurisdictions when prosecutors disagree with the perceived politics of the rioters. Most Americans see rioting as criminal regardless of its motivations, but this important norm is endangered when prosecutors are perceived as only punishing rioters with certain political beliefs.

Non-Prosecution/Non-Arrest of Demonstrators Pulling Down Statues of Historic Figures of Whom They Disapprove.

While the motivations of prosecutors in refusing to enforce laws against rioting in 2020 were obviously political, the special case of the destruction of historic statues makes these motivations even clearer. Self-described “anti-racist” protestors in 2020 targeted a wide variety of public statues, ranging from those of Confederate generals like Robert E. Lee to anti-slavery heroes like Abraham Lincoln (because he signed off on the death warrants for a group of Native Americans in 1862).⁸⁶ Statues were targeted in Albuquerque, Birmingham, Boston, Chicago, Miami, Nashville, Pittsburgh, Richmond, St. Paul, and Portland among dozens of other cities across the country.⁸⁷ In almost all cases, none of those engaged in vandalism received any punishment, with local officials often publicly sympathizing with the actions or rushing to remove statues before they could be torn down by demonstrators.⁸⁸ For example, a group of protestors in Birmingham, Alabama was in the process of destroying statues on May 31, 2020, when Mayor Randall Woodfin arrived at the scene. Instead of ordering police to arrest the lawbreakers, he promised his support: “I understand your frustration. . . . Allow me to finish the job for you.”⁸⁹ Woodfin’s promise to use city resources to remove the statues temporarily placated the group, which abandoned the scene for riper targets around the city.⁹⁰ Soon the widespread looting and arson forced Woodfin to declare a state of emergency across the city.⁹¹

In addition to its direct effect through dropping charges, non-prosecution of selected groups may also lead to de facto decriminalization indirectly if police stop making arrests because they know charges will not result. Consider how few arrests were made for the dozens of public statues vandalized or destroyed during the 2020 riots, despite the blatant vandalism often being captured on video or with police standing by.⁹² For example, Portland police did nothing to stop protestors from toppling a statue of Abraham Lincoln (not exactly a racist icon) on October 11, 2020, perhaps understanding that District Attorney Mike Schmidt would never prosecute anyone arrested.⁹³ It seems more than plausible that the DA would have had a different view, and the police would have acted differently, had a mob of Ku Klux Klan members attempted to destroy the same statue. While this is admittedly speculation, the fact it will strike many as reasonable shows how the prosecution response (or lack thereof) to rioting and statue destruction in 2020 was perceived to hinge on politics instead of an impartial consideration of the underlying conduct.

C. Legislative Decriminalization

Legislative changes in law can effectively decriminalize conduct seen by a majority of the community as criminally condemnable. Two examples of this include California’s Proposition 47 effectively decriminalizing theft under $950, and Oregon’s Measure 110 decriminalizing the possession and use of “hard” drugs, including even the wildly dangerous fentanyl. Another example is legislative change to state or local law to create immigration sanctuaries designed to prevent enforcement of federal statutes criminalizing illegal entry into the United States.

California’s Proposition 47: Decriminalizing Theft Under $950.

On November 4, 2014, 59% of California voters passed Proposition 47, which was printed on the ballot as: “Criminal Sentences. Misdemeanor Penalties. Initiative Statute”⁹⁴ and campaigned for by its supporters under the rather misleading title: “The Safe Neighborhoods and Schools Act.”⁹⁵ The titles were chosen to convey the idea that “Proposition 47 intend[ed] to ‘ensure that prison spending is focused on violent and serious offenses,’ ” which would, according to the draftees, free up millions of dollars in funding that would then be spent on schools.⁹⁶ The ballot title also conveyed the idea that criminal convictions would remain for the conduct in question, with only the punishments being lowered.⁹⁷ In addition to downgrading various drug offenses, Proposition 47 downgraded numerous different felony theft crimes to misdemeanors where the value of the stolen property is $950 or less. The newly minted misdemeanors included, among other offenses, commercial burglary, grand theft crimes, possession of stolen property, and petty theft with priors.⁹⁸

What voters were not told is that the downgrades under the statute effectively prevented police from enforcing these crimes at all. Due to California law regarding misdemeanor arrest authority, under the new classification, police were no longer authorized to custodially arrest offenders but only issue a written citation, as with low-grade traffic violations.⁹⁹ Thus, for example, while previously police could respond to a report of retail theft and immediately arrest a suspect on probable cause for burglary, after Proposition 47, a citizen would first need to detain the offender and call the police, who could then only issue a citation.¹⁰⁰ As a result of these changes, stealing under $950 became essentially an unenforceable crime—a consequence voters could not be expected to foresee when voting for “The Safe Neighborhoods and Schools Act.”

The results of the de facto decriminalization were predictable. With police unable to even make a custodial arrest if someone caught a thief, there was no reason to even call police in the first place or report the crime.¹⁰¹ Some cities, such as San Francisco, have been hit particularly hard. Almost half of downtown San Francisco retail stores have closed, with many citing theft as a cause.¹⁰² Sacramento’s Democratic Sheriff, Jim Cooper, described retail theft as “beyond crisis level” in 2023 and shared the view of many law enforcement officials as to the underlying cause:

It really started with the change in law in 2014 with Prop 47, which was voted on by the voters, I should say, and the voters were duped into voting for that. It was called the [S]afe [Neighborhoods] and [S]chools [A]ct. . . . I think they have no idea. God bless the voters. I love the voters. But it’s hard, you read the ballot and you’ve got a lot of items on there and you’ll [see] ‘The Safe Neighborhoods and Schools Act? What does that have to do with drugs and theft?’ And that’s really what it was. It decriminalized seven crimes, drug and theft crimes. And that’s why we find ourselves in the predicament we’re in today.¹⁰³

With police unable to arrest thieves, California stores have responded by locking up more products. San Francisco’s Walgreens stores average 4 times more theft than the rest of the country,¹⁰⁴ with one store being hit more than 20 times per day, leading to desperate employees putting chains on the freezer section.¹⁰⁵ The sense of impunity among shoplifters is palpable. As one shoplifter stopping to explain his actions to reporters stated: “It’s San Francisco, Bro.”¹⁰⁶

The de facto decriminalization of theft in California has sparked outrage among even left-leaning California residents. A 2022 poll found that 59% of California voters supported changing Proposition 47 to reinstate many property crimes as felonies, while only 30% supported leaving the law intact.¹⁰⁷ Democratic state lawmakers responded in 2023 by introducing measures targeting repeat shoplifting, thus tacitly admitting the damaging consequences of Proposition 47. One such bill, AB 1772, “require[s] jail time if someone were convicted of a third theft crime” regardless of the value of the goods stolen.¹⁰⁸ Similarly, AB 1708 would have changed the law to allow any third theft crime to be charged as a felony.¹⁰⁹ Other democratic officials have openly acknowledged Proposition 47’s failure. For example, San Mateo’s County Supervisor, David Canepa, once supported Proposition 47 but now admits its decriminalization of theft was a mistake: “Enough is enough! All this retail theft. All this sort of crime. Enough is enough. We really need to look at state laws. What we have in place right now is not working. . . . We can’t go on like this.”¹¹⁰ Despite strong public support for cracking down on theft, the California legislature was slow to act. However, a bipartisan ballot initiative, Proposition 36, “The Homelessness, Drug Addiction, and Theft Reduction Act” gained the necessary signatures to allow voters to reverse parts of Proposition 47 directly at the ballot box in November 2024.¹¹¹ Proposition 36 proposed to make third-time shoplifting a felony regardless of the amount stolen, allow multiple shoplifting incidents to be consolidated and charged as a felony, increase the punishment for largescale retail theft, and increase penalties for drug offenses and force addicts into rehab.¹¹² In an attempt to prevent Proposition 36 from passing, the California legislature finally enacted a series of anti-theft bills in August 2024, including a measure allowing police to make warrantless arrests of shoplifters for thefts not committed in an officer’s presence.¹¹³ But this belated effort to address community concerns and stop a wider rollback of Proposition 47 failed. Proposition 36 passed overwhelmingly, with around 70% support, confirming that Californians had never intended or supported the de facto decriminalization of theft.¹¹⁴

Oregon’s Measure 110: Decriminalizing Hard Drugs.

In 2020, Oregon became the first state to decriminalize possession of “hard” drugs, when 58% of voters approved Measure 110.¹¹⁵ The law decriminalizes possession of drugs for personal use, including methamphetamine, heroin, and other controlled substances, and even the wildly dangerous fentanyl, which has become a leading cause of death in the United States for Americans aged 18–45.¹¹⁶ The possession of such drugs became merely a violation, and people who are caught in possession of the listed drugs are issued the equivalent of a traffic ticket, with the possibility of a fine. The fine is waivable if the individual makes a call to an addiction-recovery center and asks for a health assessment.¹¹⁷ Of course, enforcing the fine against violators is practically impossible, as there is no penalty for not paying the fine.¹¹⁸ With little incentive for violators to even ask for a health assessment, the vast majority did not.¹¹⁹

Similar to how Proposition 47 de facto decriminalized theft in California through misleading promises to voters, Measure 110 de facto decriminalized hard drug possession and use. While Measure 110 won approval by promising to more effectively move people struggling with addiction into treatment and off the street, the results, perhaps predictably, were quite different. Instead of reducing addiction, the de facto decriminalization turned Oregon into a drug user’s and dealer’s “paradise” according to users themselves.¹²⁰ As one investigative report found: “There are more dealers, scouting for fresh customers moving to paradise. That means drugs are plentiful and cheap.”¹²¹

Figure 1. Multnomah County Synthetic Opioid Overdose Deaths, 2018–2022

Source: Health Department Briefs Board of Fentanyl Crisis, Multnomah Cnty. (June 29, 2023), https://multco.us/news/health-department-briefs-board-fentanyl-crisis [https://perma.cc/H77A-BJZ7].

Drug usage and overdosing significantly increased in Oregon. Figure 1 shows overdose deaths from synthetic opioids in Oregon’s most populous county over time, which reflects the general trend across the state.¹²² In 2023, Oregon’s rate of fentanyl death had surged to 13 times the national average.¹²³ There were 1,416 opioid overdose deaths in Oregon in 2023, up from 472 in 2020 before the decriminalization, an increase of 200% compared to a national increase of only 18% over the same period.¹²⁴ In fact, opioid overdose deaths kept rising in Oregon in 2023 even as they declined nationwide.¹²⁵ A special treatment hotline received fewer than two calls a week on average from drug possessors ticketed by police, as most addicts simply ignored the tickets.¹²⁶ Far from ushering addicts into treatment, Measure 110 drove them to the morgue in ever greater numbers. Opioid overdoses became so common in Portland that medical professionals encouraged everyone to carry Narcan, a drug which can reverse overdoses.¹²⁷ As one nurse explained: “My kids know how to use it and they know where it is. . . . Honestly, if you live in Portland, Oregon, you don’t know when you might be someone who comes across someone who is overdosing.”¹²⁸ In December 2023, the state decided to provide Narcan to all schools requesting it, and 500 institutions signed up in a week.¹²⁹ As a result of surging drug usage and crime caused by Measure 110, public opinion in Oregon swung decisively against the decriminalization experiment. A 2023 poll found 64% of Oregon residents supported reinstating criminal penalties.¹³⁰ In September 2023, progressive Portland passed a local ban on public drug use with a penalty of up to 6 months in jail, but the law could not take effect without a change in state law.¹³¹ Sure enough, in April 2024, Oregon’s Democratic-controlled government recriminalized drug possession, with criminal enforcement resuming in September 2024.¹³² Other states have also learned from Oregon’s mistake. Washington State’s Democratic-controlled legislature chose to increase criminal penalties for drug possession in May 2023 instead of legalizing drugs when the state’s previous drug prohibition law expired that year.¹³³

Immigration “Sanctuaries”: Decriminalizing Illegal Entry into the U.S., Even by Criminals.

Few issues are as partisan in America today as the question of immigration. Many state and local governments have expressed support for allowing undocumented immigrants to remain in the country by passing so-called “sanctuary laws.” Currently, thirteen states have explicit or de facto statewide immigration sanctuary policies (including, most significantly, California), and other jurisdictions around the country have county- or city-wide sanctuary policies (including large cities like Baltimore and New York City).¹³⁴ As of 2015, roughly 300 jurisdictions in the U.S. had some form of sanctuary policies intended to limit the enforcement of federal immigration laws.¹³⁵

Such immigration sanctuary laws usually involve two parts. One part prohibits local law enforcement from ever asking any person about immigration status, and such policies can have clear popular support in many jurisdictions as residents, perhaps correctly, believe that such inquiries would discourage crime reporting and cooperation with police by undocumented immigrants.¹³⁶ The second part of sanctuary laws is less reflective of community views, as it involves requiring local police to refuse to follow federal detainer requests for criminal undocumented immigrants who are arrested for a non-immigration offense (e.g., theft or driving under the influence).¹³⁷ A detainer is a request for local police to hold the illegal immigrant offender until federal authorities can take custody of the individual and deport or prosecute them for illegal entry into the United States. Refusing to honor detainers de facto decriminalizes illegal entry by such criminal offenders because it prevents federal authorities from effectively punishing the crime of illegal entry committed by those offenders.

California’s sanctuary law illustrates this broad form of illegal entry decriminalization. On January 1, 2018, California became a sanctuary state for undocumented immigrants as progressive lawmakers promised to put a “kink” in the federal administration’s immigration enforcement.¹³⁸ The law banned local police from honoring immigration detainers issued by U.S. Immigration and Customs Enforcement (“ICE”) (with only rare exceptions) even if the person had been arrested for committing a crime. This decriminalization appeared to conflict with community preferences even in California, as a 2017 poll found a majority of Californians opposed allowing local jurisdictions to refuse to honor ICE detainers for undocumented immigrants arrested for a non-immigration offense.¹³⁹

Sheltering criminal undocumented immigrants proved costly. On December 13, 2018, 36-year-old Gustavo Garcia, an offender with a serious criminal record who had been caught and deported twice previously, was arrested for driving under the influence in Tulare, California. ICE again issued a detainer, which the local sheriff wanted to honor because Garcia’s past record showed he was a danger to the community, but the sheriff was obliged under the sanctuary law to refuse the ICE detainer and release Garcia. The next day, Garcia went on a rampage of robbery, shooting, and murder, killing two people and seriously injuring at least four others. The local sheriff was outraged that the sanctuary law prevented him from holding Garcia and saving lives.¹⁴⁰

Many may wonder how ideologically driven policies shielding criminal immigrants serve any just purpose or even protect the undocumented communities that sanctuary laws ostensibly aim to help. This frustration is bipartisan, as prominent Democrats have called for repealing protections for criminal non-citizens. In February 2024, New York City Mayor Eric Adams urged the city to change its sanctuary law to allow police to hand over criminal non-citizens to ICE: “If you commit a felony, a violent act, we should be able to turn you over to ICE and have you deported.”¹⁴¹

More broadly, sanctuary laws run counter to the national community’s views on justice. A 2017 poll found that an overwhelming 80% of voters believed local authorities should be required to report undocumented immigrants to federal authorities.¹⁴² And a 2024 poll found 65% of Americans support deporting undocumented immigrants (one can only imagine the likely higher support for deporting such immigrants who commit non-immigration crimes).¹⁴³ Despite public opinion favoring immigration enforcement, some thirteen states and hundreds of smaller jurisdictions are immigration sanctuaries, as illustrated by Figure 2.¹⁴⁴ Over 36% of Americans live in sanctuary states, despite such policies appearing to reflect only minority views nationally.¹⁴⁵

Figure 2. Sanctuary Jurisdictions in the United States

Source: Full Screen Map: Sanctuary Cities, Counties, and States, Ctr. for Immigr. Stud., https://cis.org/Full-Screen-Map-Sanctuary-Cities [https://perma.cc/ZA2P-C3K6].

D. If Decriminalizing Is Done by a Duly Elected Official or by the Voters Themselves, How Could It Conflict with Community Views? The Messiness of Democracy in Action

It is an obvious question to wonder how decriminalization brought about by a popular referendum or state legislation (such as California Proposition 47 and Oregon Measure 110) or by democratically elected officials (such as Larry Krasner or Buta Biberaj) could conflict with the community’s views of what ought to count as criminal conduct. Should a democratic system not, by its very nature, produce criminalization in line with community views? Unfortunately, the messiness of democracy in action means the nuanced views of the community on what conduct does and does not deserve criminalization will not always translate into public policy. There are several mechanisms by which this disconnect can occur.

Misleading or Package Deals.

A majority of voters may be convinced to support broad, combined, or misleadingly advertised policies even if the change will in fact lead to consequences that conflict with the community’s views on what counts as criminal conduct. For example, as noted previously,¹⁴⁶ Proposition 47 was passed in California under the ballot title “Criminal Sentences. Misdemeanor Penalties. Initiative Statute,” which actively suggests “criminal sentences” would continue to be imposed for the conduct in question with only the penalties changing. One may reasonably speculate that if California voters had understood that reclassifying felony theft as a misdemeanor would effectively decriminalize the conduct, they would have been less inclined to support Proposition 47 given that the vast majority of Californians still clearly believe $950 theft is criminally condemnable conduct deserving of arrest and punishment (though not always a prison sentence). This lack of community support for the de facto decriminalization was reflected in the overwhelming voter support for Proposition 36’s toughening of theft penalties.¹⁴⁷ Additionally, an initiative like California’s Proposition 47 was quite broad, combining changes to drug and theft criminalization.¹⁴⁸ Voters may vote for a “package deal” even though they oppose parts of the policy changes in the package. This kind of “horse-trading” may be common in politics, but it is seriously problematic when it ends up decriminalizing what the community perceives as criminally condemnable conduct, because crossing that line has significant societal costs.¹⁴⁹

A False Promise to Reduce the Condemnable Conduct.

Another kind of explanation for why voters might support legislation that decriminalizes criminally condemnable conduct is that voters may be convinced to put aside their underlying views on criminalization as a result of rhetorical or ideological campaigns advocating the supposed societal benefits that would result from such decriminalization. For example, it appears a majority of Oregon voters were convinced to pass Measure 110, despite viewing the use of “hard” drugs like fentanyl as criminally condemnable, because the measure’s supporters promised it would allow for more effective reduction in drug usage.¹⁵⁰ As soon as it became clear the decriminalization was not reducing addiction as promised, voters apparently reverted to insisting that the condemnable conduct be recriminalized. This makes the passage of Oregon’s Measure 110 quite distinct from marijuana legalization referendums in other states which appear to reflect an actual change in societal views of the underlying behavior.¹⁵¹ In other words, it seems Measure 110 was passed in an effort to reduce the incidence of conduct the community always saw as criminal, while many marijuana decriminalization laws have been passed because the underlying conduct is no longer seen as criminal. Part II catalogues the many motivations behind decriminalization (including this harm-reduction motivation), and some of these motivations can lead majorities of voters to temporarily override their criminalization judgements. However, as Part III argues, decriminalizing conduct that the community sees as criminally condemnable is almost always a mistake—even if campaigners have temporarily convinced voters to experiment with it.

Election Package Deals.

Duly elected officials may substitute their minority judgements on decriminalization while still winning elections by either concealing their decriminalization intentions or by supporting popular majority positions on other topics that matter more to voters. For example, Buta Biberaj did not advertise her plans to effectively decriminalize most domestic violence when she won election to her office in 2019.¹⁵² Similarly, while Larry Krasner was open about his intentions to pursue non-prosecution policies, it may not have been clear to voters how extensive some of those policies would be, as reflected in the fact that many Philadelphia Democrats are seeking to reverse his reforms.¹⁵³ Additionally, Krasner, Biberaj, and other progressive prosecutors may win voter support in their jurisdictions by supporting majority positions on other topics (e.g., Democratic candidates may advertise their support for protecting abortion rights, increased welfare support, support for environmental protection, etc.). The large number of political issues voters must juggle, and the limited attention voters pay to the political process also allows legislators to regularly vote, and for prosecutors and local officials to act, in a way that is contrary to community views on criminalization.¹⁵⁴ This is especially true since officials can choose not to advertise the ways they act contrary to community views. This is one reason special interest groups can dominate the legislative process, as such groups might convince legislators to vote against public views in some areas while funding candidates’ ability to publicize popular stances in others.

Election Only by a Majority (or Even Plurality) of the Majority Party.

In jurisdictions dominated by one party, as is the case for most cities, a candidate for local office need only win a majority—or even just a plurality—of the dominant party’s primary votes (often representing the most partisan wing of that party) to win the larger election because of partisan voting in general elections. For example, Larry Krasner was first elected to office in Philadelphia after winning the Democratic primary with 38% of the vote¹⁵⁵ (a win enabled by generous outside campaign donations from progressive megadonor George Soros¹⁵⁶). Partisanship and machine politics is one reason why local officials may survive in office even while pursuing deeply unpopular policies.

Election by Only a Majority Within a Locale, in Which the Constitution Gives Criminalization and Decriminalization Authority to the State (or Federal) Government.

Decriminalization can conflict with broader community views if there is a jurisdictional conflict in which a local community has a different view of what conduct is criminally condemnable than the broader community. For example, a clear national majority in America opposes immigration sanctuary laws,¹⁵⁷ but majorities in individual cities or states may support them (or at least part of such laws). In addition to raising federalism concerns, such a clash is deeply unfortunate as it guarantees some members of the community will be dissatisfied with the criminal law. However, short-circuiting legislative change at the constitutionally authorized level of criminalization authority may do more harm than good even for those supporting decriminalization. Such local decriminalization produces a continuing string of failures of justice in the eyes of the broader community while the local community ceases to view generating broader support for decriminalization as urgent. This perpetuates a fragmented criminal law as opposed to allowing local decriminalization support to grow until it achieves majority support in the broader jurisdiction.

II. Decriminalization Motivations

There are a number of sometimes overlapping motivations and justifications that drive the decriminalization of conduct that the community sees as criminally condemnable. One of the most common is the anti-criminal justice motivation, described in Section II.A below, which generally opposes the use of the criminal justice system as a response to crime and would prefer dealing with antisocial conduct through medical, public health, or social service authorities. Section II.B discusses decriminalization motivated by a desire to advance “social justice” by reducing criminal arrests and convictions of persons seen as members of an oppressed group, such as racial minorities. Section II.C examines what might be called the “moral superiority” motivation in which those with minority views on decriminalization see themselves as holding a superior moral belief and so believe they know better than the community what should and should not be criminalized. Officials pursuing decriminalization under this view often imagine themselves as part of a brave moral minority with an obligation to move society’s laws to align with their own views of what constitutes a better society. Finally, Section II.D discusses the “local superiority” motivation: the view that no matter what the state or federal constitution may say about the proper allocation of the criminalization-decriminalization authority, the local view on decriminalization ought to prevail, even if the broader community opposes the resulting state of affairs.

Decriminalization contrary to majority community views can be driven by one or by any combination of these motivations or justifications, depending on the decriminalizer and the issue at hand.

A. The Anti-Criminal Justice Motivation: Rejecting the Importance of Doing Justice in Favor of a Medical, Social Services, or Mental Health Response to Crime

One of the most common motivations for and claimed justifications of decriminalization contrary to community views is the view that crime, or sometimes just certain crimes, are a problem best dealt with through various government agencies other than the criminal justice system.¹⁵⁸ Similar to some of the motivations that underlay the “Defund the Police” movement, this view suggests that society would be better off in the long run if crime was dealt with by medical, social services, or mental health actors rather than the criminal justice system.¹⁵⁹ Under this view, criminally condemnable conduct ought not to be dealt with based upon its moral blameworthiness but should instead be dealt with through means that provide therapeutic intervention, social services, training and education, and any other non-criminal justice mechanism that may over the long-term work to eliminate the problem of crime.¹⁶⁰ (This view is sometimes referred to as “harm reduction,” particularly in the context of drug decriminalization.)¹⁶¹ As one proponent of Oregon’s Measure 110 argued, “far from spurring recovery, arrest, incarceration and having a criminal record can exacerbate drug problems.”¹⁶² The argument goes that criminalization has failed to solve the underlying crime problem and that a new, usually public health approach, is needed. So, while drug use or prostitution, say, might not qualify as “licit” conduct, they should be decriminalized to reduce the harm caused by such condemnable behaviors. For example, supporters of Oregon’s Measure 110, discussed previously, persuaded voters to decriminalize the use of hard drugs by employing such harm reduction arguments.¹⁶³ As Mike Schmidt, the Multnomah DA and supporter of drug decriminalization, argued: “What we’ve been doing for the last number of decades has completely failed. . . . Criminalization keeps people in the shadows. It keeps people from seeking out help, from telling their doctors, from telling their family members that they have a problem.”¹⁶⁴ Supporters of Measure 110 argued that removing the criminalization of illicit drug use would cause more addicts to seek treatment, thus reducing the damage caused by drug usage.¹⁶⁵

An anti-criminal justice approach is also a hallmark of progressive prosecutors’ non-prosecution policies that seek to divert offenders into alternative programs for treatment.¹⁶⁶ Prosecutors such as Larry Krasner view punishment as pointless unless they believe it will directly reduce societal harm. For example, Krasner has defended his non-prosecution of illegal gun possession on the basis that: “We do not believe that arresting people and convicting them for illegal gun possession is a viable strategy to reduce shooting.”¹⁶⁷ In other words, for Krasner and those like him, the criminal law should only be enforced if it is instrumental toward a social goal. Under this view, criminal conduct should not be punished simply because it is criminally condemnable.

The anti-criminal justice motivation is also one reason why some support the decriminalization of domestic violence, a policy Buta Biberaj did her best to implement. As one supporter explains:

Other alternatives to prevent and address the harms of domestic violence also exist. Economic interventions could relieve some of the conditions that spur domestic violence. . . . [D]omestic violence correlates with male unemployment. Job training and employment provision programs might do more to prevent intimate partner violence than criminalization has. If prevention is the goal, public health initiatives might serve that function more effectively than criminalization. Criminalization . . . hampers the development, implementation, and evaluation of these types of alternatives. So long as criminal laws continue to exist, criminalization will be the default response that policymakers and anti-violence advocates are loath, even afraid, to abandon. And so long as funding for anti-domestic violence efforts remains focused on the criminal legal system, criminalization will deprive efforts to develop alternatives of needed resources.¹⁶⁸

While some harm-reductionists deny that their non-punishment policies will produce any additional crime,¹⁶⁹ other proponents acknowledge there may be some short-term costs. As John Chisholm, Milwaukee’s progressive DA, admitted: “Is there going to be an individual I divert, or I put into [a] treatment program, who’s going to go out and kill somebody? You bet. Guaranteed. It’s guaranteed to happen. It does not invalidate the overall approach.”¹⁷⁰

Related to anti-criminal justice motivations is the desire to decriminalize criminally condemnable conduct in order to reduce incarceration, because incarceration clearly has negative impacts on the individual offender. Many decriminalization initiatives cite reducing prison populations as their goal. For example, Larry Krasner described one of the main goals of his non-prosecution policies as reducing incarceration: “These policies are an effort to end mass incarceration and bring balance back to sentencing.”¹⁷¹

However, decriminalization is an extremely crude way to reduce prison populations. We strongly agree with efforts to reduce the use of incarceration as a method of punishment, but this goal can be achieved for all but the most serious offenses simply by shifting deserved punishment to non-incarceration forms of sanction, as we have argued elsewhere.¹⁷²

B. The Social Justice Motivation: Using Decriminalization to Reduce Sanctions Against Groups Seen as Oppressed

Many advocates of decriminalization support it as a way to advance equity by eliminating criminal penalties that appear to disproportionately impact certain groups. For example, poor people and racial minorities make up a larger share of criminal offenders than their percentage of the general population, and so they also make up a larger share of arrests, prosecutions, and imprisonments.¹⁷³ It is also often alleged that the justice system is biased in its treatment of such offenders—for example, a Black offender is thought to be more likely to be arrested, prosecuted, and imprisoned than a similar White offender.¹⁷⁴ There is good reason to be skeptical of these claims, although a full analysis of racial disparities in the justice system is beyond the scope of this Article.¹⁷⁵ Regardless of its merits, however, the claim of systemic racism in the justice system is a common theme in justifying decriminalization.

As a result, some activists believe decriminalization is a useful tool to defang a racist and oppressive system. While conduct like theft might still be criminally condemnable, it is worth decriminalizing because criminal penalties for theft fall disproportionately on individuals who are already disadvantaged and discriminated against.¹⁷⁶ Advancing equity was an explicit motivation behind the campaign for Proposition 47 in California,¹⁷⁷ and progressive prosecutors’ justification of their policies in effectively decriminalizing condemnable conduct are typically peppered with justifications based on reducing systemic disparities and oppression.¹⁷⁸

Such motivations may see the criminal justice system’s role as not to do justice for a given crime, but to consider what is equitable for a given offender—based less on their actions than their social circumstances.¹⁷⁹ At its most extreme, this group justice mentality merges with a completely anti-criminal justice philosophy because practically all criminal offenders can be seen as belonging to a societal group victimized in some way by state-sanctioned violence and discrimination.¹⁸⁰ Letting such offenders go unpunished for criminally condemnable conduct may be considered a way of attempting to rebalance the scales of social justice. As one author explains why punishment is not about crime:

On the surface, crime and punishment appear to be unsophisticated matters. After all, if someone takes part in a crime, then shouldn’t he or she have to suffer the consequences? But dig deeper and it is clear that crime and punishment are multidimensional problems that stem from racial prejudice justified by age-old perceptions and beliefs about African Americans. . . . In other words, society chose to use incarceration as a welfare program to deal with the poor, especially since the underprivileged are disproportionately people of color.¹⁸¹

This social-justice motivation has even led to proposed or implemented broad decriminalization of entire types of conduct for certain groups of persons seen as disadvantaged in some way. A 2020 proposal from Seattle’s city council, for example, would have prohibited “prosecution for misdemeanor crimes for any citizen who suffers from poverty, homelessness, addiction, or mental illness.”¹⁸² The law would “effectively legalize an entire spectrum of misdemeanor crimes, including theft, assault, harassment, drug possession, property destruction, and indecent exposure. Criminals must simply establish that they have an addiction, mental-health disorder, or [are] low income in order to evade justice.”¹⁸³ While the proposal did not pass, it was merely a more explicit form of a motivation common within decriminalization: those already disadvantaged by nature or society should not be further “victimized” through criminal punishment.

C.The Moral Superiority Motivation: Believing It Appropriate to Substitute One’s Own (Superior) Criminalization Judgments for Those of the Community

Another common motivation among decriminalization advocates is believing the majority view on criminalization is simply morally wrong and should be replaced by their own morally superior view. For example, Multnomah County DA Mike Schmidt defended his decision to effectively decriminalize rioting in the wake of George Floyd’s death based on his personal support of the motivating cause that he attributed to the rioters. As noted previously, he appeared to believe it would be morally wrong to prosecute individuals protesting (even criminally) what he saw as a societal evil: “We will undermine public safety, not promote it, if we do not take action to bring about immediate change.”¹⁸⁴ Similarly, Chicago DA Kim Foxx’s decision not to prosecute rioters was due to the “righteous anger” she saw as their motivation.¹⁸⁵ San Francisco’s progressive prosecutor, Chesa Boudin, also forged ahead with a series of non-prosecution policies after narrowly winning election to his office in 2019, despite broad community disapproval of his de facto decriminalization policies. While Boudin undoubtedly believed he was implementing morally right policy positions, residents saw things differently, choosing to decisively recall him in 2022.¹⁸⁶ As one San Francisco resident and reporter explained, “yesterday, San Francisco voters decided to turn their district attorney, Chesa Boudin, out of office. They did it because he didn’t seem to care that he was making the citizens of our city miserable in service of an ideology that made sense everywhere but in reality.”¹⁸⁷ Sometimes, progressive prosecutors show a remarkable contempt for contrary community opinions, perhaps stemming from unbridled confidence in the rightness of their own views. LA’s progressive DA George Gascon, for example, responded to critics of some of his lenient policies by stating they were “too uneducated to shut their mouths for once.”¹⁸⁸ Incidentally, LA voters did not take kindly to such views, and Gascon lost reelection in 2024 by a twenty-point margin.¹⁸⁹

Of course, there is nothing wrong with advocating for a change in community views, and as long as decriminalization advocates change public views before actually decriminalizing conduct, they may be helping society. However, when elected officials substitute their personal moral judgements for community views, they are engaging in deeply undemocratic and elitist decision-making, in addition to causing what the public will perceive as failures of justice, which are especially objectionable because they are the product of elitism. Elected officials must recognize their duty in a democratic society is not to be philosopher kings, no matter how tempting that prospect may be. While it is true that moral minorities may actually represent moral progress, it is equally true that they may represent the opposite. Historically, for every brave minority of slavery abolitionists, there have also been “brave” minorities of eugenicists, racists, and would-be genociders eager to impose their views on the criminal law and society. For example, the American eugenics movement, pioneered by intellectuals and elite progressive reformers who derided lawmaking by the people, believed eugenic policies represented the next great moral advance in the early 20th century.¹⁹⁰ Fortunately, they were never able to win over enough of the American electorate to fully implement their vision. The “brave” minority of Nazi eugenicists in Germany eliminated any such democratic constraint.¹⁹¹ Having the criminal law (and government policy more generally) track the views of the community instead of visionary elites helps defend against moral depravity even while sometimes delaying moral progress. The desire to decriminalize for the sake of minority moral beliefs is understandable, but officials in a democratic society must defer to the community they have been elected to serve and represent. If they have a good moral case for decriminalization, they are likely to persuade a majority of society in the end.

Ironically, many of the proponents of minority-view decriminalization furiously oppose other groups holding minority views from attempting to impose their views on the criminal law. For example, many on the left have (rightfully) criticized minorities on the right for attempting to pass extreme abortion criminalization legislation unrepresentative of community views.¹⁹² Additionally, almost everyone would rightfully balk at a hypothetical far-right DA choosing to decriminalize crimes motivated by racism because of his personal belief in the morality of White supremacy. Supporting local officials imposing their minority beliefs on the criminal law is so obviously short-sighted that it is hard to believe any thoughtful person could ever think it appropriate in a democratic system.

D.The Local Superiority Motivation: Believing It Appropriate to Substitute Local Criminalization Judgments for Those of the Larger Constitutionally Empowered Jurisdiction

Finally, when the officials (or sometimes even the population) in a city or county view certain conduct as not criminally condemnable while such conduct is still viewed as criminally condemnable by a clear majority of the state’s population, the local jurisdiction’s officials may rely on a notion of local superiority in making the criminalization-decriminalization decision even though the state constitution unequivocally gives that power to the state legislature. For example, Austin declared itself a “freedom city” in 2018, when the Austin City Council passed a series of measures designed to prevent enforcement of parts of the Texas criminal code.¹⁹³ The Council required police to “avoid arrests” for a variety of lower-level offenses, such as low-level theft and driving without a license or on a suspended license.¹⁹⁴ The goal was to frustrate what Austin lawmakers saw as an overly harsh state criminal code, and within 3 months, Class C misdemeanor arrests in Austin dropped by 63%.¹⁹⁵

The same substitution of local criminalization judgements for those of the broader community may occur between states and the federal government. For example, most of the provisions of California’s state sanctuary law passed in 2018 had majority support in California, even while its de facto decriminalization of illegal entry was opposed by a clear majority of U.S. residents nationally.¹⁹⁶ In the case of illegal immigration, the federal government clearly possesses the constitutional criminalization authority,¹⁹⁷ so an individual state or city’s decision to frustrate that enforcement and effectively decriminalize the conduct is problematic, even if the U.S. Constitution does not require the state to assist federal authorities in the federal law’s enforcement.¹⁹⁸

The Dangers of Local Criminalization Policies.

Substituting local judgements on criminalization for those of the broader community is dangerous when the criminalization authority is constitutionally vested in the broader jurisdiction. Such a local seizure of criminalization authority undermines principles of fairness and equal treatment which require the same standard of criminalization and punishment apply to all people within a jurisdiction. For example, why should a thief in Austin go unpunished while a thief one town over receives punishment for the same offense under the same Texas criminal code? This also creates unfairness for law-abiding residents: why should a store owner receive less protection from the criminal law depending upon city or county borders?

The state (or federal) law is meant to protect and punish equally across the jurisdiction, but patchwork local decriminalization, whether by legislatures or prosecutors, undermines this principle. Local decriminalization decisions made by prosecutors are especially damaging as they also introduce a level of unpredictable whim that undermines the rule of law. The legality principle requires fair notice of what conduct is and is not subject to criminal punishment.¹⁹⁹ But local prosecutorial decriminalization means the de facto law even within the same county is subject to change without notice as prosecutors adjust their charging policies based on the political climate or an election that ushers in a new prosecutor with different political views.²⁰⁰ The result is often a massive change in the treatment of citizens in the same place even with no change in law—the very definition of the “rule of the individual” as opposed to the rule of law.

Battles over jurisdictional control of criminalization are becoming more common, with various local officials looking to effectively decriminalize gun crimes or abortion crimes in opposition to state laws.²⁰¹ There will inevitably be tensions between local and state or state and federal communities, but those disputes in a democracy need to be resolved by democratic action at the constitutionally appointed criminalization level, rather than being resolved by local usurpation of the larger community’s criminalization-decriminalization authority.

The latter approach will only ensure a loss of system credibility at all levels. Even if the local jurisdiction’s residents support the local decriminalization in question, it would be no surprise if this local decriminalization prompted them to see the broader jurisdiction’s criminal law as illegitimate. If it is appropriate to ignore the broader community’s criminalization-decriminalization judgments on one specific issue, why not on other issues as well? Even residents in the broader jurisdiction may lose confidence in their criminal law when they see how easily and frequently it is flouted by local jurisdictions. De facto decriminalization at the local level may also prevent pressure from building to modify the criminalization at the appropriate state or federal level, thus entrenching patchwork criminalization and undercutting the purpose of having a state or federal criminal code in the first place. Abiding by the broader jurisdiction’s criminalization decisions seems wiser because it avoids these and other societal costs, many of which are discussed in the following Part.

III. The Societal Costs of Decriminalizing Condemnable Conduct: Miscalculating the Balance of Societal Interests

Decriminalizing conduct that a clear majority of the community sees as condemnable comes with a host of destructive costs commonly ignored by the backers of such decriminalization. Some of those costs are immediately apparent, as in the instant loss of deterrent effect, as discussed in Section III.A below. Soon after, the loss of incapacitation of repeat offenders starts taking effect, as rioters, looters, vandals, organized thieves, and others who would have been arrested and prosecuted for their initial offenses are free to continue their victimizations uninterrupted, now increasingly confident that they will not be arrested or prosecuted, as discussed in Section III.B.

More damaging perhaps than even the immediate societal costs of increased crime from lost deterrence and incapacitation is the long-term damage from undermining the criminal law’s moral credibility with the community by failing to criminalize and punish what is seen as criminally condemnable conduct, discussed in Section III.C, as well as the long-term damage from undermining the criminal justice system’s “legitimacy” in the eyes of the community by allowing the justice system to be governed by ideological bias rather than a commitment to a system of fairness and equality for all, a societal cost discussed in Section III.D.

What is perhaps most tragic about the unnecessary criminal victimizations that these policies cause is that the costs are borne disproportionately by the most vulnerable, as Section III.E documents. It is primarily poor and minority communities that suffer the increased victimizations, not the elites who push such policies, often in the name of equity.

As Part II discussed, some supporters of decriminalizing criminally condemnable conduct may concede the resulting costs of temporarily increased crime but argue that these short-term societal transition costs are worth the long-term benefits that they foresee, such as the benefits of replacing the criminal justice system with medical, psychological, and social services, which they believe will, in the long run, produce a society with less crime. In other words, they see the short-term costs as justified by the long-term gains.

Is it true that the short-term costs of increased crime victimizations borne disproportionately by vulnerable minority communities are justified by long-term societal benefits? We think not. As Section III.F explains, the claimed long-term benefits used to justify tolerating the additional victimizations are at best speculative, with no precedent in human history. There has never existed a society that did not have a functioning criminal justice system of rules and punishments.²⁰² Attempting to replace punishment with therapeutic intervention (as opposed to supplementing punishment with treatment) goes against the most basic human instincts. Paying destructive costs now for a speculative vision unsupported by human nature and history does not reflect a reasonable balancing of societal costs and benefits.

A.The Immediate Societal Cost of Reduced Deterrence

Decriminalizing conduct tends to lead to more of that conduct, which is disturbing when that conduct is still viewed as criminally condemnable by the community. It requires a suspension of common sense to believe the possibility of punishment for a behavior can be eliminated or drastically reduced without having some effect on the occurrence of that behavior. Indeed, the decriminalization policies discussed in this Article had just such an effect, often producing sizable or even shocking increases in lawbreaking. Theft increased statewide after California’s Proposition 47, with shoplifting surging by as much as 50% in some places.²⁰³ Drug usage increased after Oregon’s Measure 110, with overdose deaths reaching thirteen times the national average.²⁰⁴ Shootings reached record highs after Larry Krasner’s non-prosecution of gun crimes in Philadelphia, with homicides rising 78% in 2021 compared to when Krasner was elected in 2017. Non-fatal shootings also surged.²⁰⁵ Practically all criminal activity surged in Austin after it effectively deleted parts of the Texas criminal code.²⁰⁶ The violent crime rate in Austin was up almost 30% in 2021 compared to 2018.²⁰⁷ The number of homicides also rose 178%.²⁰⁸ A citizens group reported in 2021 a 38% increase in aggravated assaults, an 87% increase in burglaries, and a 124% increase in auto thefts since 2019.²⁰⁹

While defenders of these and other de facto decriminalization policies contrary to community views often quibble over the exact size of crime increases and try to introduce doubt as to what caused them, it requires ideological gymnastics not to recognize that offenders and potential offenders respond to incentives. If the legal consequences from stealing $950 worth of merchandise becomes at most the equivalent of a parking ticket and one is almost never caught, how could this not encourage more theft? If potential offenders know that the local DA is literally called “Let ‘Em Out Larry” on the street due to not prosecuting most criminal conduct, how could this not embolden offenders to carry weapons and commit other crimes they rightly believe are unlikely to lead to prosecution even if detected?²¹⁰ It should not take a study to prove the point that removing criminal consequences for a behavior will make that behavior more likely, but, of course, studies have found these effects.²¹¹

Recognizing the crime-increasing consequences of particular non-prosecution policies should in no way be partisan. Liberal Black mayors such as San Francisco’s London Breed and New York City’s Eric Adams have attacked non-prosecution policies, showing that criticism is hardly a partisan rightwing reaction.²¹² Moreover, some progressive prosecutors have reversed some of their de facto decriminalization policies in the wake of crime surges, suggesting even they understand the crime consequences.²¹³

B.The Immediate Societal Cost of Reduced Ability to Incapacitate Repeat Offenders

In addition to reducing deterrence, de facto decriminalization increases the decriminalized conduct (and overall crime) through reducing the incapacitation of criminal offenders. It is a well-known fact in criminology that a small percentage of criminals commit a large percentage of crimes,²¹⁴ meaning that much, if not most, crime is a problem of serial offenders. For example, the Charlotte-Mecklenburg Police Department found that “repeat shoplifting offenders account for 40.1% of shoplifting incidents where a suspect was identified.”²¹⁵ Similarly, “[j]ust 327 serial shoplifters accounted for a third of the more than 22,000 retail theft arrests made” by New York City police in 2022.²¹⁶ A combination of New York’s bail policies and New York City’s progressive prosecutors’ non-prosecution policies allowed such thefts to thrive because the thieves were not incapacitated. While an individual crime is unlikely to lead to punishment, a serial offender normally plays the odds until capture and punishment, which often takes the form of incarceration, thus incapacitating the offender from further crime for a period. But when even capture leads to no punishment, serial offenders can continue their busy lives of crime uninterrupted. Since frequent offenders tend not to specialize,²¹⁷ choosing to commit a range of crimes from minor to serious, decriminalizing lower-level crimes removes the possibility of incapacitating many serious offenders before they commit more or worse crimes. For example, one study tracked what happened to 231 defendants charged with a firearms violation in Philadelphia in July 2019. 16 months later, owing much to Krasner’s de facto decriminalization, 176 had been freed, and 51 “were arrested again and charged with more crimes that included gunpoint robberies, armed carjackings, rape, attempted murder, as well as a half-dozen murders.”²¹⁸ Repeat offenders will go on longer strings of offenses when the probability of punishment is decreased or eliminated—a fact that should be obvious to any good-faith observer.

C.The Long-Term Societal Cost of Reducing Criminal Law’s “Moral Credibility”

Aside from immediate reductions in deterrence and in the ability to incapacitate repeat offenders, decriminalization contrary to community criminalization judgements may have an even greater long-term cost by undermining the moral credibility of the criminal law through exposing the community to what they perceive as regular and intentional failures of justice. Reducing the credibility of the law decreases compliance, increases resistance and subversion, sparks vigilantism, and degrades the system’s ability to shape people’s internalized norms.²¹⁹ Unlike what some decriminalization advocates seem to assume, people do not simply accept what the law says without checking it against their own intuitions of justice.²²⁰ As a result, decriminalizing conduct that a clear majority of the community sees as criminally condemnable simply undermines the criminal law’s reputation as a reliable moral authority that should be obeyed. If the criminal law corrupts its “brand,” the clarity and influence of what it means to label something as “criminal” is lost. This is common sense, but these dynamics have also been confirmed by a host of controlled laboratory studies as well as shown in natural experiments. These studies and experiments, which have been described in detail elsewhere,²²¹ support three key conclusions:

Ordinary People Have Strong and Nuanced Shared Intuitions of Justice Based on Desert, Which They Believe the Criminal Justice System Should Uphold

Empirical studies show that ordinary people think of criminal liability and punishment in terms of desert—offenders should get the punishment they deserve rather than the punishment that might best deter others or best incapacitate dangerous offenders.²²² While the public believes deterrence and incapacitation is a general goal of having a criminal justice system on an abstract level, when confronted with actual case examples, people overwhelmingly assign punishment based on desert. In fact, people who self-report a preference for non-desert distributive principles still favor retribution in their actual judgements.²²³ This finding is particularly noteworthy as regards decriminalization, as it shows how some community members can paradoxically hold two views—a California or Oregon voter might intellectually believe criminal punishment should be assigned according to principles of harm reduction leading to decriminalization, but they might still intuitively feel the conduct being decriminalized deserves criminal punishment. An academic may support de facto decriminalizing theft, but when her or her friend’s purse is snatched and the thief suffers no consequences, she may still feel justice has failed.

Additionally, laypeople’s intuitions of desert are closely correlated to one another, allowing the criminal law to use those shared intuitions as the basis of a criminal code. For example, one study asked participants to rate the seriousness of 51 offenses relative to bicycle theft, and the correlation was around 0.9 across various demographics, showing an overwhelming societal consensus about what conduct is criminal and how serious that criminality is.²²⁴

Not only do people have shared intuitions about what conduct calls for criminal punishment, they also have a strong innate desire to see it done, even if they have no personal interest in the case.²²⁵ Studies have shown that even preverbal infants display a third-party desire to punish offenders—demonstrating just how deep and instinctual the human desire for desert-based justice is.²²⁶ A justice system that fails to meet that demand will lose credibility, as the next subsection shows.

When Community Members See the Criminal Law Regularly Deviating from Their Shared Intuitions of Justice, the Moral Credibility of the Law Declines

This claim is rather intuitive, but it is also supported by clear empirical evidence.²²⁷ For example, the next subsection presents a variety of studies in which experimenters varied the subjects’ perception of the criminal law’s moral credibility in order to observe the resulting shift in the subjects’ willingness to comply with the criminal law.²²⁸ How do researchers vary subjects’ sense of the criminal law’s moral credibility? Answer: By exposing subjects to more (or fewer) cases where the criminal law’s results conflict with the subjects’ own sense of justice, or by exposing subjects to cases where the conflict with their intuitions of justice is greater (or lesser).²²⁹ This standard experimental method has been shown to work reliably and predictably.²³⁰ The greater and more frequent the conflict with subjects’ justice judgments (either in the form of perceived injustices or justice failures), the greater the loss of the justice system’s moral credibility in their eyes.²³¹ Even beyond this common mechanism for manipulating the criminal justice system’s moral credibility with subjects, studies of actual participants in the justice system have shown the same dynamic.²³²

These empirical studies reinforce the notion that regular conflicts with community views as to the just allocation of criminal punishment will reduce the criminal law’s moral credibility.²³³ It is widely acknowledged that prevalent injustices (instances of over-punishment or punishing the innocent) erode the moral credibility of the justice system, but studies show ordinary people care as much about preventing justice failures (where the guilty escape their deserved punishment) as they do about preventing injustices (such as false convictions), unlike many academics who pay dramatically more attention to wrongful punishment.²³⁴

As the Moral Credibility of the Law Declines, Its Crime-Control Effectiveness Also Declines

If a criminal law is widely viewed as unrepresentative of community beliefs, this is likely to negatively influence the community’s deference to the law—a fact that hardly needs stating. In what world would citizens be indifferent to the law’s poor performance in doing justice? The evidence from history shows that when disillusionment with the criminal law sets in, lawbreaking rises.²³⁵ This is because compliance with the law is generated in significant part from the law’s reputation as a reliable indicator of what society genuinely condemns.²³⁶ If the law is seen as unrepresentative of society’s justice judgements, then its ability to harness the powerful forces of social influence are correspondingly reduced.²³⁷ As a result, decriminalization of select conduct can reduce the entire criminal law’s ability to gain compliance—a hidden cost to decriminalization that goes far beyond merely increasing the decriminalized conduct itself. This disillusionment-noncompliance dynamic can be seen in numerous controlled studies and natural experiments.²³⁸

Consider just two natural experiments where a failure of over or under criminalization reduced the criminal law’s moral credibility and compliance:

Gilded Age New York City

At the end of the nineteenth century, the New York City legislative process was notoriously corrupt: even valuable and legitimate legislation could not be passed unless the right political players were paid off.²³⁹ The result was a criminal law that failed to punish the full range of conduct that social mores at the time saw as condemnable, such as pornography, abortion, and gambling.²⁴⁰ As the criminal law came to be seen as increasingly out of touch with community norms, and unable to respond to criminally condemnable conduct, crime rates escalated—even for those crimes that were still enforced.²⁴¹ Street gangs proliferated and even shoplifting among middle-class women rose, as breaking the law was viewed as less serious. With some “crimes” being effectively legal, all crimes seemed more permissible.²⁴²

American Prohibition

In 1920, Congress criminalized the sale of alcohol despite the fact most Americans did not view alcohol as a matter for the criminal law. Demand for alcohol remained high, and lax or corrupt enforcement contributed to a sense that the law was illegitimate. As trust in the law eroded, Americans increasingly violated the law. Most importantly, the disillusionment tainted not only the alcohol-prohibition rules, but it also reduced compliance with criminal law rules generally,²⁴³ including those unrelated to alcohol.²⁴⁴

In addition to these historical case studies, social science research suggests that the relationship between the law’s moral credibility and community deference and compliance is widespread and nuanced.²⁴⁵ Even small incremental losses in moral credibility can produce corresponding incremental losses in deference and compliance.²⁴⁶ Consider a 2010 study using a within-subjects design in which participants were asked questions about how they related to the justice system, such as questions about their willingness to report crimes, assist justice system officials, and view the law as a reliable moral guide. After baseline responses were collected, subjects were then disillusioned by exposing them to real accounts of the system’s failures of justice and injustices. Retesting revealed the measures of deference, compliance, and internalization of norms had decreased among the disillusioned subjects.²⁴⁷ This finding is particularly important because it indicates that no matter the current state of a criminal justice system’s moral credibility with an individual or the community, any incremental reduction in credibility (such as by decriminalizing only seemingly minor condemnable conduct) can produce an incremental reduction in deference, and any increase can produce an increase in deference.

Reducing the moral credibility of the law not only makes people more likely to engage in general criminal conduct, but it also creates another cost through sparking vigilantism because community members increasingly believe justice will not be served by the system.²⁴⁸ Some of those who care strongly about doing justice sometimes turn to appointing themselves judge, jury, and even executioner when the criminal justice system seems unable or unwilling to do justice.²⁴⁹ De facto decriminalization contrary to community views can make outraged citizens turn to violence, and even murder. For example, when Dallas County DA John Creuzot announced a non-prosecution policy of theft under $750 dollars in 2019, there was a predictable surge in shoplifting.²⁵⁰ Dallas residents became increasingly angry as they watched more store items being locked behind glass even as no thieves were locked behind bars. In December 2022, one Dallas resident, Kevin Jacobson, had had enough. When he saw a thief fighting with store clerks, he ordered the clerks to move and gunned down the shoplifter in an act of vigilante “justice.”²⁵¹ Such disproportionate violence is a hallmark of vigilantism, which despite being romanticized in popular culture, only contributes to a decline in the law’s moral credibility by showing the justice system has lost its monopoly on criminal punishment.²⁵²

D.The Long-Term Societal Cost of Reducing the Criminal Justice System’s “Legitimacy”

Another long-term societal cost of some decriminalization may be reducing the justice system’s reputation for fair adjudication—what has been termed the system’s “legitimacy,” a related concept to the system’s moral credibility discussed in the previous Section. While moral credibility refers to the system’s reputation for just results, legitimacy refers to the system’s reputation for a fair adjudication process. Tom Tyler and other scholars have shown that a decline in the system’s reputation for fairness and professionalism in adjudication reduces compliance, deference, and acquiescence.²⁵³

Their research shows people are more likely to defer to the decisions of legal authorities when the decision-making process appears to lack bias, focuses on objective facts, recognizes citizens’ rights, and treats people with dignity.²⁵⁴ The authors also found that “[People] depend heavily upon their inferences about the intentions of the authority. . . . If the authorities are viewed as having acted out of a sincere and benevolent concern for those involved, people infer that the authorities’ actions were fair.”²⁵⁵

The justice system’s reputation for a fair and honest adjudication process can be undermined by de facto decriminalization of conduct the community sees as criminally condemnable, particularly de facto decriminalization of conduct committed by certain political or ideological groups undertaken because the criminal justice officials share the perceived ideological view of the offenders. As noted previously,²⁵⁶ prosecutors have at times chosen to decriminalize conduct only for certain groups, as with the selective decriminalization of rioting in 2020, but this appears to violate fairness in adjudication and can do significant damage to the justice system’s legitimacy, at least among those members of the public who are not part of the favored group. This is likely reflected by surveys showing Republican confidence in the justice system fell sharply after 2020. In June 2020, 38% of Republicans had “a [g]reat deal” or “[q]uite a lot” of confidence in the justice system.²⁵⁷ By June 2021, just 17% did.²⁵⁸ Notably, the share of Democrats having “a [g]reat deal” or “[q]uite a lot” of confidence in the justice system barely changed,²⁵⁹ suggesting the fall in Republican confidence was not simply due to a change in presidential administration. Donald Trump had also not been indicted by June 2021,²⁶⁰ meaning a likely culprit for the decline in conservative confidence was the perceived politically motivated decriminalization of rioting in 2020.

The costs of such lost legitimacy should not be ignored, regardless of one’s political persuasion. For example, many Democrats might feel unconcerned by prosecutors refusing to apply the law to protestors or rioters acting for a “good cause” such as racial justice, but they should recognize that such a decision makes it more likely that those on the political right will feel justified in rioting or supporting criminality perceived as advancing a rightwing cause. In fact, one might reasonably speculate that the disturbingly high amount of support shown for January 6th rioters among conservatives is likely due in no small part to the justice system’s legitimacy being degraded. The attitude of “if they can do it, why shouldn’t we” is a destructive but inevitable consequence of biased criminalization and punishment decisions.

More generally, de facto decriminalization policies implemented by prosecutors can undermine system legitimacy by replacing the consistent and impartial application of the law with the personal beliefs of the prosecutor. Even if a citizen disagrees with the criminal law, they can at least respect its impartial application across a jurisdiction as being in some sense “fair.” However, when prosecutorial decriminalization leads to a patchwork criminalization scheme dependent on the personal or political beliefs of powerful individuals, this sense of a fair process is undermined. Upholding the “rule of law” as opposed to the “rule of the individual” is important for maintaining the system’s legitimacy.

The larger point is that when the criminal law fails to do justice as the community sees it, the result is numerous serious societal costs, ranging from eliminating deterrence and incapacitation for the specific conduct to increasing crime generally through eroding the moral credibility and legitimacy of the law. These costs must be included in any cost-benefit calculation surrounding decriminalization, but they are too often ignored by elitist policymakers unconvinced or unaware of basic facts of human nature and basic principles of criminal law. The fact that policymakers ignore these costs is all the more unconscionable when one realizes, as the next Section shows, that the costs of increased crime are borne disproportionately by vulnerable minorities—not the privileged proponents of decriminalization far removed from the consequences of their policies. It is one thing to support decriminalization from the safety of an ivory tower or gated neighborhood—it is quite another to experience its effects in one’s daily life.

E.Societal Costs of Increased Crime Are Disproportionately Borne by Vulnerable Minorities

The increased crime sparked by decriminalizing condemnable conduct is all the more tragic and counterproductive to the aims of many decriminalization efforts due to its disparately large impact on racial minorities and the economically disadvantaged. This disparate impact plays out in several ways.

First, the violent crime rate is disproportionately higher in poor neighborhoods,²⁶¹ and the people who live in those areas are often racial minorities. Thus, the criminogenic effect of lost deterrence, incapacitation, moral credibility, and legitimacy is highest in these neighborhoods and disproportionately suffered by minorities. For example, several studies, including one by the Department of Justice, found that from 2008 through 2012, Americans living in households at or below the Federal Poverty Level (less than $15,000 for a couple) had more than double the rate of violent victimization as persons in higher-income households ($75,000 or more).²⁶² In 2020, one-third of violent crimes reported to police had a Black victim—almost twenty percentage points higher than the Black share of the population.²⁶³

But the difference in victimization rates is only part of the problem. The crime clearance and conviction rates are also lower, meaning that criminals are more likely to get away with their crimes, thus leading to more crime through failing to incapacitate dangerous offenders, failing to deter would-be offenders, and failing to support the public’s confidence in (and thus cooperation with) the justice system. Crime clearance rates are significantly lower in poorer areas with high racial minority populations than in White middle-income and high-income areas.²⁶⁴ For example, one analysis of fifty-two of the United States’s largest cities found that police arrested someone in 63% of homicides that killed White victims, compared with just 47% of homicides of Black victims, a sixteen-percentage-point difference in clearance rates.²⁶⁵ Data from Chicago indicates that homicide cases involving a White victim are solved 47% of the time, cases involving a Hispanic victim are solved 33% of the time, and cases involving a Black victim have a clearance rate of a mere 22%.²⁶⁶ In other words, poor and minority Americans are victimized at higher rates and with greater impunity than other Americans, thus leading to a vicious cycle of more crime and less justice.

De facto decriminalization policies that incentivize gun violence, such as Larry Krasner’s non-prosecution of many gun crimes in Philadelphia, thus hurt minorities the most. For example, over 75% of Philadelphia’s homicide victims are Black, and Black Philadelphians are about five times more likely to be victims of gun violence than White residents—meaning the cost of Krasner’s policies is borne largely through an increase in the unsolved murder of Black lives.²⁶⁷ If the slogan “Black Lives Matter” is to be taken seriously, then policymakers must consider the consequences, and not merely the intentions, of their decisions. Indeed, even those generally supportive of the progressive prosecutor movement have cautioned against ignoring the racial injustice of underenforcement (and thus a lack of protection) in minority communities.²⁶⁸

Additionally, specific decriminalization, such as decriminalizing theft, has a disproportionate impact on poor residents who may end up living in “food deserts” because the stores in the area have closed or because they are paying more for basic items due to the costs of increased shoplifting. While food deserts are largely unknown in wealthier neighborhoods, they can have devastating impacts on already disadvantaged neighborhoods. As Ruth Jones Nichols, president and CEO of a regional food bank in Virginia explains, Black people “will be disproportionately impacted by having this new food desert in our community, and we can’t ignore that reality. . . . This isn’t just a food access or social justice issue. It really is a racial justice issue.”²⁶⁹ While food deserts can have many causes, tolerating shoplifting is a significant contributing factor in many communities.²⁷⁰

The de facto decriminalization of rioting in 2020 by authorities who sought to show solidarity with the rioters’ perceived political cause had the actual effect of exposing minority neighborhoods and minority-owned stores to the greatest harm, as authorities allowed rioters to demand justice for George Floyd by day while looting and burning Black-owned businesses at night.²⁷¹

F.Balancing Societal Costs and Benefits in Decriminalizing Criminally Condemnable Conduct

As Parts I and III document, the effect of decriminalizing conduct viewed as criminal by the community has not been a reduction in such conduct but rather a proliferation of the damage caused by it. Nor is this simply a matter of a few mistakes being made in implementation: the theory behind replacing criminal punishment with treatment ignores the human demand for justice documented previously.²⁷² Ignoring that demand will lead to the entire criminal law’s moral credibility eroding, thus replacing any rehabilitated offenders with new offenders inspired to commit crimes by the breakdown in respect for the law, in addition to crime increases caused by the loss of deterrence. An easy thought experiment illustrates the point: Imagine murderers could be rehabilitated with 100% effectiveness after undergoing one therapy session. Would replacing current criminal punishments for murder with this mandatory therapy session reduce murder or not? The answer is obviously no. While there would be no repeat murderers, far more people would be inspired to murder from both the loss of deterrence and the sense that the criminal law no longer saw murder as conduct that deserves punishment. Indeed, general lawbreaking would likely rise as people perceived the criminal law as being outrageously out-of-touch with justice (if the law thinks murder deserves therapy, why not steal—it’s not even as bad!). Of course, the results in the real world are even worse: most therapeutic and rehabilitative interventions for criminal offenders have dismal success rates.²⁷³ As a result, replacing criminal punishment with therapeutic intervention will not even eliminate repeat offenders, but it will increase crime through decreasing deterrence and the law’s credibility. These facts explain why there has never been a successful abolition of criminal punishment in any human society, despite a number of small experimental societies attempting such a “progressive” reform.²⁷⁴

Another claimed benefit of decriminalization contrary to community views is increasing equity by reducing the punishment of poor, minority, or other disadvantaged offenders.²⁷⁵ However, as the previous Section shows, the costs of the resulting increased crime fall disproportionately on disadvantaged communities (such as poor people and racial minorities), thus actively decreasing equity.²⁷⁶ Some proponents of decriminalization seem to make the bizarre, and implicitly racist, assumption that minority communities are best-served by prioritizing the well-being of minority offenders over the well-being of innocent minority victims.²⁷⁷ But, for example, what is best for “the Black community” is not necessarily best for “Black criminals” despite some activists seeming to equate the two in a deeply disturbing and offensive manner. Most crime is intraracial, meaning that refusing to punish minority offenders for the sake of “equity” is actually refusing to grant justice to minority crime victims and protection to innocent minority communities.²⁷⁸

When one weighs the costs of decriminalization contrary to community views against the proposed benefits, the balance of interests clearly favors adhering to the community’s criminalization judgements. If decriminalization really is the correct course of action for society, then reformers ought to actively work to persuade the community that the conduct in question ought not be seen as criminal—rather than trying to force that view upon them with all its societal costs.

IV. Equal Application of these Principles to the Reverse Situations: Demanding Broader Defenses and Mitigations and the Decriminalization of Conduct Not Seen as Condemnable by the Community

The principles discussed in Part III need to be applied broadly to the full range of reverse situations to ensure the criminal law tracks the community’s nuanced justice judgements. As Section IV.A below explains, criminalizing conduct in line with community judgements must be supplemented by recognizing sufficiently broad defenses and mitigations reflective of community views, which the current criminal law does not do. On the other hand, as Section IV.B discusses, there are limits to the appropriate breadth of excuse defenses, and those limits do not extend to such things as recognizing a “rotten social background” defense or similar such ideologically based defenses, as some have argued for.

The need for a broad application of the principles in Part III also applies to a different kind of reverse case. While it is essential for society to avoid decriminalizing conduct seen as criminally condemnable, it is equally true that society should stop criminalizing conduct that the community no longer sees as condemnable, a problem discussed in Section IV.C.

A.Enlarging Defenses and Mitigations to Match Community Views

One reason some well-intentioned people might support decriminalization contrary to community views is they see it as a utilitarian way to avoid injustices where the system over-punishes.²⁷⁹ Such thinking may see decriminalization leading to some failures of justice but avoiding enough injustices to make it worth it. It is certainly true that if one is serious about the central importance of doing justice, one must be as careful to avoid injustice as to do justice. But instead of blanket decriminalization contrary to community views, it makes more sense to pursue expanding defenses and mitigations to avoid injustices under the current criminalization. Just as the criminal law must not decriminalize conduct that the community thinks is criminally condemnable, it must also strive to recognize a full range of defenses and mitigations where the community’s shared intuitions of justice would hold an offender blameless, or less blameworthy, for their criminalized conduct. Empirical research suggests that current law falls short in this respect with regard to a wide variety of criminal law doctrines.²⁸⁰

First, consider the matter of formal legal defenses. As one of us has detailed elsewhere, modern American criminal codes based upon the American Law Institute’s Model Penal Code have appropriately broadened the formulation of justification and excuse defenses beyond those recognized at common law.²⁸¹ Unfortunately, a quarter of the states sat out the modern recodification movement of the 1970s and 1980s and therefore have “pre-modern” criminal codes that commonly are inadequate in recognizing a full range of general defenses.²⁸² And even many Model Penal Code jurisdictions have rejected the Code’s modern formulations in favor of retaining the common law rules.²⁸³ We would urge a reconsideration of these decisions that leave the formulation of general defenses too narrow to fully avoid unjust punishment under the community’s intuitions regarding exculpation.

While fully adopting the Model Penal Code’s defenses would be a step in the right direction, even the Model Penal Code fails to fully appreciate the breadth and nuance of ordinary people’s assessment of what defenses may remove the blameworthiness for criminal conduct. For example, consider community views on three kinds of situations in which an individual ought to have a justification for the use of force: when the force is used in self-defense, in defense of property, or to apprehend a person fleeing from a crime he has committed. Studies show that the community judges that these justifications are more compelling than what legal codes are willing to grant.²⁸⁴ Respondents frequently assign no liability in cases to which the Model Penal Code attaches liability. Even when respondents assign liability, they typically assign considerably less punishment than would be suggested by criminal codes.²⁸⁵

Similarly, research has shown that ordinary people would support a significant expansion of disability excuse defenses, far beyond what the current law allows.²⁸⁶ The disability caused by mental incapacity is recognized by people as a valid reason to exculpate an offender, with people formulating mental incapacity in a way that recognizes both a control and cognitive deficit; people seem to judge that dysfunction of either type is a valid trigger for exculpation. The greater the dysfunction, the greater the likelihood of a defense. And people seem satisfied with a substantial rather than a complete dysfunction as establishing exculpation.²⁸⁷ These results suggest states should adopt a considerably broader insanity defense than is recognized by most criminal codes in the U.S. today. The same is true for the excuse defense of involuntary intoxication. It is commonly analyzed in legal codes on similar principles to the ones used for the analysis of mental illness defenses, and ordinary people agree that equating the two is generally appropriate.²⁸⁸ But since criminal codes’ current recognition of mental-illness-based defenses is more limited than public intuitions,²⁸⁹ the same is true of the involuntary intoxication defense.

While some progressive activists mistakenly assume that relying upon community views would necessarily produce an unforgiving draconian criminal law,²⁹⁰ the empirical evidence suggests otherwise, as illustrated by the above discussion of justification and excuse defenses. There is clearly significant room for expanding criminal defenses to match public views, thus adding more nuance to current criminalization and punishment practices without generating the failures of justice inherent in blanket decriminalization or anti-criminal justice policies.

B.The Proper Limits of Exculpation: Rejecting a “Rotten Social Background” Defense

While the criminal law ought to embody a far-reaching and nuanced assessment of an offender’s blameworthiness, there are important limits to doctrines of exculpation. As with criminalization-decriminalization decisions, legal defenses and mitigations must reflect the community’s shared intuitions of justice rather than the ideological preferences of a minority, as seems to be the case, for example, in proposals to recognize an excuse defense for having a “rotten social background.”²⁹¹

It should be no surprise that many of the decriminalization motivations discussed in Part II were at work in the proposal and promotion of this defense. While some academics and activists may continue to support recognition of such a defense,²⁹² no legislature or judicial system has yet done so, probably due to fear of triggering the destructive societal costs, reviewed in Part III, of generating increased crime and lost credibility in the eyes of the community. A system that recognizes a “rotten social background” defense would quickly sacrifice its moral legitimacy with the community and lose its ability to harness the powerful forces of social influence, to say nothing of the loss of deterrence and incapacitation such a defense would cause.

Traditional excuse theory, reflected in the excuses of essentially all American jurisdictions, provides an excuse upon the satisfaction of two requirements: (1) a disability must cause (2) a recognized excusing condition.²⁹³ In the excuses of insanity and involuntary intoxication, for example, (1) a mental disease or defect or involuntary intoxication must cause (2) the offender at the time of the offense to suffer sufficient cognitive or control dysfunction so as to be unable to sufficiently understand or control their offense conduct.²⁹⁴ The first element, a disability, requires a real world abnormality or dysfunction. The second element, a resulting excusing condition, requires a normative judgment that the extent of the dysfunction is sufficiently severe to render the offender blameless for their conduct.²⁹⁵ To give another example, the duress excuse requires that the offender (1) committed the offense under the effects of duress or coercion and (2) that the coercion, in the words of the Model Penal Code, was such that “a person of reasonable firmness in [the offender’s] situation would have been unable to resist.”²⁹⁶

To obtain a defense, it is obviously inadequate to show simply that the offender was acting under some coercion or had some mental illness or was involuntary intoxicated to some extent. An excuse is merited only if the extent of the resulting dysfunction meets a critical level: such that the actor could not reasonably have been expected to have remained law-abiding. This is true even if the disability is a “but for” cause of the offense. The offender who steals perfume from the department store because his girlfriend threatened to break up with him if he did not may well be able to claim that he would not have committed the offense but for the coercion, yet no one would seriously consider that the effect of the coercion was sufficient to merit an excuse. The same is true of the man who is involuntarily intoxicated by his friends and, in that state, has non-consensual intercourse with a semi-conscious woman. He may well be able to honestly claim that he would not have committed the crime but for his involuntary intoxication, but whether he is entitled to an excuse requires us to know more about the actual effects of the involuntary intoxication on him at the time of the offense conduct.²⁹⁷ Was the effect simply a pleasant buzz and a minor loosening of restraints? Could he reasonably have been expected to have avoided the offense conduct despite the involuntary intoxication? Even if the offender would not have committed the offense but for the disability, he does not necessarily merit an excuse. The extent of the dysfunction—the extent of cognitive or control impairment—must be shown to have been sufficiently severe as to render him blameless for his offense conduct.²⁹⁸

Before Judge Bazelon’s support for a “rotten social background” defense in the Alexander case, he supported the adoption of the Durham “product test” for insanity, under which an offender is to be given an insanity defense if he would not have committed the offense but for his mental illness.²⁹⁹ But that formulation of the insanity defense stands in direct conflict with the standard requirements for modern excuses by leaving off the second element for an excuse: the requirement that the disability (such as mental disease or defect) cause a sufficiently dysfunctional effect so as to render the offender blameless (as opposed to the disability merely being a contributing cause of the conduct). By dispensing with the excusing condition requirement, Bazelon’s theory of excuses would seem to provide a defense to the mildly coerced perfume thief and the involuntarily intoxicated rapist, a difficult position to sustain.

In United States v. Brawner, the D.C. Circuit rejected Bazelon’s Durham arguments, for predictable reasons similar to those above.³⁰⁰ The blamelessness that underpins an excuse defense does not derive from the bare existence of an actor’s disability or even from the fact that the disability may influence the actor toward the violation. The conclusion of blamelessness comes from the existence of the excusing condition—the finding that the effect of the disability on the actor was so severe that he could not reasonably be expected to have remained law-abiding.

Judge Bazelon repeated his error in United States v. Alexander, in which “rotten social background,” rather than mental illness, was said to have caused the offender’s offense conduct.³⁰¹ In that case, a group of unarmed uniformed marines had stopped at a hamburger shop. Alexander began staring at the marines, and when one of the marines returned the stare, Alexander approached and confronted them, leaning in and poking one in his name tag, calling them to come outside. When words were exchanged, Alexander and co-defendant Murdock each pulled a handgun and began firing into the group, killing two of the marines.³⁰²

Murdock’s claim was that he should be excused because his rotten social background (“RSB”) had caused his offense conduct.³⁰³ This is similarly the thrust of the rationale for an RSB excuse offered by some scholars:

An environment of extreme poverty and deprivation creates in individuals a propensity to commit crimes. In some cases, a defendant’s impoverished background so greatly determines his or her criminal behavior that we feel it unfair to punish the individual. . . . [B]lame is inappropriate when a defendant’s criminal behavior is caused by extrinsic factors beyond his or her control.³⁰⁴

There is good evidence that a rotten social background can have potentially significant effects on a person’s behavior. Murdock grew up nearly penniless in the violent, chaotic Watts section of Los Angeles. He was not responsible for being born into the situation. He had no obvious or easy means of escaping from it. On the other hand, most people who grow up in such circumstances do not become criminals.³⁰⁵ That is, even admitting that environment can seriously influence the development of personality, the strength of that influence to committing the offense at hand is at best unclear. But most importantly, even if it could be shown that Murdock’s rotten social background was a but-for cause of his shooting the marines, such a but-for cause by itself is insufficient for excuse, just as it is in the insanity (or duress or involuntary intoxication) situation. To qualify for an excuse—to be blameless for the offense—would require Alexander to show that the overwhelming effect of the RSB was such that he could not reasonably have been expected to not shoot the marines.

Indeed, one could argue the case in Alexander does not even meet Bazelon’s watered-down “product test” requirement, or at least not to an extent that any criminal justice system would want to use it as a basis for excuse. “Poverty causes crime” is a favorite mantra among many activists and academics, and it does seem to be wise crime-control policy to shape society to minimize rotten social backgrounds.³⁰⁶ But there is little empirical support for the proposition that a generally impoverished upbringing can itself cause a specific crime so as to render the offender blameless. There may be a general correlation between RSB and crime,³⁰⁷ but general correlation does not make specific causation.³⁰⁸

To illustrate the point, consider the strong correlation between being male and committing crime.³⁰⁹ If correlation sufficed to create an excuse, then the prime candidate for excuse under Bazelon’s “product test” would be being male, which has been shown to have an a very strong correlation to crime. A person with XY chromosomes is nearly four times more likely to commit a criminal offense and more than six times more likely to commit an offense of physical aggression and violence.³¹⁰ Not surprisingly, men are dramatically overrepresented in the prison population. They make up 84% of the country’s correctional population and are twelve times more likely to be convicted of a violent felony than women.³¹¹ Seventy-eight percent of all arrestees, 84% of all convicted felons, and 86% of all violent offenders are male.³¹²

The facts indicate that being male is a significant contributor to criminality, but would anyone really want to conclude that, given the correlation between being male and committing crime, we should provide a “being male” excuse defense? A vast number of male offenders can plausibly argue that they would not have committed their offense if they had been born female, so presumably Bazelon’s product-test theory of excuses would support a defense for them.

Of course, traditional excuse theory rejects any such defense. It is not enough that a defendant was abnormal in some way or in an abnormal situation at the time of the offense. To be excused, the law requires that the abnormality or the abnormal situation have such a dramatic effect in reducing the offender’s ability to remain law-abiding that they can no longer be seen as blameworthy for their offense.³¹³ There is little evidence that being male causes such sufficient cognitive or control dysfunction (although some women might dispute this point). Certainly, there is nothing to suggest a dysfunction so substantial that men cannot reasonably be expected to remain law-abiding. When it comes to a rotten social background, social variables that suggest a RSB—growing up in a broken home, inadequate care or child abuse, early socialization to violence—actually have a much more limited association with criminality than chromosomes do.³¹⁴ If we should not give a defense for “being male,” there is even less reason to consider a defense for having a rotten social background.

Recognizing an RSB defense might well serve the ideological goal of replacing criminal law with an anti-criminal justice social policy, one of the motivations for decriminalization described in Part II, but this would be wholly inconsistent with the notion of deserved punishment to which the community is deeply devoted.³¹⁵ The same activists who support RSB-type defenses are also likely to support decriminalization as a way to avoid “criminalizing poverty.” The overlap in motivations for an RSB defense and for decriminalization was exemplified by the previously mentioned 2020 proposal from Seattle’s city council to create a “poverty defense” that would prohibit “prosecution for misdemeanor crimes for any citizen who suffers from poverty, homelessness, addiction, or mental illness.”³¹⁶ While the proposal was fortunately not implemented due to outcry from even liberal Seattleites, it illustrates how the motivations behind many of today’s decriminalization efforts are a logical parallel to recognizing an RSB defense—ditching the criminal justice system’s focus on doing justice in favor of social engineering through converting crime into a purely public health issue no matter how egregious the conflict with the community’s commitment to the importance of justice.³¹⁷

C.The Need to Decriminalize Conduct That the Community No Longer Sees as Criminally Condemnable

We have argued that there are significant and often ignored societal costs to decriminalizing conduct that the community sees as criminally condemnable. For many of the same reasons, the criminal law has an obligation to officially decriminalize conduct whenever community views shift to view that conduct as no longer sufficiently condemnable to merit criminal prohibition and punishment.

To take an obvious example of an outdated criminalization, adultery is still an offense on the books in some states.³¹⁸ The 2003 conviction of one John R. Bushey, Jr. under Virginia’s adultery statute generated national attention in large part because such prosecutions were no longer ever brought.³¹⁹ (Bushey was sentenced to community service and the misdemeanor charge was expunged from his record upon his completion of the program).³²⁰ The newsworthiness of the case itself highlights the fact that adultery has been de facto decriminalized even in states with an existing adultery statute.

Some people may see this situation—criminalization without enforcement—as having some practical value. They may argue that, while adultery is rarely prosecuted, its continuing presence in the criminal code provides an official source of moral condemnation of the conduct.³²¹ Decriminalizing adultery, they might worry, could be seen as morally approving it. And few legislators probably want to be on record as the “adultery legalizer,” lest the change suggest the legislator approves of such conduct.

But criminalization-without-enforcement creates a range of problems, many of which echo some of the concerns around decriminalization discussed previously. First, the failure to decriminalize mistakes the proper role of criminal law. There are numerous social institutions that have as their role building or reinforcing social norms—religious institutions are a prime example and more recently social media and special interest groups—but criminal law’s role is more limited. It seeks to identify the wrongdoing that is so condemnable as to deserve the stigma and punishment of criminal liability. As noted previously,³²² there are many sorts of conduct that may be socially and even morally objectionable, but just because the conduct is wrong, it does not follow that its wrongfulness rises to the level of criminality.

To criminalize adultery when the community sees it as insufficiently serious to deserve actual prosecution and punishment is to dilute the power of the “criminal” label. By representing adultery to be a criminal offense, when in practice it really is not seen as meeting those requirements, is to risk people concluding that other offenses in the criminal code are similarly only pretend offenses. Creating that kind of ambiguity undermines the social influence of criminal prohibitions generally.

Further, the criminalization of conduct that is not really meant by the legislature to be prosecuted has the effect of shifting the criminalization power to prosecutors and risking arbitrary enforcement. The rule of law is undermined when the criminal law includes a variety of outdated offenses that can be dusted off and used to prosecute a disfavored defendant according to the whims of the prosecutor.

To protect its moral credibility, as well as its legitimacy based upon fairness in adjudication, a criminal law ought to say what it means and mean what it says. It ought to carefully define as criminal only the conduct that the community believes deserves the sanction of criminal liability and punishment and, when such conduct occurs, it ought to prosecute violations to the extent feasible, unless some special circumstance of the particular case justifies non-prosecution.

Perhaps a more relevant example of the need to decriminalize conduct no longer seen as criminally condemnable is the case of marijuana. Recent surveys have confirmed that a clear majority of Americans support marijuana legalization. For example, a 2022 poll showed that 59% of Americans supported complete marijuana legalization, another 30% supported making marijuana legal for medical use only, and only 10% supported the total marijuana criminalization still common in many states.³²³ As with adultery, some lawmakers are reluctant to decriminalize marijuana lest they be seen as encouraging marijuana use, but such concerns should be outweighed by the loss of credibility the law suffers by being out of step with community views. Lawmakers concerned about marijuana as a gateway drug should be more concerned about marijuana as a possible gateway crime that accustoms people to breaking the law generally.³²⁴ This is also a possible lesson of Prohibition, discussed earlier,³²⁵ in which creating a comfort level with lawbreaking through drinking alcohol may have been one of the reasons crime rates for non-alcohol offenses increased.

That said, policymakers should be cautious approaching decriminalization purely on the basis of opinion polls that can fail to capture the community’s true and nuanced criminalization and justice judgements. The result of polling is too often the result of how questions are asked. Further, polling may not reflect the community’s judgment about whether specific conduct is criminally condemnable but rather whether the community is open to an alternative approach to eliminating that criminal conduct—if the alternative approach works. For example, while polls show an increasing share of Americans seem open to “hard” drug decriminalization, this does not necessarily mean their criminalization judgements are changing in the same way as with marijuana.³²⁶ As discussed earlier regarding Oregon’s Measure 110,³²⁷ people may be willing to experiment with decriminalizing conduct they still see as criminally condemnable under the belief it will reduce the conduct in question. When the conduct is not reduced (a seemingly predictable result), people revert to desiring criminalization because they still view the underlying conduct as criminal. By contrast, increasing marijuana use in states that have legalized marijuana has not caused voters to demand reinstating criminalization—perhaps because they now see marijuana use as more akin to drinking or smoking—unhealthy, but not criminally condemnable. The best test for whether the community truly no longer sees the conduct as criminally condemnable is whether they would still support decriminalization if they knew the conduct would significantly increase post-decriminalization. The public seems by-and-large accepting of weed brownies and pot-smokers proliferating, but the prospect of fentanyl cookies and heroin addicts coming to a street near you is likely to face rejection.

Decriminalization advocates would also do well to remember how nuanced public judgements can be—the same behavior can be viewed very differently depending on context. There is a reason sex in the bedroom is legal but sex on the street is not. The community might see private drug use as not being a matter for the criminal law so long as that drug use does not significantly affect the rest of society. At the same time, the community might very strongly believe public drug use deserves criminalization and strict enforcement.

An additional distinction to consider is that the public may not so much support decriminalization as a change in punishment. For example, California voters who passed Proposition 47 may have been expressing a desire to allow for more non-prison punishments for theft—not a desire to let theft go unpunished. Sometimes a change in the statutory punishment, instead of de facto decriminalization, is what best reflects community views. We strongly support efforts to revise laws to reflect changing community views about what constitutes a just punishment for criminal conduct, but the need for such reform should not be used as an excuse to enact unsupported decriminalization.

While policymakers must decriminalize conduct that is no longer seen as criminally condemnable to avoid damaging the law’s moral credibility, they must also recognize that the law’s credibility is equally imperiled by enacting decriminalization contrary to the community’s nuanced criminalization judgements. Respecting the community’s views is the only way to avoid both errors.

Conclusion

The increasing trend toward de facto decriminalization of condemnable conduct is driven in large part by a failure to appreciate the importance of doing justice according to the community’s shared justice judgments. Those pressing for such decriminalization are often those who oppose punishment generally and who imagine a society in which condemnable conduct is “corrected” through medical and social interventions instead. No matter how beautiful this vision may sound, it is dangerous because of its disregard for human nature. There never has been and never will be an orderly society without criminal punishment.³²⁸ What thoughtful reformers can do is promote a society that is extremely careful about punishing only in strict proportion to an offender’s blameworthiness. This includes making sure only conduct the community views as criminally condemnable is criminalized and that a full set of defenses and mitigations representative of society’s justice judgements are recognized. Additionally, as we have argued elsewhere,³²⁹ reformers can do much to reduce imprisonment in favor of non-prison sanctions the community will still see as just, as well as making sure the imprisonment that is imposed is as effective as possible in rehabilitating offenders, be it through education, training, treatment, or anything else that proves effective. As should be clear, this Article has no quarrel with a desire to change or reduce punishments for particular crimes, but only with the desire to do so by decriminalization contrary to the community’s views.

We can and should do our best to eliminate both injustices and failures of justice from the legal system. But converting anti-criminal justice or anti-prison preferences into the decriminalization of condemnable conduct makes the system less just for all, ensures more crime, and inflicts more suffering on the most vulnerable. There is nothing just or equitable about well-intentioned policies with terrible results. They are just terrible policies. That is a lesson reformers must learn in order to avoid compounding the problems they try to solve.

98 S. Cal. L. Rev. 585

Download

Paul H. Robinson*

*Colin S. Diver Professor of Law, University of Pennsylvania.

Jeffrey Seaman†

†Bachelor of Arts in Philosophy, Politics, and Economics, 2022, and Master of Science in Behavioral and Decision Sciences, 2023, University of Pennsylvania. Levy Scholar at University of Pennsylvania Law School. Sarah M. Robinson made significant contributions to this Article, for which we thank her. We also thank Hannah Agarwal and Hugh Rennie for their assistance. The authors express their appreciation to the participants of programs in which this paper was presented at Heinrich-Heine-Universität Düsseldorf, Notre Dame Law School, and Vanderbilt Law School.

The Embodiment of Sovereignty: Outrages, Section 1983 Litigation, and the Federalism Revival

April 2025April 2025Norman W. Spaulding*

A remarkable number of canonical Section 1983 cases and many other less well-known civil rights cases involve extraordinary violence—brutality that cannot be described either as a bureaucratically rational, dispassionate expression of state power, or as marginally disproportionate in relation to the legally authorized level of force incident to the state’s monopoly on violence. They instead involve what the drafters of Section 1983 called “outrages”—extraordinary excess, a level of violence that is not merely disproportionate, but unbridled, not merely painful, but severe, degrading, abusive, extended, grotesque, gratuitous, orgiastic, carnivalesque, barbaric, and all too often fatal for Black people. What accounts for this cruelty (and the apparent pleasure taken in inflicting it) on the part of officers of the state? Why does it persist?

Reading across a range of interdisciplinary sources on extravagant violence, I argue that it involves the interaction of a number of elements: the fantasy of finding/generating legal certainty in the imposition of bodily suffering, the enactment of social and political domination (including controlling color and gender lines), the role of terror and physical abuse in relationships structured by dependency, and ambivalence in officers’ embodiment of sovereignty (officers who are both bound by and always potentially above the law). Extraordinary violence was also well known to the drafters of Section 1983. Unfortunately, the modern Court’s limitations on the Section 1983 cause of action and remedies betray the text of the statute and its unmistakable legislative purpose to provide redress for “outrages.”

INTRODUCTION

The exception explains the general and itself. . . . It brings everything to light more clearly than the general itself. After a while, one becomes disgusted with the endless talk about the general—there are exceptions. If they cannot be explained, then neither can the general be explained.

—Søren Kierkegaard¹

The landscape of Section 1983 precedent is littered with dead, mutilated bodies. It is saturated with blood. The children strewn across this landscape who are not dead themselves (shot in the back, even the face)² have been beaten, forcibly sterilized,³ and abused in other ways.⁴ Private homes have been ransacked,⁵ burst asunder.⁶ The houses of detention that dominate the landscape⁷ reek of fecal matter, the stench of burning flesh, pestilence, and countless other indignities.⁸ Time itself is seized and obliterated for the wrongfully convicted who linger for decades in prison.⁹

I refer here not to cases involving ordinary displays of the state’s monopoly on violence—the monopoly which political scientists and sociologists tell us must be held by the state and its law enforcement officers.¹⁰ Nor do I refer to the fact, insisted upon by Robert Cover, that while the exercise of judicial review may check the state’s monopoly on violence, it also rationalizes that violence.¹¹ “Legal interpretation,” Cover trenchantly observes, “takes place in a field of pain and death.”¹² Courts are implicated, inexorably, he insists, in the state’s monopoly on violence.¹³

These are not the cases to which I refer. However painful and even fatal the ordinary operation of law can be, the cases to which I refer do not involve bureaucratically rational, efficient, dispassionate expressions of state power.¹⁴ Nor are they cases in which brutality is marginally disproportionate in relation to the legally authorized level of force incident to the state’s monopoly on violence—cases, that is, in which rational, efficient, dispassionate expressions of state power happen to exceed the legal threshold of “excessive force” or “cruel and unusual punishment” or “discriminatory purpose.”¹⁵ These cases, tragic as they are, are instances of ordinary legal error, lying at the “sometimes ‘hazy border between excessive and acceptable force.’ ”¹⁶

The class of cases to which I refer involve what the drafters of Section 1983 called “outrages”—extraordinary excess. Violence that is not merely disproportionate, but unbridled, not merely painful, but severe, degrading, abusive, extended, grotesque, gratuitous, orgiastic, carnivalesque, barbaric, and all too often fatal.¹⁷ Less commonly observed is the fact that this form of violence is also very often deeply intimate, even erotic in the sense in which George Battaille understood erotic excess (a “transgression of borders [that] sets us in intimacy with the other”), though unlike the loving transgressions he describes, the “sovereignty” exercised in these cases is not in the service of “friendship, truth, compassion, and the welcoming of differences.”¹⁸ Difference is instead annihilated, and suffering is inflicted proximately, often by hand, far from conventional bureaucratic forms of action at a distance. Extraordinary excess is not merely intimate, but in some cases sadistic—perpetrated, that is, by officers who appear to take pleasure in the suffering they inflict.¹⁹ And while this violence can be brazenly public, a spectacle,²⁰ it is very often privately inflicted, sometimes withheld from public scrutiny by conspiracies of silence and deceit.²¹

What explains this form of state violence in Section 1983 cases, many of which are landmarks in the development of the law and procedure of civil rights enforcement? If James Monroe must be arrested on false suspicion of murder, why must his wife be made to stand naked in the living room during the arrest, his six children beaten?²² If Adolph Lyons has not perfectly complied with an officer’s request to raise his hands during a traffic stop, if he complains after his hands are slammed back to his head, why must he be placed in a chokehold until he passes out, “spitting up blood and dirt,” having “urinated and defecated” on himself when he regains consciousness?²³ If Larry Hope must be disciplined for falling asleep and arriving late to forced labor on a chain gang in a prison, why must he be handcuffed to a hitching post shirtless in the sun for seven hours?²⁴ Why must he be deprived of bathroom breaks? And why would a corrections officer carry water out to the hitching post only to taunt Mr. Hope by giving it to a prison dog and then spilling it on the ground before Mr. Hope in the sweltering heat? If Keith Hudson legally deserved punishment for getting into an altercation with prison guards, why, after he’d been restrained in handcuffs and shackles, was he repeatedly punched in the mouth, eyes, chest, and stomach while a guard held him down?²⁵ Why did the supervisor on duty not only watch the beating, but tell the officers involved with acidic irony “not to have too much fun”?²⁶ If Javid Iqbal must be detained on immigration charges in a maximum security facility on completely false suspicion of connections to the September 11th attacks, why did the officers kick him “ ‘in the stomach, punch[] him in the face, and drag[] him across’ his cell without justification”?²⁷ Why must he be subjected to “serial strip and body-cavity searches when he posed no safety risk to himself or others”?²⁸ Why must he be denied the right to pray and subjected to religious slurs?

Why are there so many replications of these cases?²⁹ And why are these outrages so often color-coded?³⁰

In some of the canonical cases, the Supreme Court has upheld a right to relief for the underlying constitutional violations. The experience of Mr. Monroe and his family underlies the Court’s landmark 1961 decision reviving Section 1983 as a federal cause of action from its Jim Crow dormancy. The Court upheld the use of Section 1983 for constitutional violations by state actors even when the plaintiff would presumably have a cause of action under state common law for at least some of the relevant harm.³¹ Larry Hope’s suffering underlies the Court’s determination that, for purposes of overcoming the affirmative defense of qualified immunity, some conduct is so obviously unlawful that there need not be a near-exact precedent on the books alerting the officer to this fact.³² Any reasonable officer would know the conduct was unlawful.

In other cases, however, the Court has denied relief and denied even the right to seek federal judicial review. These are the circumstances in Lyons, the 1983 case requiring that standing be separately established for injunctive relief (even when the plaintiff has standing for money damages) and requiring that the plaintiff show substantial risk of recurrence of harm.³³ The holding is part of the reason why deadly police chokeholds, used in the absence of proportionate physical threat to officers, remain so prevalent.³⁴ In the absence of standing to seek an injunction to ensure that police-use-of-force policies comply with the Fourth Amendment, use-of-force policies are difficult to challenge.³⁵ In Iqbal, the Court dismissed the complaint as implausible under Rule 12(b)(6).³⁶ Even after Inspector General reports confirmed Iqbal’s assertions that high level government officials were well aware of the conditions in the prison and did nothing about it, the Court dismissed suits by similarly mistreated detainees on the ground that there is no Bivens cause of action (the federal officer analogue to Section 1983) for such harm, and that there was no cognizable cause of action for a Section 1985 conspiracy.³⁷

But put to one side, for the moment, whether legal redress is or ought to be available for such conduct. And put to one side as well the simmering debate on the affirmative defense of qualified immunity when money damages are sought in civil rights cases,³⁸ the near universal indemnification of officers who are ultimately held liable,³⁹ and the uncertain fate of the Bivens cause of action for constitutional torts committed by federal officers.⁴⁰ Put aside what it means to act “under color” of law for the purposes of Section 1983 liability⁴¹ and what must be shown to obtain an injunction to reform police departments and prisons.⁴² Suspend all of these vitally important, hotly contested doctrinal questions. Pause to take stock of the extraordinary, dehumanizing level of violence inflicted at the hand of the sovereign, the repetition compulsion, the fetishization of excess. Consider how and why Trent Taylor was forced to spend four days in a cell

covered, nearly floor to ceiling, “ ‘in massive amounts’ of feces”: all over the floor, the ceiling, the window, the walls, and even “packed inside the water faucet.” Fearing that his food and water would be contaminated, Taylor did not eat or drink for nearly four days. Correctional officers then moved Taylor to a second, frigidly cold cell, which was equipped with only a clogged drain in the floor to dispose of bodily wastes. . . . Because the cell lacked a bunk, and because Taylor was confined without clothing, he was left to sleep naked in sewage.⁴³

What explains this level of mistreatment? Not just the wretched, “shockingly unsanitary” conditions (conditions, note, in which the corrections officers themselves must work as they move and supervise prisoners), but the taunting by the officer who, upon placing Taylor in the first cell, informed him that he was “going to have a long weekend,” and another officer who, upon placing Taylor in the second cell, said he “hoped Taylor would f***ing freeze.”⁴⁴

What accounts for this humiliation and cruelty (and the apparent pleasure taken in inflicting it) on the part of officers of the state?

No meaningful conversation about jurisdiction, appropriate forms of action, or remedies can begin without grappling with the nature and sources of these constitutional violations and the mutual dehumanization they cause. Even if there are necessary tragedies attendant to the state’s monopoly on violence, what explains this level of violent excess, depravity, and depredation?

The very first thing to say is that the harm cannot be understood if it is not accurately described, if the depth of the violence involved is diminished, obscured, or antiseptically elided.⁴⁵ The majority opinion in Lyons is an exercise in elision, rendering Mr. Lyons’ request for injunctive relief to prevent another such incident irrational, even paranoid, rather than a reasonable response to policies and practices regarding use of force that disproportionately affect Black drivers in Los Angeles.⁴⁶ The chokehold, the majority concedes, “caus[ed] damage to his larynx” and “render[ed] him unconscious,” and the majority mentions the evidence of more than a dozen deaths resulting from the use of chokeholds.⁴⁷ But it does so only in passing in a paragraph concerning whether the case was mooted by alterations in the city’s policies.⁴⁸

It is instead in Justice Marshall’s dissent that one learns that when Lyons “regained consciousness, he was lying face down on the ground, choking, gasping for air, and spitting up blood and dirt,” and that he had “urinated and defecated” on himself as a result of the chokehold.⁴⁹ We learn from Justice Marshall that Lyons was not charged with any crime (let alone a crime such as resisting arrest that might suggest the chokehold was a proportionate response). He was simply “issued a traffic citation [for a burnt out taillight] and released.”⁵⁰ And it is the dissent which cites record evidence that Black people made up seventy-five percent of deaths from chokeholds in a city in which they were just nine percent of the population.⁵¹ The majority simply ignored the racial elements of the case, as well as the statistical evidence showing that three-fourths of all police/citizen “altercations” involved use of a chokehold—data suggestive of a policy or practice of resorting to deadly force when officers are met with simple dissent.⁵² This colorblindness obscures record evidence of racially discriminatory policing.

Similarly, in Monroe, it is only from Justice Frankfurter’s dissent that we learn that Mr. Monroe had six children, all of whom were rousted along with his wife, that Mr. Monroe was struck several times by Detective Pape “with his flashlight, calling him [n-word] and ‘black boy,’ [and] . . . that other officers hit and kicked several of the children and pushed them to the floor.”⁵³ Here too, the element of race and its conjunction with unprovoked violence would be all but invisible absent the dissent’s more faithful recitation of the allegations of the complaint. So too the fact and nature of the violence visited upon Mr. Monroe’s children.

Nor can the nature or source of the harm be understood if excesses are dismissed as the unfortunate result of individual rogue officers. To be sure, not every constitutional violation springs from the orders of supervisors or from official policy, practice, or custom, and comparatively few in number involve extraordinary excess. Neither is it the case, however, that every violation springs from circumstantial contingencies, isolated error, or the pathological deviance of individual officers.⁵⁴ Whether or not one believes that supervisors and government agencies should be held legally accountable for the conduct of their officers (either on a respondeat superior theory or some more exacting standard such as the one developed from Monell v. Dep’t of Social Services and its progeny⁵⁵), bad apples cannot explain all cases.⁵⁶ Indeed, the theory is contradicted by cases such as Taylor and Iqbal in which supervisors are plainly involved, as well as a vast body of research in social psychology and organizational behavior concerning institutional failure and crimes of wrongful obedience.⁵⁷ Ordinary people of good will, the research shows, are capable of inflicting pain and suffering on others, including extraordinary levels of pain and suffering, when working in settings where harmful conduct is expected, rationalized, or ignored.⁵⁸

In the following pages I examine “outrages” in civil rights cases and the nature of such extraordinary excess. The first level of work is mainly negative, identifying in Part I what this excess is unlike—how distinctive it is even according to social and political theories that are explicitly concerned with the ways in which people performing seemingly ordinary bureaucratic tasks contribute to monstrous deeds and systems of subordination. Part II turns to theories that focus on direct participation in extraordinary, dehumanizing excess and subordination to explore what kind of legal officer is capable of inflicting such severe suffering, under what circumstances, what purposes and forms of power might be served, and what kind of legal subject is produced by inflicting suffering in this way. I concentrate on Franz Kafka’s short story The Penal Colony, which imagines an extravagantly painful form of capital punishment, Patricia Williams’s essay on the irrational fears that animate racial prejudice and police brutality, and studies on the nature of “intimate partner” violence.

Reading across these interdisciplinary sources on extravagant violence I suggest that outrages involve the interaction of a number of elements: the fantasy of finding/generating legal certainty in the imposition of bodily suffering, the enactment of social and political domination (including controlling color and gender lines), the role of terror and physical abuse in relationships defined by dependency, and ambivalence in officers’ embodiment of sovereignty (libidinally charged oscillation between self-restraint and extravagant excess on the part of officers of the state who are themselves both bound by and always potentially above the law).⁵⁹ In the final section, Part III, I turn from these theories of extraordinary violence to civil rights enforcement doctrine. I argue that extraordinary, racialized violence resulting from corruption of office was well known to the drafters of Section 1983, that the statute was specifically designed to remedy what the Reconstruction-era drafters called “outrages,” and that the modern Court’s use of federalism principles to limit the Section 1983 cause of action and remedies directly contradicts the text of the statute and this unmistakable legislative purpose.

The literature on the drafting of Section 1983 is extensive, but it concentrates mainly on whether the language “under color” of law was designed to restrict the cause of action to circumstances in which the federal right-holder would not have a parallel claim under state law. This approach is understandable in light of Justice Frankfurter’s emphasis on this reading of the statute in his dissenting opinion in Monroe v. Pape and the debate his opinion fueled.⁶⁰ What has not received adequate attention beyond conventional references to the role of Klan violence in the legislation of Section 1983 is the specific connection between the persistence of “outrages,” government officer complicity in such violence, and the failure of state courts and law enforcement to address the corruption of state legal systems arising from resistance to Reconstruction.⁶¹

Both Section 1983 scholarship and empirical analyses of officer brutality also tend to lump modern cases involving extraordinary violence together with all cases on excessive force and forms of prison violence that lack a legitimate penological purpose.⁶² The effect is to assimilate outrages to a broader category of state violence. However instructive this is regarding the overall problem of modern police/carceral violence, it tends to obscure the distinctive phenomenon of outrages, the unique harm they cause, and their role in prompting Congress to establish Section 1983 in the first place.

I. FALSE STARTS IN THEORIZING EXTRAORDINARY VIOLENCE

Modern social and political thought has been preoccupied with the irrational and dehumanizing effects of institutionalized violence. But it is notable that among the most authoritative accounts—Arendt’s theory of the banality of evil, Foucault’s theory of disciplinary power, and Kafka’s theory of perishing “before the law”—outrages are largely irrelevant.

A. The “Banality of Evil”

Section 1983 cases involving extraordinary violence do not seem to result from what Arendt famously described as the “banality of evil,” a theory that has been used as a framework for understanding how relatively ordinary people become instruments in the perpetration of monstrous deeds. Observing the trial of Adolph Eichmann for his role in organizing the transportation of Jewish people and others to Nazi death camps, Arendt was struck by the extent to which Eichmann was neither “perverted” nor “sadistic” but rather terrifyingly “normal.”⁶³ He apparently acted mainly to advance his bureaucratic position.

The deeds were monstrous, but the doer—at least the very effective one now on trial—was quite ordinary, commonplace, and neither demonic nor monstrous. . . . Despite all the efforts of the prosecution, everyone could see that this man was not a ‘monster,’ but it was difficult indeed not to suspect that he was a clown.⁶⁴

Arendt uses the term “clown” because Eichmann was unable to avoid self-contradiction, lying, and cliché in matters of morality. His “inability to speak,” Arendt emphasizes, “was closely connected with an inability to think; that is, to think from the standpoint of somebody else.”⁶⁵ Eichmann knew (“of course”) that “he had played a role in the extermination of the Jews,” that without performing this bureaucratic function Jewish people “would not have been delivered to the butcher.”⁶⁶ But he scoffed at the idea that this established his personal culpability (“What is there to admit?” he asked when pressed on the matter⁶⁷).

Chillingly, he “remembered the turning points in his own career” more accurately than “the turning points in the story of Jewish extermination, or, as a matter of fact, [] the turning points in history.”⁶⁸ The ideological machinery of fascism, Arendt contends, efficiently mobilized the petty self-serving idealism of bureaucrats like Eichmann, drawing them into the enthusiastic erection and maintenance of the machinery of death. Inability “to think” and dislocation of moral responsibility in the very seat of identity combined to produce a monstrous and yet banal form of evil.⁶⁹

Eichmann’s role in the machinery of death of the Holocaust was, however abhorrent, principally administrative in nature—distant from the scenes of murder. He orchestrated a process of stripping Jews of their property and rights to prepare them for expulsion from their communities to death camps. Arendt’s analysis reminds us that acts of violence have bureaucratic predicates in which many people are complicit, but the distinction between the acts and their bureaucratic predicates matters to a full understanding of extraordinary violence. Many people who are capable of performing roles at a distance from violence are utterly incapable of the violence itself. Eichmann, for example, testified that if he was shown “a gaping wound, I can’t possibly look at it. I am that type of person.”⁷⁰ Indeed, this is one of the most important insights of Arendt’s theory of the banality of evil—that people who recoil from direct acts of violence are instrumental to the administrative architecture which ultimately produces and sustains it. In the Section 1983 cases involving extraordinary violence, by contrast, the named officers are not acting at a distance—their hands are saturated with blood. Even when high level officers are also named, as in Iqbal and Abassi, the core allegation is that the policies and practices immediately responsible for the harm (not intermediate bureaucratic steps) flow from the ink of their pens and associated decisions.

B. “Disciplinary Power” and the “Spectacle of the Scaffold”

The exceptional violence in Section 1983 cases does not flow from “disciplinary power” either. In Discipline and Punish, Michel Foucault describes ways in which surveillance, ranking, and the arrangement of space in modern carceral institutions produces efficient compliance, or “docility”:

[T]he systems of punishment are . . . situated in a certain “political economy” of the body: even if they do not make use of violent or bloody punishment, even when they use “lenient” methods involving confinement or correction, it is always the body that is at issue—the body and its forces, their utility and their docility, their distribution and their submission.⁷¹

A panopticon, the famous prison design of Jeremy Bentham from which Foucault theorizes, produces docility among inmates mainly by creating uncertainty about whether and when one is being surveilled. Indeed, uncertainty about surveillance prompts inmates to monitor themselves, obviating the need for actual force to be deployed by prison guards. This was its supposedly modern genius as compared to the use of dungeons, torture, and spectacles of death on the scaffold. It achieves this by “induc[ing] in the inmate a state of conscious and permanent visibility that assures the automatic functioning of power,” such that

the surveillance is permanent in its effects, even if it is discontinuous in its action; that the perfection of power should tend to render its actual exercise unnecessary; that this architectural apparatus should be a machine for creating and sustaining a power relation independent of the person who exercises it; in short, that the inmates should be caught up in a power situation of which they are themselves the bearers.⁷²

Foucault connects the panopticon and other systems oriented toward the production of docility (in schools, factories, hospitals, military barracks) to “progressive forms of training,” measurement, ranking, and other “techniques for assuring the ordering of human multiplicities” in a “disciplinary society.”⁷³ Disciplinary power thus develops alongside mass society and the industrial revolution to ensure social order and efficient labor.⁷⁴ In “every society, the body was in the grip of very strict powers, which imposed on it constraints, prohibitions[,] or obligations.”⁷⁵

In all of these accounts, what makes disciplinary power distinctive is precisely the absence or minimal role of violence, its “subtle coercion” through the “efficiency of movements[,] . . . an uninterrupted, constant coercion, supervising the processes of activity,” which the subject eventually internalizes.⁷⁶ The extraordinary violence in Section 1983 cases is quite different. It involves direct, physical, unmediated infliction of pain. It is therefore closer to what Foucault calls “sovereign power.”⁷⁷ Throughout Discipline and Punish and in other works Foucault is careful to distinguish disciplinary and sovereign power. Sovereign power is what we conventionally associate with the state’s monopoly on violence—state action in the standard sense of armed officers acting in an official capacity. It is comparatively inefficient and discontinuous, requiring “sumptuous expenditure[s],” relative to the operation of disciplinary power.⁷⁸

Precisely because it is so inefficient, Foucault argues that sovereign power historically relied on spectacles of excess to produce compliance. These spectacles were designed to display the “unlimited power of the sovereign,”⁷⁹ “an emphatic affirmation of . . . its intrinsic superiority . . . not simply that of right, but . . . the physical strength of the sovereign bearing down upon” deviant subjects to “revenge[] its authority.”⁸⁰ As David Owens has shown, recorded and publicly disseminated images of anti-Black police violence fit the “spectacle of the scaffold” which Foucault associates with early modern European expressions of sovereign power.⁸¹ There is, however, no public scaffold when extraordinary violence is inflicted behind prison walls and in unrecorded or unpublished encounters with police officers, a problem to which I return in Part II.

C. Being “Before the Law”

In Franz Kafka’s famous parable, Before the Law,⁸² the protagonist is denied access to law by a low-ranking but “mighty” gatekeeper who says that he is but one of many gatekeepers, “each one mightier than the one before.”⁸³ The gatekeeper shares that “[e]ven the sight of the third is more than I can bear,” so terrible are his powers.⁸⁴ The gatekeeper is thus an example of sovereign power, a power apparently so awesome in its capacity for violence that no one dares resist it. Significantly, for our purposes, however, the gatekeeper’s capacity for violence is so overwhelming that it need not be exercised upon the body. The gatekeeper also informs the protagonist that the gate is for him alone, not others, suggesting that the gate is not public. The protagonist decides to wait for permission and the gatekeeper “gives him a stool and allows him to sit down beside the door.”⁸⁵ But the wait is unending, and ultimately fatal. The gatekeeper stands impassively as the protagonist gradually withers “before the law” and dies. In the meantime, the gatekeeper takes all the protagonist’s money, “so that you don’t think there’s something you’ve omitted to do.”⁸⁶

There are many readings of the parable as a critique of the terrible, dehumanizing force of law,⁸⁷ but the protagonist’s suffering, albeit unto death, is not caused by any act of violence on the part of the gatekeeper. His suffering is shaped instead by the possibility of violence, implied in the gatekeeper’s admonition about his awesome power and the even greater powers of other gatekeepers beyond the portal before which he stands. The moral outrage the story evokes lies in the gatekeeper’s startling indifference. It is suggestive of the thoughtless, cold, bureaucratic unseating of empathy with which Arendt is concerned, but not the extravagant infliction of physical suffering we see in Section 1983 cases.

II. THE BODY AND BEING ABOVE THE LAW

If targets of extraordinary violence are not “before the law,” subjected to disciplinary power, or victims of the “banality of evil” and its action at a distance, what explains their victimization?

A. In the Penal Colony

Another famous story by Franz Kafka takes us closer to understanding the phenomenon. In the Penal Colony describes an “apparatus” created by a former colonial commandant to suppress disorder in the colony.⁸⁸ The apparatus is a machine that “carr[ies] out . . . the sentence” by inscribing its text on the body of the condemned person with needles.⁸⁹ The condemned person is strapped to a “bed” above which hangs a “harrow”—a metal frame holding many needles.⁹⁰ The machine, called the “engraver,” moves the bed and the harrow, “vibrat[ing] both sideways and up and down, in tiny, very rapid movements” each of which is “very carefully calibrated,” inscribing the sentence in thrusts and cuts of gradually increasing depth.⁹¹ During the process, which takes twelve hours, the condemned person slowly experiences the sentence “on his body” as the thrusts and cuts deepen until he bleeds to death.⁹² The sentence is “put to him physically.”⁹³

Extraordinary excess is a defining feature of every phase of this punishment. The legal “process” that leads to punishment is itself excessively deferential to officer discretion. A condemned person receives no trial, no defense, and is not informed of the sentence before it is inscribed on their body. Guilt is decided by those who control the apparatus, and because there is no opportunity to present a defense, guilt “is always beyond doubt.”⁹⁴ A foreign observer is informed that in the process of inscription there are elaborate embellishments by the needles upon the body of the condemned: “many[,] many ornaments surround[ing] the script proper.”⁹⁵ The apparatus not only writes the punishment with increasingly deep incisions in the body of the condemned, it uses “the rest of the body” to surround the “actual text” with extravagant “decoration.”⁹⁶ The pain is equally extravagant. About halfway through the execution of the sentence, the traveler is told, the pain is so excruciating that the condemned person eventually stops screaming and falls silent as he “begins to decipher the script . . . with his wounds.”⁹⁷ The officer who operates the apparatus views this as the climactic moment when justice is realized, describing it as a “transfiguration” in which even “[t]he very dimmest of them begins to understand.”⁹⁸ Their faces take on a look “that might seduce one to take their place” in the apparatus, so transcendent is the truth they experience.⁹⁹

In many ways, In the Penal Colony represents the apotheosis of the spectacle of the scaffold. The officer, overcome with nostalgia, reports that during the reign of the former commandant, executions were public events with massive public attendance (“the whole valley was packed . . . everyone came to spectate . . . they all knew: justice is being enacted”) and the apparatus not only inscribed the sentence, its needles dripped with a painful, corrosive acid:

It was impossible to find room for all those who wanted to view the proceedings from close . . . . How we watched the transfiguration in the tormented faces [of the condemned], how we held our cheeks in the glow of this arduously achieved and already passing justice! I tell you, comrade, those were times!”¹⁰⁰

This is a genuine public spectacle, designed to demonstrate the awesome power of sovereignty, to secure compliance with law by making the administration of justice not only summary (no trial, no defense, no notice of the sentence) but immediate (operating directly upon the body of the condemned, enacting the sentence upon the body using the exquisitely painful machinery of the apparatus), absolute (every sentence is a death sentence), and yet also temporally and substantively extended (extending beyond the “actual text” of the sentence to lavish “decorative” embellishments, and beyond mere execution to excruciatingly slow torture ending in death), all resting on the sadistic, metaphysically sublime premise that the true nature of justice appears and is “understood” in and through the body during the inscription.¹⁰¹

Kafka implies that the apparatus is not, in fact, essential to maintain order in the colony even though the officer claims that under the former commandant who built it, administration of the entire colony was “seamlessly efficient.”¹⁰² We learn early in the story that the new commandant, much to the chagrin of the officer, regards the apparatus as barbaric and unnecessary, resists expenditure for its upkeep, has the support of the people, and has invited the traveler to observe the execution in order to use the traveler’s views to ban its use permanently. Indeed, much of the dramatic tension in the story rests not just on the design and use of the apparatus for what becomes its final execution, but on whether and how the traveler will announce his moral opposition to it to the new commandant (the traveler is, we gradually learn, steadfastly “opposed” to the entire process: “[t]he injustice of the procedure and the inhumanity of the execution were incontestable”¹⁰³). When the officer fails to persuade the traveler of the value of the apparatus, the officer spares the condemned man and instead inserts himself into the apparatus. The apparatus impales him almost immediately because it has not been possible to properly maintain it and he dies within minutes, leaving “no trace” that he has experienced the “promised transfiguration.”¹⁰⁴

The violence of the apparatus is of course in important ways quite different from extraordinary violence in Section 1983 cases. It is, to begin with, imaginary, and also a relic of public executions, whereas the violence in Section 1983 cases is all too real, and there is often no audience. The apparatus also mediates the relationship between the body of the condemned and the officer. The officer, at one point, has to probe deep into the apparatus to attempt a repair, but it is neither a cyborg nor simply an instrument of torture wielded by the officer. Once the instructions of the “sentence” are inserted into the apparatus and the condemned person is strapped onto the bed, the machine makes the inscription automatically. Finally, many Section 1983 cases involve deaths at the hands of officers of the state, but the cases generally do not involve executions following judgments imposing capital punishment as a sentence.¹⁰⁵ Instead, when deadly force is used, officers are usually in the process of mundane traffic stops, executing warrants and making arrests, or engaged in prison administration away from death row. And of course, many Section 1983 cases involving extraordinary violence and suffering do not result in death.

The point of connection between the story and Section 1983 cases lies instead in its account of the officer’s attachment to the apparatus as an instrument of excess (even to private scenes of excess after public attendance has been banned), and to the seductive power of the summary administration of justice. The officer who maintains the apparatus doesn’t just appreciate it; he is obsessively attached to it. By the time the traveler comes to see it, the officer is “its sole defender.”¹⁰⁶ While he barely notices the condemned man, the officer is distraught that the new commandant has withheld funds for maintenance of the apparatus, he laments that there are no longer proper replacement parts, and he is particularly distressed that a more humane execution procedure is being developed.¹⁰⁷ He refers to the diagrams that typeset the apparatus as “the most precious things I have.”¹⁰⁸ He also desperately seeks to draw the traveler into a conspiracy to undermine the new commandant and his plan to decommission the apparatus. The completeness of his obsession with the machine is revealed by his fatal devotion to it—once he realizes that it will no longer be used and that his efforts to persuade the traveler of its value are in vain, the officer makes his own execution the machine’s last, a process that results in the machine falling apart. His devotion is, one might say, not merely complete but fatally excessive.

Underneath devotion to the apparatus is attachment to the power to inflict bodily pain in the administration of justice—an attachment grounded in the idea that infliction of bodily pain serves as a lexical index of justice, that pain is an instrument of justice, the way to see justice done. This idea animates the officer’s sadism and the erotic charge traveling with the physical imposition of legal judgment—literally, the inscription of sovereign power upon the body of the subject. Similar attachments appear to motivate the officers in Section 1983 cases who take pleasure in the suffering they inflict (e.g., the officer who hoped Taylor would “f***ing freeze” in the “frigidly cold cell . . . equipped with only a clogged drain in the floor to dispose of bodily wastes”¹⁰⁹), as well as chokehold cases like Lyons in which the body of the officer is (must be) personally, immediately, and intimately engaged with the exercise of sovereign power, inscribing it upon the body of the supposedly resistant subject.

Attachment is not just to embodying the instrument of sovereign power, but to legal certainty—inscribing the law on the subject’s body. A legal order that seeks perfect compliance (along with the erotic charge of potentially ‘complete’ repression) is a form of excess in relation to human fallibility. In Kafka’s story, there is a cathartic element in the dyadic economy of a suffering/redeemed/dead “perpetrator” and suffering/relieved victims and bystanders. Indeterminacy, the terror of it, is executed, purged from the community, along with the condemned. Thus, even as other more restrained forms of policing and punishment emerge in the penal colony, the imagined moral clarity and instrumental utility of the apparatus, and the attendant desire to preserve it and all the accoutrement of its scaffolding, to speak its language of certainty, to inscribe it automatically upon the bodies of deviant subjects, remains. And this remainder—the longing it prompts for revival of such a system—haunts new, ostensibly more humane and fallible regimes. This remainder is also latent in the “thin blue line” discourse of law and order insofar as it rationalizes police and carceral violence—what “law and order” promises (the absence it mobilizes politically and in the psyche) is the certainty of determinate exercises of sovereign power and the compliance such certainty ostensibly guarantees.¹¹⁰ Embellished, gratuitous, and grotesque bodily infliction of pain reflects this longing for compliance and legal certainty. Bodies must be fixed, transfixed, in compliant recognition of sovereign power.

The officer in the story is so committed to this work that he appears to have lost all sense of connection to the dignity and humanity of the subjects of the colony he participates in governing, not least of which the condemned man. His almost exclusive concern is with maintaining and caring for the apparatus. Indeed, he exhibits paranoia about the new commandant’s plans and is unable to cope with a future in which those plans will unfold.¹¹¹ Even at lower levels of intensity, sociologies of policing suggest that perspective and judgment are warped when officers succumb to an embattled, binary, “blue line” mindset in which the people one serves become a foreign, abstract out-group.¹¹² Ordinary emotional responses become grotesquely inverted—instead of revulsion, pleasure is taken in inflicting pain; instead of empathy and a sense of duty toward subjects, there is antipathy or brutally cold indifference; instead of care, there is the belief that suffering is not just deserved but in the best interests of the subject’s legal “transfiguration.”¹¹³

The officer’s obsession in Kafka’s story is not just nostalgically motivated, it is an expression of “imperialist nostalgia”—seething resentment at the new regime and its lenience is expressed in inverted form as intimate attachment to the apparatus believed to be central to the authority of the ancien regime over the colony.¹¹⁴ John Zilcosky admonishes that it is a mistake to separate the erotic element of the apparatus’s sadistic excess from its value as an instrument of colonial oppression. “The problem with [some] postcolonial readings” of In the Penal Colony, he insists, “is that they fail to address the promise of sado-masochistic pleasure that is so central to the story’s effect. Just as the earlier psychoanalytical interpretations repressed politics, these political readings repress desire.”¹¹⁵ The officer’s obsessive attachment (being willing indeed, not just to die for the apparatus, but with it, in it) fuses the erotic and politically instrumental in the form of a death wish executed by fatal penetration (as the apparatus malfunctions, one of the needles plunges through the officer’s skull). Obsessive attachment is in this way represented as an erotically charged and (at least until the officer’s death) a politically useful byproduct of a system that scaffolded spectacles of sadistic excess and permitted officers to impose summary justice.

In any system in which extravagant sovereign power has been replaced with a new regime of limits, especially externally imposed limits, resentment can run deep. In the American context, one might, for instance, read resistance to Reconstruction enforcement in the nineteenth century and resistance to the Warren Court’s revival of Section 1983 along these lines—new, decidedly unwelcome federal (and federal judicial) interventions into matters of traditional state and local executive branch concern in the administration of criminal justice.¹¹⁶

B. “Spirit-Murder”

Kafka leaves the role of racial subordination in spectacular violence implied in In the Penal Colony. Early on in the story the traveler remarks that the officer’s uniform is “much too heavy for the tropics.”¹¹⁷ The officer replies that they wear them nonetheless because they “signify home, and we don’t want to lose touch with home.”¹¹⁸ In this sentence, in another about the fact that both the traveler and officer speak French while the condemned man and attending soldier do not,¹¹⁹ and in the officer’s reference to the traveler’s moralistic “European perspective” on capital punishment,¹²⁰ a geography of colonial and racial exploitation is made clear.

In one of her most famous essays on law and policing in America, Patricia Williams is explicit, directly addressing the role of extraordinary violence in racial subordination.¹²¹ She examines the failed criminal prosecution of a white police officer who shot and killed a sixty-seven-year-old, arthritic, Black woman in her Bronx apartment.¹²² The officers were there to evict Eleanor Bumpurs and she resisted.¹²³ The writ of possession authorizing the eviction was issued via a default judgment, which means that Mrs. Bumpurs never received a hearing in court before the officers showed up to evict her. Indeed, it appears that she was not properly notified of the case and there were “serious doubt[s]” about the validity of the landlord’s grounds for eviction.¹²⁴ This summary process without notice may have played a role in her confusion and resistance.

According to the Police Commissioner, six officers were in the apartment, all armed, two wielding large plastic shields, another a “restraining hook.”¹²⁵ Mrs. Bumpurs, wielding a knife “bent” in the confrontation by one of the plastic shields, escaped from the hook and was shot by the officer farthest from her with a shotgun.¹²⁶ The blast “removed half of her hand, so that, according to the Bronx District Attorney’s Office, ‘it was anatomically impossible for her to hold the knife.’ ” The officer nevertheless “pumped his gun and shot again,” this time killing her.¹²⁷

She asks what “animus . . . inspired such fear, and such impatient contempt in a police officer that the presence of six other heavily armed men could not allay”:

Why was the sight of a knife-wielding woman so fearfully offensive to a shotgun-wielding policeman that he felt that blowing her to pieces was the only recourse, the only way to preserve his physical integrity? What offensive spirit of his past experience raised her presence to the level of a physical menace beyond real dimensions? What spirit of prejudgment and of prejudice provided him with such a powerful hallucinogen?¹²⁸

The essay goes on to explore the ways in which racism authorizes not just violence, but extraordinarily excessive anti-Black violence. She considers examples in which both white police officers and white citizens were exonerated in the press and in court for such violence. Echoing Arendt’s critique of Eichmann’s inability to think and the penal colony officer’s perverse certainty that torture produces transfiguration, Williams ultimately concludes that white supremacy creates distorted mental states (“formalized distortions of thought” that “produce[] social structures centered around fear and hate”) in which “in the minds of whites[,] . . . blacks become large, threatening, powerful, ubiquitous, and supernatural.”¹²⁹ The Black body must not only be subdued, its far larger spectral presence which operates “beyond real dimensions” must be annihilated.

The connection between extravagant violence and “embellished fear” in this process of annihilation comes into sharpest relief when Williams recites “a news story from [her] fragmentary grammar school recollections of the 1960’s,” according to which

a white man acting out of racial motives killed a black man who was working for some civil rights organization or cause. The man was stabbed thirty-nine times, a number which prompted a radio commentator to observe that the point was not just the murder, but something beyond. What indeed was the point, if not murder? I wondered what it was that would not die, which could not be killed by the fourth, fifth, or even tenth knife blow; what sort of thing that would not die with the body but lived on in the mind of the murderer.¹³⁰

Williams’s answer is that white supremacy demands the death of the “spirit”—that part of our being which lies within and yet extends beyond the body, the part of ourselves which “is beyond the control of pure physical will and resides in the sanctuary of those around us.”¹³¹ “Spirit-murder” must be exceptionally excessive to meet the seemingly bottomless fear and rage that animate prejudice. It requires a level of violence capable of producing degradation, of destroying the specter of Black power which lurks (however counter-factually) as a “predatory” threat in the aggressor’s mind.¹³²

Williams situates spirit-murder in the long, painful legacies of spectacular violence visited upon Black people. She interweaves examples of modern racism in New York, including her own experiences as a Black woman, with the civil rights era. And like Kafka, who opens the story of the penal colony after the old commandant who designed the apparatus has been replaced, Williams represents contemporary spectacles of racial violence as haunting remainders of an “unwanted past,” an ancien regime that will not end:

Failure to resolve the dilemma of racial violence merely displaces its power. The legacy of killing finds its way into cultural expectations, archetypes, and ‘isms.’ The echoes of both dead and deadly others acquire a hallucinatory quality; their voices speak of an unwanted past, but also reflect for us images of the future. Today’s world condemns those voices as superstitious and paranoid. Neglected, they speak from the shadows of such inattention, in garbles and growls, in the tongues of the damned and the insane.¹³³

The historian Allen Trelease formally locates this “unwanted past” in slavery and resistance to emancipation. He richly details the excessive, “pathological fear . . . not at all justified by the actual danger” felt by whites who defended the institution.¹³⁴ Both before and after emancipation “whites were apt to react with disproportionate severity” to Black people.¹³⁵ But emancipation and the “advent” of Reconstruction produced “enhanced anxiety[,] . . . intolerance[,] and violence as never before.”¹³⁶ Violent repression, Trelease emphasizes, was directed less toward ensuring that Black labor remained a “docile” labor force (“in fact [racial terror] did much more to drive off labor and disrupt economic life”), than toward “keeping blacks within their proper social bounds,” especially preventing voting, officeholding, and crossing social color lines.¹³⁷

If Negroes had been feared even when they were subjected to the control of masters and prewar slave codes, they were infinitely more dangerous now that these controls were lifted. . . . A former Mississippi Klansman justified violence as the only way of protecting white families against what he conceived to be “the threatened and rising arrogance of the negroes.” Cruelty, seemingly beyond the gravity of the provocation, he said, was “justified by the fact that every little insolence, if left unnoticed, would be bragged about by its perpetrator and fellow observers . . . . The news would spread with great rapidity, and there was no telling where it would end. . . .” By this reasoning white men could and did commit the most brutal forms of aggression, convinced all the while that they were acting defensively.¹³⁸

As early as 1865, the year the Civil War ended,

Carl Schurz reported . . . [that] “the maiming and killing of colored men seems to be looked upon by many as one of those venial offenses which must be forgiven to the outraged feelings of a wronged and robbed people.” . . . Certainly whipping and corporal punishment were regarded as the white man’s right and duty, emancipation or no emancipation; organized regulators or vigilantes took up this task . . . and the Klan further institutionalized the practice,” adding other forms of torture and death to the list of “atrocities.”¹³⁹

Page after page of the 1871 U.S. Senate Report and Testimony on the Alleged Outrages in the Southern States, which supported the passage of Section 1983, confirms the extraordinary level of violence that followed. Even a cursory review of the reported outrages in Alamance County, North Carolina between 1868 and 1870, indicates the distinctive role of extraordinary violence.

[A] negro man by the name of Outlaw . . . was taken from his house . . . about one o’clock at night, by a band of from eighty to a hundred men, and hung upon an elm tree, not very far from the court-house door. . . . The newspapers have said that he was guilty of having shot at a band of Ku-Klux that passed through the town some time previous; but that was not true.

Caswell Holt, (colored,) taken from his house in the night, was whipped until blood was streaming from his back; he was then made to bend down, while one of the parties tortured him by rubbing a rough stick up and down his back.

Joseph Harvey, (colored,) taken by fifteen or twenty in disguise, carried about three quarters of a mile from his house, dragged over fences, rocks, through briars, and about one hundred and fifty lashes laid upon his bare back.

Nathan Trollinger, (colored,) taken from his house by men in disguise, severely whipped, and afterwards made to mutilate his own private parts with his pocket knife.

Sally Hall and her two daughters thrown out of their house and whipped, and one of them made to exhibit her person, while the fiends proceeded to inflict blows upon her private parts.

John Bass, (colored,) “potterized.”¹⁴⁰

The Klan also targeted whites who supported Black people and Reconstruction:

Alonzo B. Corliss, a northern man, who was teaching [in] a colored school . . . was dragged from his bed by twelve or fifteen disguised men, amid the shrieks and cries of his distressed wife, carried a little distance into the woods and severely whipped, the hair and whiskers shaven from one side of his head and face, his head and face cut and disfigured in a most cruel manner, after which he was painted black and left in this condition.¹⁴¹

The “ultimate purpose” was to “overthrow the reconstruction acts down South.”¹⁴² In some places “a majority of men” in the Klan had served in the Confederate Army.¹⁴³ Having lost the war, they were now refighting with the tools of racial terror.

Not only did white supremacist “aggression [take] every conceivable form,” it was seldom punished; “it simply was not a punishable offense in the eyes of many whites to attack or even kill a Negro.”¹⁴⁴ In many regions of the South, “civil authorities, like the general white public, were either in sympathy . . . or intimidated.”¹⁴⁵ Lynchings were regularly conducted in front of courthouses to reinforce the message that the Klan controlled and was above the law.¹⁴⁶ It was also common for law enforcement officers, courts, the press, and the public to blame Reconstruction “radicalism and negroism, which in the south are one and the same thing,” rather than the white perpetrators of violence.¹⁴⁷ This was true even when law officers and judges became targets of Klan violence for their support of Reconstruction or their efforts to restrain the Klan.¹⁴⁸ Local Klan members perjured themselves as witnesses and violated their oaths as jurors in order to protect each other from prosecution.¹⁴⁹ Others who gave testimony were attacked.¹⁵⁰ In Gullford County, North Carolina, a lawyer who was asked whether the laws were administered “properly and thoroughly” answered that

where parties, for instance, are concerned in outrages committed by the order of which we have been speaking, my observation is that there are no convictions. I know that people have been whipped and scourged, and some have been killed; but the guilty have never been brought to punishment. Of course, the courts are the proper tribunals to award the punishment, but they have never been able to do it in any of those cases. . . . An alibi is proved in nearly every case . . . without any difficulty.¹⁵¹

Lawyers who did agree to become involved in the prosecution of Klan members were notified in writing that “they would be visited with punishment” unless they “ceased.”¹⁵² Some prosecutors not only abandoned their cases but felt “compelled” to move to other states.¹⁵³ Other officers collaborated with the Klan. The mayor of a town in Alamanee County, North Carolina testified that the “sheriff was connected with [the Klan], and every deputy . . . . [T]hey always hang around the courts to get on the jury . . . . [Y]ou cannot convict them.”¹⁵⁴ Impunity in the infliction of extraordinary violence went hand in hand with the complicity of state law enforcement and judicial officers, and functional immunity from liability in the rare cases brought against officers due to judicial bias, witness intimidation, and jury tampering.¹⁵⁵

When Southern resistance prevailed and the retreat from Reconstruction became national policy later in the 1870s, anti-Black violence continued. Lynchings morphed from nighttime affairs into full blown spectacles—mass events in broad daylight, publicized in newsprint and on the radio.¹⁵⁶ Experts on the history of lynching emphasize that even though “Black men and women were much more likely to become victims of personal assault, murder, or rape than lynching, and . . . withstood all sorts of injuries and insults on a daily basis[,] . . . lynching held a singular psychological force.”¹⁵⁷ It “came to stand as the primary representation of racial injustice and oppression as a whole,” assuming “this tremendous symbolic power precisely because it was extraordinary and, by its very nature, public and visually sensational.”¹⁵⁸

I have elsewhere written about the complicity of law enforcement and other legal professionals in lynching, including spectacle lynchings.¹⁵⁹ Others have documented in vivid detail how crime statistics and policing became instruments of Jim Crow segregation as unreconstructed whites seized control of the levers of sovereign power within and well beyond the South.¹⁶⁰ Inscription of the color line on Black bodies was charged not only with racial anxieties but also the erotic, most obviously in the exceptionally violent retaliation against Black men for real and perceived interactions with white women, in the frequency with which anti-Black violence was both actually and symbolically castrating, and in both the sexualization and sexual assault of Black women. The erotic is also present in the proximity, intimacy and raw physical effort required to administer whippings and other common forms of white supremacist torture, as well as in the gruesome acquisition of body parts and photographs of naked, mutilated black bodies as “souvenirs” of spectacle lynchings.¹⁶¹ A gendered expression of sovereign power is present in the invasion of Mrs. Bumpurs’s apartment as well—the officers’ penetration without notice or consent of Mrs. Bumpurs’s home, and her bent-edged, arthritic resistance met by their wildly disproportionate, ejaculatory, fatal, violence.

C. Disappointed Dominion

“[A]ny account of violence that cannot explain the strike, the blow, the act of sexual violence (including rape), or that fails to understand the way violence can work in the intimate dyad or the face-to-face encounter, fails descriptively, and analytically, to clarify what violence is.”¹⁶² A responsible theory of violence, including extravagant violence, must therefore grapple with rape and other forms of intimate partner violence.¹⁶³ While this kind of violence may initially seem quite far afield from cases involving extraordinary violence in policing and prison administration, feminist and queer theory have long identified connections between patriarchal power, the “cage of sex . . . in which all other [cages] are enclosed,”¹⁶⁴ “the family as a male-dominant institution[,] and male-dominant sexuality as the deus ex machina of male power across society.”¹⁶⁵ Gendered and sexual violence share with the extraordinary violence of policing and prison administration not only elements of excess, but also the erotic charge of intimately inflicted physical harm (patriarchal power in society and politics is, on this view, “eroticized power”¹⁶⁶) and, significantly, the targeting of the body as a principal object and site of subjection, control, and domination.¹⁶⁷

With respect to excess, women and gender non-conforming people are more likely to suffer extreme violence. They are more likely to be murdered and more likely to suffer physical injury that requires medical treatment from intimate partner sexual and physical assaults than heterosexual men.¹⁶⁸ Although near fatal and fatal injuries are relatively rare as compared to scratches, bruises, and welts in intimate partner abuse, serious injuries include lacerations from knives, broken bones, dislocated joints, head and spinal cord injuries, burns, broken teeth, and brain injuries from being knocked unconscious.¹⁶⁹ Intimate partner violence is pervasive in the United States and other western countries, an “epidemic” affecting millions annually despite hard won reforms since the 1970s that have replaced a permissive legal and cultural regime (in which the “corrective” battery of domestic dependents was lawful and considered a private “family matter”) with criminal prohibitions, networks of women’s shelters, and other support.¹⁷⁰

Qualitative research emphasizes that intimate partner violence is usually repeated, that “almost all” people targeted “suffer[] some physical violence as a part of the abuse, and while this was often described as ‘not the worst part’ of the abuse in comparison to less visible aspects, it invoke[s] a great deal of fear.”¹⁷¹ Even though it is common for abuse to become “routine” in intimate partner violence, “incidents of physical violence remain a shock, often experienced as coming ‘out of the blue.’ Interviewees describe both physical shock[]and a feeling of betrayal of the expectation[s] of trust and care in intimate relationships . . . as the hardest to come to terms with.”¹⁷² This kind of abuse is not “something you ‘get used to,’ ” despite the adoption of “skills of precaution and management to try and maintain personal security for themselves and their children.”¹⁷³ As one interviewee in a qualitative study reported, her partner “would just blow . . . . [Y]ou stand there and you can’t believe, you cannot believe that this is happening.”¹⁷⁴

Extraordinary excess lies not just in the gendered nature of such violence—the fact that greater, more harmful, and sexualized violence is visited upon women and gender non-conforming persons by men—but in the underlying breach of trust. Intimate relationships in which violence should be absent, governed by care, devotion, affection, and affirmation of personal autonomy, are perverted into degrading, abusive sites of domination—haunted by foundational acts of physical violence and the ever-looming threat of escalation. Intimate knowledge of a partner’s desires, fears, and vulnerabilities is itself abused—physically, sexually, emotionally, and psychologically. Children are often made hostages and objects of violent escalation to amplify fear and power.

There are competing theoretical accounts of how and on precisely what terms masculine “dominance, potentiated and centrally entitled and expressed sexually” in intimate partner violence translates to social and political domination.¹⁷⁵ But no one doubts the singular terror and harm this violence creates in intimate relationships. There is also broad consensus that intimate partner violence is “fundamentally a product of gender inequality[,] . . . a powerful tool in enforcing and reinforcing gender hierarchies and gender inequalities.”¹⁷⁶ Revealingly, what often sets such violence in motion is “the perception (by men) that women are trying to act ‘equal,’ that they have lost sight of their ‘proper place.’ ”¹⁷⁷ In this respect, the parallels to racialized violence are unmistakable:

[J]ust as Ku Klux Klan violence arose during the Reconstruction period in the South to enforce racial boundaries threatened by the emancipation of millions . . . to remind African Americans of “their place”—so, too, gendered violence can be understood as a tactic for reminding women of their place.¹⁷⁸

It occurs in intimate relations precisely because “the need to enforce between-group boundaries is greatest when there is a high degree of intimacy between group members.”¹⁷⁹

The body is so often the target because it represents, synecdochally, all the capacities of subordinate subjects to move, press, and rise beyond “their place.” Once we see the body—its movement, capacity, and crucially, its placement—as a site of control in social hierarchy, deeper connections between gender, race, sexuality, and class unfold. Abuse committed in intimate relations might be read as a distorted private, compensatory expression of both real and (mis)perceived public affronts to the “masculinity” of aggressors (for example, discrimination, economic exploitation, political disenfranchisement). Abuse against people believed to be lower status provides for the assertion of dominion in forms that are unavailable, lost, or threatened in society, work, and politics.¹⁸⁰ Rather than patriarchy radiating outward from the sexual politics of the family, on this account dominion both retreats and amplifies inward, displacing public slights, disappointments, and defeats (which, again, may be real or merely perceived) onto the comparatively safer targets of vulnerable domestic dependents and intimate relations in the private sphere.

There are vast differences between the domestic scene, on the one hand, and police and prison encounters, on the other. But what they share, at least when victims are unarmed or restrained, is the relation of dependence, intimate bodily proximity, status hierarchy, and overweening power. This may help explain why even restrained or otherwise non-threatening, unarmed arrestees and prisoners—people whose lives are literally in the hands of officers—nevertheless become objects of extraordinary violence.¹⁸¹ The problem is not, or at least not always, “embellished fear” or paranoia on the part of the officer, but rather acute awareness of the target’s relative helplessness and vulnerability,¹⁸² the target’s assumed culpability (social and moral, if not legal),¹⁸³ the reception of the target’s assertion of their own humanity, interests, or rights as audacious defiance of the officer’s status and authority,¹⁸⁴ and the opportunity this presents to exercise dominion believed to have been diminished or lost in other ways (through the officer’s subordination to the hierarchy of command, dealing respectfully with other citizens, suspects, and prisoners who have been openly defiant of the officer’s authority, or other social encounters).¹⁸⁵ Put differently, extraordinary excess can arise from displaced aggression, just as it can arise from exaggerated, misplaced fear.¹⁸⁶ And both can be personally and politically useful in keeping others in their place.

As importantly, aggression can be more than a form of compensatory displacement, more than a proportionate reaction to the perceived disappointment of dominion in other relations. It can also be an erotically charged, transgressive embodiment of sovereign power itself—raw, unchecked desire for dominion. To be sovereign is both to be the source of the law and, at least in democratic societies, to be bound by law—bound not just by training, and the law itself, but by oath. Every officer, and certainly those armed with the implements of sovereign power, inhabits and must mediate this contradiction. Having invested energy upholding compliance with law, oneself and for others, sometimes at great personal peril, extraordinary violence holds the seductive appeal of being, however transiently, above the law, of experiencing this singularly ecstatic form of sovereignty (the rapture of “sole and despotic dominion”) in and through the body of another, seizing and inscribing upon it the dictates of one’s will. The perverse pleasure expressed by some officers in the infliction of suffering and the violent fury with which all but total submission is greeted may derive from this source, the desire to embody unbound sovereign power.

III. THE BODY POLITIC

The foregoing analysis draws into relief why Section 1983 liability for extraordinary violence matters. Cases involving licentiousness among those who hold the state’s monopoly on violence are among the most important for the exercise of judicial review in a free society. Little is more destructive of constitutional faith in a republican form of government, where sovereignty ultimately rests with the people, than unconstitutional violence committed “under color” of law by those who wield and embody powers delegated by the people.¹⁸⁷ Self-restraint regarding lawfully delegated authority legitimates possession of the implements of sovereign power. Cases involving extraordinary violence represent the antithesis of self-restraint—not mere errant, or debatable uses of force, but flagrant disregard for the constraints of the Constitution on state action.

The Reconstruction Congress knew this. Whatever else may be said about ambiguities in the text and legislative history of Section 1983, regarding, for example, the availability and scope of municipal liability, or the power to reach private actors, one thing is beyond doubt: The statute operates in derogation of state sovereignty to enforce powers conferred by Section 1 of the Fourteenth Amendment in order to rectify and prevent “outrages” by state officers.¹⁸⁸ Although legislative debates on the Ku Klux Klan Act of 1871 are relatively sparse and opaque on some interpretive issues, “the debates contain extended discussion of conditions in the South, the breakdown of law and order, the acquiescence of Southern authorities, Klan terrorism against Blacks and republicans, and the consequent need for federal action.”¹⁸⁹ The civil cause of action created by the statute addresses the role of state actors in these “outrages.” The operative language (“deprivation of any [federal] rights, privileges, or immunities” by persons acting “under color of” state law, “custom, or usage”¹⁹⁰) sweeps beyond outrages, but officer-involved outrages were foundational, animating wrongs for which the statute provides civil liability. Suits seeking relief for the infliction of such exceptional harm thus warrant particular judicial solicitude.

The text of Section 1983 as enacted by Congress is unequivocal that the new federal cause of action operates in derogation of state sovereignty.

[A]ny person who, under color of any law, statute, ordinance, regulation, custom, or usage of any State, shall subject, or cause to be subjected, any person within the jurisdiction of the United States to the deprivation of any rights, privileges, or immunities secured by the Constitution of the United States, shall, any such law, statute, ordinance, regulation, custom, or usage of the State to the contrary notwithstanding, be liable to the party injured in any action at law, suit in equity, or other proper proceeding for redress . . . .¹⁹¹

In view of the text, legislative purpose, and historical context, the modern Court’s reliance on federalism as a ground norm to cabin Section 1983 claims (in cases involving standing to prevent recurrence of harm,¹⁹² officer immunity in damages suits,¹⁹³ municipal liability,¹⁹⁴ abstention,¹⁹⁵ class actions,¹⁹⁶ summary judgment,¹⁹⁷ the standard for equitable relief,¹⁹⁸ the interaction of habeas and 1983 damages claims,¹⁹⁹ and the substantive Fourth Amendment doctrine of reasonable force²⁰⁰) is the jurisprudential equivalent of lost cause ideology. It relies on the very interpretive tools used to end Reconstruction, return to “home rule,” and place the anti-Black violence of Jim Crow segregation beyond the reach of federal courts.²⁰¹

Two attributes of lost cause historical consciousness are particularly relevant. The first is the idea that the Civil War and Reconstruction did not alter antebellum understandings of state sovereignty—that the exertions of federal power necessary to save the Union did not diminish the sovereignty of the states. On this view, return to “home rule,” which the Court actively supported in the 1870s and 1880s by, among other things, striking down other sections of the Ku Klux Klan Act,²⁰² confirmed that antebellum federalism principles survived the Civil War. The fact that return to home rule was instrumental to Jim Crow segregation is frequently elided by the modern Court when it invokes federalism principles. Justice Frankfurter’s dissent in Monroe is emblematic, insisting that

[t]he jurisdiction which Article III of the Constitution conferred on the national judiciary reflected the assumption that the state courts, not the federal courts, would remain the primary guardians of that fundamental security of person and property which the long evolution of the common law had secured to one individual as against other individuals. The Fourteenth Amendment did not alter this basic aspect of our federalism.²⁰³

The effect is to limit the powers conferred by the Reconstruction Amendments to remedy civil rights violations by relying on structural constitutional principles that betray the letter and spirit of these amendments, principles affirmed in precedents that blinked at and emboldened extralegal, violent resistance to Reconstruction. The Civil War and Reconstruction did alter basic aspects of “our federalism,” expanding federal court jurisdiction and with it access to the expertise, “experience, solicitude, and hope of uniformity that a federal forum offers on federal issues.”²⁰⁴ But instead of solicitude, civil rights litigants all too often find federal courthouses inhospitable.

The second attribute of lost cause ideology is the view that the Reconstruction Congress overreached by legislating rights and remedies on the premise that states could not be trusted to redress federal constitutional violations.²⁰⁵ Section 1983 is itself excessive on this view, a remainder of Reconstruction and the exceptionally violent resistance that defined it. The Reconstruction Congress legislated no sunset clause in Section 1983. But that has not stopped the modern Court from treating the cause of action like an anachronism, an unwelcome remainder of a state of exception—one of the few statutory enforcement tools that was not struck down by the Court in its retreat from Reconstruction. Within this interpretive horizon, embellished fears about excessive liability and excessive interference with officer discretion and local control become predominant concerns, rather than state sanctioned violence and defiance of federal law. The modern Court has in this way developed a jurisprudence of excess in Section 1983 cases, just not one concerned with the state officer outrages against which the Reconstruction Congress legislated.

This is most evident in qualified immunity doctrine, where concern with excessive liability and judicial interference with executive officer discretion dwarfs concern with whether officers have complied with the Constitution and other laws.²⁰⁶ But it can be seen in other doctrines,²⁰⁷ and it resonates with Justice Frankfurter’s view in Monroe that the statute was intended to provide a cause of action only when state law or custom authorizes unconstitutional acts—that is, in the presumptively rare situation in which a state openly defies federal law and state common law provides no redress.

The modern Court has functionally endorsed Frankfurter’s dissent not by revisiting the interpretation of “under color” of law and overturning Monroe, but rather by expanding the affirmative defense of qualified immunity in officer suits, elevating the causation threshold for municipal liability, and limiting access to injunctive relief that could prevent recurring violations. Justice Frankfurter may have lost the battle of statutory interpretation in Monroe, but the modern Court has ensured that he won the war on the use of federalism to limit Section 1983. The result is expansive discretion in officer decisions about the use of force and inconsistency even in cases involving extraordinary violence. Tellingly, the Court has applied the Hope v. Pelzer exception to qualified immunity in only one case since 2002 when Hope was decided, it has emphasized that Hope’s exception for obvious unconstitutionality applies only in “rare” cases, and Hope has featured most prominently in opinions dissenting from the grant of qualified immunity.²⁰⁸ Thus, the Court has been relatively passive with a doctrine well-suited to address extraordinary violence. In the context of prison violence, the Eighth Amendment standard of liability for the use of force is explicitly tied to “malicious” and “sadistic” violence,²⁰⁹ but the test requires evidence of the officer’s subjective state of mind, and it allows officers to avoid liability for extraordinary violence skirting the line of malicious and sadistic harm. Even when the line is crossed, prisoners face far greater procedural obstacles than victims of police violence raising Fourth Amendment claims.

These obstacles exist because the only area in which Congress has expressly endorsed the Court’s subordination of Section 1983’s remedial purposes to federalism is in state prison conditions cases.²¹⁰ Passed during the most expansive period of mass incarceration²¹¹ when prison litigation reached one-fifth of the total number of civil cases filed in federal courts,²¹² the Prison Litigation Reform Act of 1995 (“PLRA”) includes an administrative exhaustion doctrine requiring strict compliance with state grievance procedures, an unwaivable filing fee even for prisoners who meet the in forma pauperis standard, a physical injury rule for money damage claims, and a right-remedy nexus requirement that sharply circumscribes federal equity power.²¹³ The effects on prison litigation have been palpable.²¹⁴ Immediately after the statute passed, decades-old continuing decrees in prison cases were vacated.²¹⁵

In Brown v. Plata,²¹⁶ decided fifteen years after the statute took effect, the persistence of breathtakingly cruel, unremedied unconstitutional conditions in California state prisons prompted the Court to uphold federal equity power notwithstanding the strict limits of the PLRA. The case is a parable of the Court’s jurisprudence of excess. Prison conditions in Plata shock the conscience, rivaling those which existed in state prisons under the so-called “hands-off” doctrine prior to Cooper v. Pate when federal courts simply refused to entertain prison conditions claims on federalism grounds.²¹⁷ As a result of their woefully inadequate mental health care system, California’s prisons had a suicide rate “80% higher than . . . for prison populations” nationally.²¹⁸ Data showed that “72.1% of suicides [in California prisons] involved ‘some measure of inadequate assessment, treatment, or intervention.’ ”²¹⁹ Shocking, systemic deficiencies in physical health care (inadequate facilities, doctors, nurses, and delays of months for acute conditions and “extreme departures from the standard of care”) caused “a preventable or possibly preventable death . . . once every five to six days.”²²⁰ Beyond preventable deaths, prisoners “suffering from severe but not life-threatening conditions[] experienced prolonged illness and unnecessary pain.”²²¹ The record is an intricately detailed catalogue of outrageous, arbitrary suffering.

However, to justify upholding the release of tens of thousands of prisoners notwithstanding the PLRA’s strict right-remedy nexus,²²² the majority could not just rely on the extensive record of suffering. The decree ordered release of healthy prisoners as well as unwell prisoners whose mistreatment did not violate the Eighth Amendment’s guarantee of minimally adequate medical care under Estelle v. Gamble.²²³ To uphold such sweeping relief, the majority adopted a prophylactic theory of the substantive right. The majority reasoned that the health care system was so dysfunctional that there was a “substantial risk” of prisoners suffering a violation of their rights under Estelle.²²⁴ Releasing prisoners (and moving some to local jails) would increase the state’s capacity to adequately address the health needs of those who remained.

Justice Scalia’s dissent sharply criticized this end run around the statute and its federalism principles. He insisted that Estelle claims exist only when individuals who already have serious medical conditions do not receive treatment because of the state’s deliberate indifference.²²⁵ Thus, no injunction to release prisoners merely at risk of an Estelle violation could be described as “narrowly drawn, . . . extend[ing] no further than necessary to correct the violation” or the “least intrusive means.”²²⁶ Moreover, Rule 23 provides no right to aggregate resolution of such claims given the particularity of each prisoner’s medical condition and their inadequate treatment.²²⁷ Decisions about how to rectify the health care problems in the prison system must remain with state prison officials. The lower court’s prison release order displacing local administrative discretion was “perhaps the most radical injunction issued by a court in our Nation’s history,” a “structural injunction” that deems an “entire” state-run prison “system . . . unconstitutional because it may produce constitutional violations.”²²⁸

Two years earlier, in Haywood v. Drown, the Court overturned parallel state legislative efforts in New York to eliminate Section 1983 money damage claims against corrections officers filed in state court.²²⁹ On a legislative record asserting that such claims were predominantly frivolous, the state stripped its courts of jurisdiction for both Section 1983 and state civil rights claims seeking money damages.²³⁰ Prisoners could still sue for damages, but only against the state itself, not the officer responsible for the harm, only in a separate court, not the state’s courts of general jurisdiction, and only subject to a range of procedural and remedial limitations that do not apply to the Section 1983 cause of action.²³¹ The Supreme Court struck down the law on the ground that it violated the Supremacy Clause by denying federal right holders access to a congressionally legislated cause of action, forum, and relief to enforce constitutional rights.²³² Like Plata, the case suggests a commitment on the part of the Roberts Court to preserve jurisdiction and remedial power notwithstanding the limitations Congress imposed on prison condition litigation in the PLRA. However, both Plata and Haywood were decided by bare 5–4 majorities, and as in Plata, the dissent in Haywood rested on federalism principles—specifically, the autonomy of states to control state court dockets by refusing jurisdiction of unwelcome federal claims.

The crucial swing vote in both cases was provided by Justice Kennedy. That fragile majority is sure to be tested in a reconfigured Roberts Court more committed to the dissents’ federalism principles. Notice the effects on prison conditions litigation if Plata and Haywood were to be reversed: Federal courts would be virtually closed to claims for injunctive relief under the PLRA and modern standing doctrine; strict officer immunity and municipal liability doctrines already render both state and federal courts inhospitable to Section 1983 money damage claims; and the dissent’s view in Haywood would permit state courts to simply refuse jurisdiction of Section 1983 prisoner claims. Prisoners would be left to state law claims in state court.

This outcome would be consistent with a central premise, arguably the central premise of federalism-based limitations on Section 1983 claims: that leaving federal constitutional right holders to state positive or common law is adequate to protect their federal constitutional rights. This was the structural constitutional core of Frankfurter’s dissent in Monroe. The conventional form of the argument is that when a state officer defends a state law claim (for example, a state common law claim for negligence rather than a federal claim for excessive use of force), the federal constitutional right will arise if the officer asserts that the action was sanctioned by state law.²³³ The defense can be rebutted by showing that what the officer did was, in view of the plaintiff’s federal constitutional rights, ultra vires. Federal law is then adjudicated, just not via a cause of action bottomed on the federal right. In theory, all that is lost is the federal right holder’s option to file in federal court.²³⁴

The cost of losing access to federal district court as the court of first instance matters if state courts are less hospitable to the claims of federal right holders. This was all too familiar during Reconstruction and Jim Crow segregation. More recently, state courts have faced extraordinary budget crises, forcing some local courts into a form of fiscal parasitism upon the very communities they are supposed to serve.²³⁵ And as Haywood and the cases on which it relies show, state legislative efforts to restrict Section 1983 litigation in state courts are not uncommon.

Notably, however, Section 1983 does not create exclusive federal jurisdiction—it does not rest on the assumption that state courts are categorically untrustworthy. Concurrent state court jurisdiction decentralizes judgments about which forum is appropriate, allowing the federal right holder to choose. The plaintiff, on advice of counsel, has the right to assess the specific circumstantial risks of hostility in state court, the costs and benefits of filing in federal court, and any other doctrinal and strategic considerations that may inform choice of forum.²³⁶ Using a broad structural constitutional principle such as federalism to displace congressionally endorsed, decentralized decision-making by the federal right holder with the Court’s arm chair conclusions about the fidelity and solicitude of state courts is precisely the kind of “Mr. Fix-It-Mentality” Justice Scalia rightly criticized as imperious judicial policymaking in a civil liberties case.²³⁷ It also flatly contradicts the Court’s near complete deference to Congress’s remedy-creating power in other settings. Most prominently, the Court has deferred to any evidence that Congress has established an alternative remedial scheme when the question is whether the Court should recognize an implied right of action to enforce constitutional rights violated by federal officers,²³⁸ an implied right of action to enforce statutory rights,²³⁹ or a right of action grounded in equity under Ex parte Young.²⁴⁰ If judicial restraint demands deference to the decision of Congress to create remedies in these settings, deference is owed to the forum-selection authority Congress granted federal right holders seeking relief under Section 1983. Conspicuously, the Court’s Section 1983 jurisprudence runs headlong in the opposite direction.

Also lost in enforcing federal civil rights through state law claims is the nexus between the federal right, the cause of action, and the remedy Section 1983 creates. The nexus between injury, right, cause of action, and remedy might seem like an abstraction. After all, what should it matter if a plaintiff receives $100,000 in damages on a state tort claim or $100,000 on a federal Fourth Amendment claim under Section 1983; so too a state injunction reaching the same conduct as a federal injunction? The difficulty is that state law and constitutional harms are not necessarily equivalent. There may be no state cause of action that overlaps with certain constitutional injuries. Even when there is, the elements of a state law cause of action, state procedure, and state remedies doctrine may defeat or limit recovery for certain violations of federal law. Ultimately, state law is not designed to implement federal interests and enforce federal rights.²⁴¹

State law may even be hostile to federal rights.²⁴² Causes of action designed to address harm inflicted by private persons do not reflect the anti-democratic harm arising from the violation of federal rights by an officer of the state acting “under color of law” with the badges, implements, and armory of the state’s monopoly on violence.²⁴³ Section 1983 claims, by contrast, name the abuse of sovereign authority directly and force the officer or municipal government to answer for it. If the essence of corrective justice is that “liability rectifies the injustice inflicted by one person on another . . . simultaneously taking away the defendant’s excess and making good the plaintiff’s deficiency[,]”²⁴⁴ then the right, cause of action, and remedy should correlatively address the injury and injustice of abuse of sovereign power.

The point is not that corrective justice requires perfect mirroring between the cause of action and the injury, but rather that the Court owes deference to the congressional judgment to make injury, right, and remedy mutually correlative rather than contingent on state common law where constitutional rights are concerned. Whatever wisdom and policy justified the common law approach to enforcing federal law in officer suits during the antebellum period,²⁴⁵ Reconstruction revealed the deadly consequences of its potential for abuse (in the form of “outrages”) and the Reconstruction Amendments empowered Congress to provide a less circuitous remedial path. The presumption in favor of state common law causes of action is precisely what Section 1983 rejects by bottoming a cause of action on the federal right itself. Lastly, it is no answer to say that the Supreme Court can exercise appellate jurisdiction to correct underenforcement of federal civil rights when it decides so few appeals from state courts.

These arguments have little resonance within the interpretive horizon of the modern Court’s federalism revival. Suit in state court under state law is presumptively superior as measured against the potential “excesses” of federal judicial intervention. Structural constitutional attachments have in this way come to dominate substantive constitutional rights, legislative text, and federal court remedial power. Unlike Kafka’s penal colony, however, where the apparatus responsible for the extraordinary excesses of the ancien regime was being dismantled and the officer’s perverse attachment to it was reassuringly anachronistic, here, now it is the second Reconstruction that is being dismantled, our body politic still transfigured by the inscriptions of sovereign power’s outrageous excess.

98 S. Cal. L. Rev. 367

Download

Norman W. Spaulding*

* Sweitzer Professor of Law, Stanford Law School. I am grateful for excellent assistance with research provided by Owen Foulkes, Anna Kuritzkes, Joelle Miller, and Molly Shapiro.

Property and Prejudice

April 2025April 2025Fatma Marouf* & Vanessa Casado Pérez†

“Alien land laws”—laws restricting noncitizens from owning real property—are back. A dozen states have enacted such laws during the past year, and over thirty states have considered such bills. These new bills are rooted in xenophobia, much like their predecessors, but they also have unique characteristics. They single out governments, citizens, and corporations of specific countries perceived to pose a threat; they impose ownership restrictions based on arbitrary distances to U.S. military bases and critical infrastructure; they inflict particularly harsh penalties; and they try to ferret out foreign control in complex corporate structures. The purported justifications are national defense, food security, and prevention of absentee ownership. But these laws completely fail to achieve their asserted goals. The poor means-end fit, combined with the availability of far less restrictive alternatives, leaves the new laws vulnerable to legal challenges under the Equal Protection Clause and the Fair Housing Act. But century-old Supreme Court precedents and gaps in legal doctrine may still make it difficult for such challenges to prevail. Preemption arguments based on immigration law, the foreign affairs power, and federal laws governing foreign investment, as well as Dormant Commerce Clause arguments, also involve legal hurdles. This Article analyzes these legal arguments, evaluates potential obstacles, and charts possible paths forward. Regardless of the legal viability of these laws, this Article cautions that they will perpetuate prejudice, open the door to a new form of segregation, and limit who can achieve the American Dream.

INTRODUCTION

Sun Guangxin, a Chinese real estate tycoon, owns 140,000 acres of land in Val Verde County, Texas, near an Air Force base close to the border.¹ He spent approximately $110 million on real estate purchases, paying above-market prices for plots that were not on the market.² But Mr. Sun did not buy this land himself. He used a Texan intermediary, who bought the land and transferred it to Mr. Sun’s company, GH America Energy LLC, a subsidiary of the China-based Guanghui Energy Company.³ The plan was to establish a wind farm and produce renewable electricity for the Texas grid.⁴

Environmentalists opposed the wind farm, but their concerns did not gain traction until they framed the wind farm as a threat to national security due to its location.⁵ That got the attention of Senator Ted Cruz and state legislators, who began campaigning against the wind farm.⁶ This campaign became a catalyst for several bills in Texas that restricted foreign ownership of land.⁷ The bill that received the most traction prohibited real property ownership by any businesses headquartered in China, Iran, Russia, and North Korea or owned or controlled by citizens of those countries, as well as by individual citizens and government actors from those countries.⁸

Texas is not alone. In the past year, bills have been proposed in over thirty states that would restrict foreign ownership of land, real estate, and natural resources.⁹ To date, a dozen of them have been enacted into law.¹⁰ Many of these laws single out specific countries perceived to be hostile, including, but not limited to, China, Iran, Russia, and North Korea. Some bills name countries directly, while others reference various federal designations, such as federal lists of “foreign adversaries” and “countries of particular concern.”¹¹ A few bills are a bit more subtle, restricting ownership by “state-controlled enterprises,” which are most common in China,¹² or citing statutes that address only Chinese military companies.¹³

These laws fan the flames of rising anti-Chinese sentiment. Over 80% of the U.S. population currently holds an unfavorable view of China.¹⁴ Fear of China’s economic and military power,¹⁵ disapproval of China’s foreign policies and human rights abuses,¹⁶ media reports blaming China for the COVID-19 pandemic,¹⁷ and angst over espionage,¹⁸ as well as explicit or implicit biases,¹⁹ fuel these views. Of course, most Chinese investors seeking to buy property in the United States are not acting as pawns of the Chinese Communist Party. Instead, they may be families trying to move their money beyond the reach of the Chinese government, investing to ensure that their children get a good education, or hoping to establish themselves in the United States.

Despite the new context, these laws conjure up one of the darkest periods of U.S. immigration history, involving Chinese Exclusion²⁰ and an Asiatic Barred Zone that swept across a continent.²¹ The history of alien land laws is intertwined with racial exclusions from U.S. citizenship and the creation of hierarchies based on race, national origin, and alienage.²² As California’s Attorney General said in 1913 when he championed the state’s alien land law aimed at limiting the presence of Japanese immigrants: “[T]hey will not come in large numbers and long abide with us if they may not acquire land.”²³ A century ago, the U.S. Supreme Court upheld California and Washington’s alien land laws, and it has never revisited the issue.²⁴ These lingering precedents from an unabashedly racist era are now being relied on by states eager to stretch the limits of traditional state powers like regulating the transmission of property and to influence the federal domains of immigration, national security, and foreign affairs.

This new wave of alien land laws differs from prior waves in important respects.²⁵ First, the naming of specific countries and use of certain federal lists reflects a new form of national security creep. This national security slant also appears in the heightened restrictions placed on property located within a certain distance of critical infrastructure, such as military bases and weather stations. While some states have found ten miles to be a safe distance, others require fifty miles, suggesting an arbitrariness to the restrictions imposed. The new laws also seek to ferret out foreign control in more complex corporate structures than ever before. And they punish violators with harsher criminal penalties than in the past.

While the laws purport to protect national security and food security, and to prevent absentee landownership, they are poorly designed to achieve these aims. Foreign ownership of U.S. real property is minimal. Only 2.9% of privately held agricultural land²⁶ and 1.8% of residential real estate²⁷ is foreign-owned. Additionally, the major foreign owners of agricultural land are not from the countries targeted by the new state laws. While China is second only to Canada on the list of foreign countries whose citizens are buying U.S. residential properties,²⁸ their share of US land is very small. Foreigners own 31% of the land in the U.S., but Chinese investors represent only 1% of all foreign-owned land.²⁹

But even assuming there are compelling government interests at stake, the means used to achieve them are ineffective. These laws will not solve the problem of foreign interests and corporate consolidation driving the real estate and agricultural markets, as sophisticated players can easily circumvent the restrictions. For example, because most of the laws do not restrict leases, a foreign-owned business could just lease land from local landowners. The restrictions on landownership will also not increase national security in an era of cyber warfare, drones, and spy balloons. Furthermore, some of these alien land laws target only ownership and not leases. A tenant occupying a property near a military base can be as dangerous as the owner of that land, if not more. This new wave of alien land laws also fail to prevent absentee landownership because they generally exempt noncitizens residing in other U.S. states, along with all U.S. citizens and permanent residents regardless of their location. Less restrictive alternatives to some of the proposed or enacted laws could include simply limiting the amount of land that foreigners may own, requiring owners to reside or work on the land to avoid absentee ownership, or creating exceptions for residences if the main concerns are agriculture and food.

Given the poor means-end fit, the true purpose of the laws appears to be symbolic. These laws may simply be a way for politicians to capitalize on the xenophobic sentiments of their electoral base. Sadly, their nefarious social effects will extend well beyond the real estate market.³⁰ Like racist property restrictions of the past, the new laws will subordinate minorities. Excluding people from home ownership keeps them out of communities, deters immigration, impedes intergenerational transfers of wealth, and obstructs personal flourishing. Even people who are not directly affected by the new laws will suffer due to the chilling effect on the real estate market. Sellers will be hesitant, at best, to engage in transactions with anyone from a targeted country.

This Article examines potential legal challenges to the new wave of alien land laws. Part I provides historical background about prior waves of alien land laws. Part II describes the distinctive characteristics of the current wave. Part III explores possible statutory and constitutional arguments for challenging the new laws. First, Part III explores whether these laws violate the Fair Housing Act, which was enacted as part of the Civil Rights Act of 1968 and prohibits discrimination in housing based on race and national origin.³¹ Second, Part III examines whether the new laws violate the Equal Protection Clause, highlighting the underdeveloped nature of equal protection jurisprudence on alienage and national origin classifications. This Section also stresses the lack of means-end fit, which we argue should result in the laws being struck down under either strict scrutiny or rational basis review.³² Next, this Article analyzes whether the new state laws are preempted by federal immigration law, the federal foreign affairs power, or the federal regulatory framework involving the Committee on Foreign Investment in the United States (“CFIUS”).³³ Finally, this Article analyzes whether the new laws violate the Dormant Commerce Clause with respect to both domestic and foreign commerce.

Legal challenges to the new alien land laws will not be easy. A federal district court has already refused to enjoin Florida’s law, which not only restricts individuals and companies domiciled in certain countries but also singles out those domiciled in China for especially harsh treatment.³⁴ The legal questions raised by alien land laws will likely reverberate in other important contexts as well. States like Texas and Florida are increasingly looking for ways to use well-established state powers, including police and property powers, to challenge the federal government’s authority over international borders and immigration.³⁵ Alien land laws represent one, but by no means the only, way for states to do this. If no restrictions are placed on alien land laws by courts or the federal government, states could use them to create new forms of segregation, excluding immigrants from their territories by denying them a place to live. In short, these laws once again instrumentalize property for racial prejudice.

I. A BRIEF HISTORY OF ALIEN LAND LAWS

Alien land laws in the United States date back to colonial times and to the influence of the English feudal system.³⁶ English feudal laws were designed to secure allegiance to the Crown and initially prohibited aliens from purchasing land; then, the laws prohibited them from inheriting it.³⁷ England eventually abolished those restrictions by statute in 1870.³⁸ But alien land laws continued in the United States, sanctioned by common law.³⁹ Some early land laws were incorporated into state constitutions in explicitly racial terms. For example, in 1859, Oregon amended its constitution to prevent any “Chinaman” from owning property in the state and granted only “white foreigners” the same property rights as citizens, a provision that was not repealed for over one hundred years.⁴⁰

Scholars have previously categorized alien land laws into several waves.⁴¹ During the first wave, which extended from approximately 1880 to 1900, eleven states restricted alien ownership of real property in response to a depressed agricultural economy and concerns over absentee landowners.⁴² Congress also passed the Territorial Land Act of 1887, which “forbade extensive alien landholding in the organized territories, except by immigrant farmers who had applied for citizenship.”⁴³ The federal law aimed to prevent large, foreign-owned ranches from jeopardizing statehood for the territories.

The second wave of alien land laws were passed in the 1920s, as a result of resentment toward Japanese immigrants engaged in farming in California, Oregon, and Washington.⁴⁴ California’s law “was enacted and . . . enforced solely as a discriminatory law directed against the Japanese.”⁴⁵ California’s Attorney General at the time, Ulysses S. Webb, was transparent about its purpose, framing the central issue as “race undesirability.”⁴⁶ The California law carried criminal penalties and resulted in successful prosecutions;⁴⁷ it also led to severe financial losses with over 30,000 Japanese farmers abandoning “nearly 500,000 acres of California’s richest crop lands.”⁴⁸ Beyond these penalties, the law had a severe psychological impact, demoralizing and subordinating Japanese Americans.⁴⁹

Alien land laws passed at this time often excluded Japanese and other Asians by precluding noncitizens “ineligible for citizenship” from owning land.⁵⁰ As Keith Aoki observed, “ ‘aliens ineligible to citizenship’ was a disingenuous euphemism designed to disguise the fact that the targets of such laws were [Japanese].”⁵¹ Laws dating back to 1790 and 1870 excluded Asians from naturalizing.⁵² In 1922, the U.S. Supreme Court confirmed that a Japanese person could not be naturalized because he was not “white.”⁵³ The following year, the Court reached the same conclusion regarding someone from India.⁵⁴

That same year—1923—the U.S. Supreme Court upheld Washington’s and California’s alien land laws.⁵⁵ Both cases involved U.S. citizens who wanted to lease land to Japanese farmers. In Terrace v. Thompson, the Court reasoned that Washington had “wide discretion in determining its own public policy and what measures are necessary for its own protection and properly to promote the safety, peace and good order of its people.”⁵⁶ The Court explained that “in the absence of any treaty provision to the contrary, [a state] has power to deny to aliens the right to own land within its borders.”⁵⁷ Similarly, in Porterfield v. Webb, the Court found California’s law limiting property rights to those “eligible to citizenship” to be constitutional.⁵⁸ Two other U.S. Supreme Court cases decided that year upheld laws restricting the transfer of shares of a landowning corporation to aliens⁵⁹ and prohibiting food crop contracts with aliens.⁶⁰

But Supreme Court decisions issued in 1948 cast doubt on whether Terrace and Porterfield remained good law. In Oyama v. California, the Court invalidated a provision of California’s alien land law that deprived a U.S. citizen of Japanese descent of agricultural land paid for by his father.⁶¹ The Court found that the state had failed to offer any compelling justification for discriminating against a citizen “based solely on his parents’ country of origin.”⁶² The Court recognized that restrictions based on ineligibility for citizenship constituted discrimination based on “racial descent.”⁶³ That same year, in Takahashi v. Fish and Game Commission, the Court declared unconstitutional a California law that allowed only U.S. citizens to get fishing licenses, which was aimed at discouraging Japanese immigrants from returning to the state after their exclusion from the West Coast and internment.⁶⁴ Justice Black, writing for the Court, explained that “the power of a state to apply its laws exclusively to its alien inhabitants as a class is confined within narrow limits.”⁶⁵

In the years following Takahashi, the supreme courts of Oregon, California, and Montana invalidated those states’ alien land laws, recognizing their racist nature and finding them unconstitutional.⁶⁶ The Supreme Court of California opined that the law imposed on noncitizens “an economic status inferior to that of all other persons living in the state. ”⁶⁷ Other states decided to simply repeal their laws.⁶⁸ The Immigration and Nationality Act of 1952, which made naturalization race-neutral, rendered meaningless any remaining state laws that still tied property ownership to eligibility for citizenship.⁶⁹ But various other types of alien land laws remained. For example, in 1943, Wyoming had enacted an alien land law that prohibited Japanese Americans who had been in internment camps from buying land in the state, which was not repealed until 2001.⁷⁰

During the Cold War, a third wave of state laws emerged limiting the rights of foreigners to receive land by inheritance.⁷¹ The purpose of these laws was to keep U.S. wealth from communist regimes rather than to prevent noncitizens from owning land.⁷² This practice ended after the U.S. Supreme Court’s 1968 decision in Zschernig v. Miller, which invalidated an Oregon statute that conditioned a noncitizen’s inheritance right on reciprocal rights being granted to U.S. citizens.⁷³ The Court found that the Oregon law was preempted because it intruded on the federal government’s authority over foreign affairs.

A fourth wave of alien land laws occurred during the 1970s in response to media reports of increased foreign investment in U.S. farmland.⁷⁴ These laws generally restricted the type and amount of land that noncitizens could purchase. Media reports stoked fears that family farmers in the U.S. were threatened by foreign investment.⁷⁵ In 1972, the Wisconsin Supreme Court upheld an alien land law with “no racial implications” that restricted only the amount of land that could be owned by foreign investors, finding the law “sufficiently related to the state’s asserted desire to limit possibly detrimental absentee land ownership.”⁷⁶

The current wave of land laws has much in common with these prior waves. Anti-immigrant biases, xenophobia, and fears regarding the fate of family farmers all appear to be playing a role. But as discussed below, the new bills and law also have their own distinct characteristics.

II. RECENT BILLS AND LAWS: THE FIFTH WAVE

The fifth wave of alien land laws began around 2020 and rapidly gained momentum. In 2022 and 2023, dozens of bills were proposed across the country restricting the ownership of real property by individual noncitizens, foreign companies, and foreign governments.⁷⁷ To date, twelve of those bills have been enacted into law in Alabama, Arkansas, Florida, Idaho, Indiana, Louisiana, Montana, North Dakota, Oklahoma, Tennessee, Utah, and Virginia. These laws, like their predecessors, vary widely, both in terms of whom they restrict and what is restricted.

Some of the newly enacted laws focus on foreign governments and businesses rather than individuals.⁷⁸ Among the laws that apply to individual noncitizens, most restrict only “non-resident aliens,” while exempting “resident aliens.” Residence in this context generally refers to domicile in the United States,⁷⁹ but a couple of laws define a “resident alien” to mean a noncitizen who lives in the state.⁸⁰ Some of the laws require “resident aliens” to dispose of their real property within a certain amount of time if they no longer qualify as residents of the state.⁸¹

Other laws turn on immigration status rather than residence. For example, Louisiana’s law exempts anyone “lawfully present” in the U.S.⁸² Tennessee’s definition of a “sanctioned nonresident alien” explicitly excludes legal permanent residents.⁸³ North Dakota, like Minnesota, exempts not only legal permanent residents but also noncitizens who enter with certain types of temporary investor or trader visas that are available only to citizens of specific countries that have special treaties with the United States.⁸⁴

Like prior waves, many of the new laws place restrictions specifically on agricultural land and other natural resources.⁸⁵ Some are even more specific. Indiana, for example, has prohibited foreign business entities from owning agricultural land for the purpose of crop farming or timber production.⁸⁶ However, there are also novel types of restrictions. Notably, many of the new laws restrict ownership of land within a certain distance of a military installation or other “critical infrastructure.”⁸⁷ Other bills and laws apply broadly to any type of land or real property.⁸⁸

The following Sections take a closer look at some of the distinct characteristics of the new wave of alien land laws and proposed bills. These include singling out specific countries or nationalities by name, focusing on foreign adversaries, prohibiting landownership within a certain distance of military installations or critical infrastructure, focusing on agricultural land, imposing more severe penalties for violations, and targeting all types of foreign control in complex corporate structures.

A. Singling Out Specific Countries

Bills proposed in at least a dozen states (including Alabama, Arkansas, Colorado, Iowa, Florida, Georgia, Maryland, Mississippi, South Carolina, Texas, West Virginia, and Wyoming) singled out specific countries for property restrictions.

For example, Alabama enacted a law that defines a “foreign country of concern” as “China, Iran, North Korea, and Russia.”⁸⁹ Bills considered in Arkansas,⁹⁰ Georgia,⁹¹ and Texas⁹² similarly placed restrictions on citizens of these four countries. In Colorado, West Virginia, and Wyoming, proposed bills placed restrictions on citizens of China, Russia, or any country designated as a “state sponsor of terrorism.”⁹³

Florida enacted an alien land law that defined a “foreign country of concern” to mean China, Iran, North Korea, Russia, Cuba, the Venezuelan regime of Nicolás Maduro, and Syria.⁹⁴ Florida’s law is harshest, however, on citizens of China, placing more severe restrictions on them and subjecting them to stiffer penalties for violating the law.⁹⁵ A bill proposed in Arizona included the same seven countries on Florida’s list plus Saudi Arabia.⁹⁶ The Arizona bill emerged after a Saudi Arabian company made headlines for leasing Arizona public lands and pumping exorbitant amounts of groundwater to grow alfalfa for export to Saudi Arabia.⁹⁷

Many other bills singled out China alone, including bills proposed in Iowa,⁹⁸ Maryland,⁹⁹ Mississippi,¹⁰⁰ South Carolina,¹⁰¹ and Washington.¹⁰² Two Arizona bills,¹⁰³ as well as a bill proposed in Hawaii,¹⁰⁴ refer specifically to the Chinese Communist Party and its members. A Utah bill indirectly references Chinese companies by defining a “restricted foreign entity” as a company that the Secretary of Defense is required to report as a military company, which includes only Chinese military companies.¹⁰⁵ The intense focus on China across so many of these bills and laws is reminiscent of the anti-Asian sentiment that fueled alien land laws long ago. Alien land laws singling out specific countries are less likely to pass constitutional muster than more evenhanded laws.¹⁰⁶

B. Targeting Foreign Adversaries

Prior to the most recent wave, only five states had alien land laws that restricted land ownership by citizens of foreign adversaries.¹⁰⁷ None of those laws explicitly referred to foreign adversaries, much less attempted to name them. Instead, they benignly extended equal property rights to “alien friends” (New Jersey),¹⁰⁸ “[a]liens who are subjects of governments at peace with the United States and this state” (Georgia),¹⁰⁹ or any alien who is “not an enemy” (Kentucky, Maryland, and Virginia).¹¹⁰

In 2023, however, numerous state legislatures considered or passed laws restricting property ownership rights of citizens and companies of countries designated by the federal government as hostile to the U.S. or its values in some way. These bills and laws use various federal lists that were created for completely different purposes.

Laws enacted in Louisiana,¹¹¹ North Carolina,¹¹² and Virginia,¹¹³ as well as bills proposed in Kansas,¹¹⁴ Montana,¹¹⁵ Ohio,¹¹⁶ South Carolina,¹¹⁷ and Wisconsin,¹¹⁸ refer to the Secretary of Commerce’s designation of certain countries as “foreign adversaries” in the Code of Federal Regulations.¹¹⁹ This designation is based on the Secretary’s determination that a foreign government or foreign nongovernment person has “engaged in a long-term pattern or serious instances of conduct significantly adverse to the national security of the United States or security and safety of United States persons.”¹²⁰ Currently, this designation applies to six countries: China (including Hong Kong), Cuba, Iran, North Korea, Russia, and “Venezuelan politician Nicolás Maduro (Maduro Regime).”¹²¹

The Ohio bill and Louisiana law restrict not only “foreign adversaries” as defined by Secretary of Commerce but also the much longer list of foreign governments sanctioned by the Office of Foreign Assets Control (“OFAC”), which adds Afghanistan, Belarus, Burma, Central African Republic, Democratic Republic of Congo, Ethiopia, Iraq, Lebanon, Libya, Mali, Nicaragua, Somalia, Sudan, South Sudan, Syria, Yemen, and Zimbabwe.¹²² A law enacted in Tennessee, on the other hand, refers to citizens of foreign governments sanctioned by OFAC but does not include “foreign adversaries” designated by the Secretary of Commerce.¹²³

Other bills and laws refer to various U.S. State Department designations. For example, a bill proposed in New York¹²⁴ refers to a “foreign country of particular concern,” which currently includes twelve countries designated by the State Department: Burma, China, Cuba, Eritrea, Iran, North Korea, Nicaragua, Pakistan, Russia, Saudi Arabia, Tajikistan, and Turkmenistan. Bills proposed in Colorado, West Virginia, and Wyoming reference a completely different U.S. State Department designation—“state sponsors of terrorism”—a list that currently includes only four countries: Cuba, Iran, North Korea, and Syria.¹²⁵

States have also incorporated other federal definitions into their bills and laws. For example, the law passed in Arkansas references not only foreign countries of “particular concern” but also includes citizens or residents of countries subject to the International Traffic in Arms Regulations.¹²⁶ Meanwhile, a bill proposed in Texas referred to countries identified by the United States Director of National Intelligence as posing a risk to the national security of the United States in each of the three most recent Annual Threat Assessments of the U.S. Intelligence Community.¹²⁷ At least one law, enacted in Indiana, does not refer to federal definitions at all and instead allows the governor to designate certain countries as a threat to critical infrastructure.¹²⁸

A few of the proposed bills simply make vague references to “hostile” countries without providing a clear definition of the term. For instance, a Mississippi bill restricts ownership by “citizens of a country that is hostile to the interests of the United States or a country that is a known violator of human rights,” without explaining how such countries should be identified.¹²⁹ Similarly, a Hawaii bill that restricts land ownership by members of the Chinese Communist Party also refers to “other hostile foreign influence,” providing only a vague definition of this term.¹³⁰

C. Proximity to Military Installations and Critical Infrastructure

Additionally, many of the recent bills and laws limit landownership near military installations or other critical infrastructure. Considerable variation exists among the bills regarding what types of facilities are included under these terms as well as what constitutes an acceptable distance from them.

For example, a bill proposed in California prohibits foreign actors from owning or leasing land within fifty miles of a U.S. military base or California National Guard Base.¹³¹ A bill proposed in Louisiana restricts foreign ownership of “immovable property located within [fifty] miles of any federal or state military land, . . . weather station[], . . . or any facility operated by the Civil Air Patrol.”¹³² A bill proposed in Mississippi prohibits nonresident aliens from owning land within fifty miles of a military installation under the jurisdiction of the Department of Defense, the U.S. Coast Guard, or the Mississippi National Guard.¹³³ A South Carolina bill prohibits companies owned by China or the Chinese Communist Party, or whose principal place of business is in China, from controlling any land or real estate “within fifty miles of a state or federal military base or installation for the purpose of installing or erecting any type of telecommunications or broadcasting tower.”¹³⁴

Bills proposed elsewhere specify shorter distances from military installations. For example, a Georgia bill prohibits nonresident aliens from possessing any land within twenty-five miles of any military base, military installation, or military airport.¹³⁵ A North Carolina bill prohibits adversarial foreign governments from purchasing or holding land within twenty-five miles of a military base or airport.¹³⁶ The law enacted in Florida generally prohibits foreign land ownership within ten miles of a military installation or critical infrastructure facility.¹³⁷ Florida’s choice of ten miles is particularly interesting given that the legislative history indicates that a major concern was a Chinese company’s purchase of land located twelve miles from an air force base in North Dakota.¹³⁸ Meanwhile, a bill proposed in Hawaii considered just two miles from federal land or critical infrastructure to be a safe distance.¹³⁹

D. Harsh Penalties

Criminal penalties and prosecutions for violations of alien land laws are not new. In California and Arizona, such criminal prosecutions were common during the 1920s and 1930s, but those laws were subsequently repealed.¹⁴⁰ Penalties for violating a state’s alien land laws have generally been civil. Forfeiture of the property or sale at auction with proceeds escheating to the state were commonly specified as penalties in state laws. Under some laws, such as Wisconsin’s, a civil fine could be imposed, ranging from $500 to $5,000.¹⁴¹ Criminal penalties existed but were rare.¹⁴²

In the most recent wave of bills, criminal penalties have gained popularity, and civil fines are steeper. Additionally, some of the new bills and laws impose penalties on the sellers as well as the buyers. For example, the alien land law enacted in Arkansas makes a violation a felony punishable by two years in jail and a $15,000 fine.¹⁴³ Being a “resident alien” is mentioned as an “affirmative defense” to the charge.¹⁴⁴ Florida has also made it a criminal offense to violate its new law, which imposes harsher criminal consequences on Chinese purchasers of land than purchasers of other nationalities.¹⁴⁵ Violators who are domiciled in China may be charged with a third-degree felony, punishable by up to five years in jail and a $5,000 fine, while violators domiciled in the other countries named in Florida’s law may be charged with only a second-degree misdemeanor, punishable by sixty days in jail and a $500 fine.¹⁴⁶ This disparity extends to sellers. Selling real property to individuals or companies domiciled in China is a first-degree misdemeanor, punishable by one year in prison and a $1,000 fine, while selling property to individuals or companies domiciled in other countries is only a second-degree misdemeanor.¹⁴⁷

E. Targeting Corporations

Finally, the current wave of alien land laws targets all forms of foreign control in complex corporate structures. The laws restrict not only foreign corporations but also companies incorporated in the U.S. if they are controlled by noncitizens who would not be allowed to purchase the real estate themselves. The expansive language used in some of these laws reflects an attempt to close the loopholes in previous laws that allowed foreigners to acquire land simply by channeling their investments through the veil of a U.S. corporation. This was one of the main drivers behind the recent alien land law passed in Oklahoma, which specified that “[n]o alien or any person who is not a citizen of the United States shall acquire title to or own land in this state either directly or indirectly through a business entity or trust.” ¹⁴⁸

Similarly, a Tennessee bill defined a “foreign business” as “a corporation incorporated under the laws of a foreign country, or a business entity whether or not incorporated, in which a majority interest is owned directly or indirectly by nonresident aliens.”¹⁴⁹ The bill further explained, “Legal entities, including, but not limited to, trusts, holding companies, multiple corporations, and other business arrangements, do not affect the determination of ownership or control of a foreign business.”¹⁵⁰ But this bill still has a major loophole—its definition of a foreign business is limited to owning a majority interest and does not address control. Nonresident aliens could control a corporation based on voting power, even if they do not own a majority of the stock.¹⁵¹

Many other bills closed that loophole. A Washington bill, for example, prohibited acquisition of agricultural land by a foreign-controlled enterprise and defined a controlling interest to mean “possession of more than [fifty] percent of the ownership interests in an entity, or an ownership interest of [fifty] percent or less if the persons holding such interest actually direct the business and affairs of the entity without the consent of any other party.”¹⁵² A law enacted in North Dakota adopts a nearly identical definition.¹⁵³

While the definitions in the new bills and laws vary and are not perfect, they clearly seek to capture all kinds of businesses in which noncitizens play a decisive role. Of course, if a corporation is forty-nine percent owned by U.S. citizens and fifty-one percent owned by noncitizens, the U.S. citizen owners are also likely to suffer financial setbacks as a result of such laws.

III. ARE ALIEN LAND LAWS LEGAL?

Commentators have taken different perspectives on the legality of alien land laws in the past.¹⁵⁴ Some have argued that alien land laws would violate the Equal Protection Clause if they singled out specific countries.¹⁵⁵ Others contend that only restrictions on lawful permanent residents would raise equal protection concerns, and even those may be permissible.¹⁵⁶ Preemption concerns and Dormant Commerce Clause concerns have also been raised.¹⁵⁷ Because of significant variations among the laws, it is difficult to analyze these legal issues for the laws as a whole. Nevertheless, this Part attempts to parse some of the legal challenges that the new wave of alien land laws may face.

A. Statutory Violations

Alien land laws may conflict with federal statutes that prohibit discrimination such as the Fair Housing Act (“FHA”)¹⁵⁸ and the Civil Rights Acts of 1866¹⁵⁹ and 1870.¹⁶⁰

1. The Fair Housing Act

The FHA, enacted as part of the Civil Rights Act of 1968, seeks to prohibit unlawful discrimination by landlords. Under the FHA, it is discriminatory “[t]o refuse to sell or rent after the making of a bona fide offer, or to refuse to negotiate for the sale or rental of, or otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.”¹⁶¹ Although alienage is not specifically mentioned, the U.S. Department of Housing and Urban Development (“HUD”) has stated that “[a] requirement involving citizenship or immigration status will violate the [FHA] when it has the purpose or [unjustified] effect of discriminating on the basis of national origin.”¹⁶² Private parties would be violating the FHA if they comply with state laws that restrict who can buy or lease real estate based on national origin. States may enhance the protections of the FHA but cannot reduce them. Section 816 of the FHA declares invalid any state law that requires or permits any action that would be a discriminatory housing practice under the FHA.¹⁶³

One aspect of the FHA that makes the inquiry different from an equal protection claim is that claimants do not need to prove discriminatory intent. A facially neutral law may violate the FHA if it has “discriminatory effects.”¹⁶⁴ This is useful in challenging a law like Florida’s, which may be perceived as discriminating based on domicile rather than national origin. By prohibiting sales of real estate to individuals and companies domiciled in China, Florida’s law clearly has discriminatory effects related to national origin: China has over one billion inhabitants, of whom only .05% are not Chinese.¹⁶⁵ Similarly, other countries identified as “foreign adversaries” under Florida’s law have a very small percentage of foreigners. Less than 0.1% of Cuba’s population are immigrants, for instance.¹⁶⁶

A law that has a discriminatory effect on a protected class is unlawful if it is not necessary to achieve a substantial, legitimate, nondiscriminatory interest, or if a less discriminatory alternative could serve that interest.¹⁶⁷ As discussed further under equal protection below, alien land laws are not necessary to achieve the asserted interests, and less discriminatory alternatives are, in fact, available.

An important limitation of the FHA, however, is that it only applies to “dwellings,” that is, to real estate capable of being used as a residence.¹⁶⁸ Thus, while broadly written alien land laws that restrict real estate (or real property in general) remain vulnerable to FHA challenges,¹⁶⁹ those that restrict only agricultural land cannot be challenged under the Fair Housing Act.¹⁷⁰ The Civil Rights Acts of 1866 and 1870 may help fill this gap, although, as explained below, these laws have their own limitations.

2. Civil Rights Acts of 1866 and 1870

The Civil Rights Act of 1866 provided that “citizens . . . shall have the same right, in every State and Territory in the United States, to make and enforce contracts . . . as is enjoyed by white citizens.”¹⁷¹ The Civil Rights Act of 1870 made a significant revision by changing “citizens” to “persons.”¹⁷² This language is now codified in 42 U.S.C. § 1981 (“section 1981”). The revised language made it clear that noncitizens, as well as citizens, are protected by the law’s equality mandate.¹⁷³ Courts have also construed section 1981 as prohibiting discrimination based on alienage.¹⁷⁴ Alien land laws may therefore run afoul of section 1981.¹⁷⁵

One limitation of section 1981 is that it applies only to individuals “within the jurisdiction of the United States.” While this phrase includes noncitizens in the United States,¹⁷⁶ it would likely exclude noncitizens residing abroad, the group most affected by alien land laws. Corporations headquartered abroad that are “foreign adversaries” under Montana’s law therefore may not be able to bring challenges under section 1981, although if they have U.S.-based subsidiaries, such challenges may still be possible. Other states, like Indiana, have broad definitions of “qualified entities.”¹⁷⁷ Many alien land laws tackle corporations controlled by foreigners. Any qualified entities based in the U.S. should be able to bring section 1981 challenges, even if they are owned or controlled by citizens of Iran, North Korea, or China.

Another potential limitation of section 1981 is that a separate provision of the Civil Rights Act of 1866, now codified at 42 U.S.C. § 1982 (“section 1982”), specifically addresses property and extends equal protection only to U.S. citizens.¹⁷⁸ Specifically, section 1982 provides: “All citizens of the United States shall have the same right, in every State and Territory, as is enjoyed by white citizens thereof to inherit, purchase, lease, sell, hold, and convey real and personal property.”¹⁷⁹ That language was not altered by the Civil Rights Act of 1870. Courts could therefore interpret section 1982 as a limited exception to section 1981’s more general rule about contracts, excluding contracts pertaining to property from the alienage equality principle found in section 1981.

Even under this interpretation, however, section 1981 is still relevant, since some of the recently enacted laws not only prohibit buying and selling real property, but also prohibit forming other types of contracts. For example, Indiana’s, Montana’s, and Texas’s new alien land laws prohibit certain foreign entities from countries like China from entering into agreements regarding critical infrastructure (energy grid, water treatment plants, and so on).¹⁸⁰

Additionally, one could argue that section 1982 prohibits the restrictions that alien land laws place on U.S. citizen sellers and landlords, as well as U.S. citizen-owned or controlled realty and title companies. From the perspective of U.S. citizens who want to sell properties, the restrictions imposed by states are restraints on alienation.¹⁸¹ The laws shrink their market, and if the claims about Chinese investors flooding the market and paying exorbitant prices are true,¹⁸² then real estate owners and companies who cater to this population will lose a profitable share of potential buyers. One complication with this argument is that U.S. citizen sellers are not necessarily being treated differently from other “white citizens” under the language of section 1982. For the argument to work, the focus would likely have to be on non-white U.S. citizen sellers, for example, U.S. citizen sellers of Chinese descent whose clientele potentially include a substantial number of Chinese citizens or companies domiciled in China. These U.S. citizen sellers of Chinese descent could argue that they are being deprived of the same opportunities to sell real property that are enjoyed by white citizens who do not have clientele in China.

Another possible legal hurdle is that a disparate impact claim under section 1981 or section 1982 requires showing that the disparate impact is traceable to a discriminatory purpose.¹⁸³ This is more limiting than a disparate impact claim under the FHA. Nevertheless, the legislative history and rhetoric surrounding the passage of some of the laws may help demonstrate a discriminatory purpose. State legislators and executive officials discussing alien land laws have used inflammatory rhetoric coated with national security concerns. Feeding on the anti-Asian sentiment fueled by dubious theories about the origin of COVID-19 and compounded by economic fears concerning China’s influence, their statements are reminiscent of the language used in the era of the “Yellow Peril.”¹⁸⁴ Although alien land laws may seem somewhat removed from the original purpose of the Civil Rights Acts, which was to prevent discrimination against African Americans in the wake of the Civil War, the rhetoric surrounding these laws reflects a form of racial discrimination.

B. Equal Protection Concerns

The Equal Protection Clause applies to all persons within the United States, including all noncitizens.¹⁸⁵ But noncitizens abroad generally are not regarded as having a right to equal protection,¹⁸⁶ although open questions about extraterritorial rights certainly remain.¹⁸⁷ This may be a threshold hurdle for bringing an equal protection challenge, since many of the alien land laws apply only to “nonresident aliens” and define “resident aliens” as noncitizens living anywhere in the U.S.¹⁸⁸ If an alien land law restricts only foreigners abroad, an equal protection challenge would likely need to be brought by the individuals and companies based in the U.S. that are prohibited from selling or leasing real property to foreigners abroad.¹⁸⁹

Another major challenge in bringing an equal protection claim will be the century-old Supreme Court precedents in Terrace and Porterfield upholding alien land laws, which have never been overruled.¹⁹⁰ Of course, in the 1920s, equal protection jurisprudence was quite different than it is today. Segregation, Jim Crow, and racially restrictive covenants were all legal.¹⁹¹ Levels of judicial scrutiny were not introduced until 1938, in the famous footnote four of United States v. Carolene Products, in which Justice Stone mentioned certain circumstances that may call for a “more searching judicial inquiry,” including cases involving “prejudice against discrete and insular minorities.”¹⁹²

In 1948, when the Supreme Court applied this type of searching judicial inquiry in Oyama, it invalidated as racially discriminatory a part of California’s alien land law that deprived U.S. citizens of Japanese descent of property rights.¹⁹³ But the Court stopped short of invalidating the law altogether.¹⁹⁴ That same year, in Takahashi, when the Court struck down a California law that prohibited those “ineligible for citizenship” from obtaining fishing licenses, it rejected California’s reliance on the Terrace and Porterfield cases, finding them not controlling even “[a]ssuming the[ir] continued validity.”¹⁹⁵

The modern strict scrutiny test did not emerge until the 1960s.¹⁹⁶ And it was not until 1971 that the Supreme Court applied strict scrutiny to alienage classifications.¹⁹⁷ In a watershed decision, Graham v. Richardson, the Court found that “[a]liens as a class are a prime example of a ‘discrete and insular’ minority . . . for whom such heightened judicial solicitude is appropriate.”¹⁹⁸ Applying this new, rigorous standard of review, the Court struck down Arizona and Pennsylvania statutes that favored citizens over noncitizens in welfare benefits.¹⁹⁹ Richardson rejected the states’ argument that the restrictions were justified by “a State’s ‘special public interest’ in favoring its own citizens over aliens in the distribution of limited resources such as welfare benefits.”²⁰⁰ The Court also flatly rejected “fiscal integrity” as a compelling justification, stating that “aliens lawfully within this country have a right to enter and abide in any State in the Union ‘on an equality of legal privileges with all citizens under non-discriminatory laws.’ ”²⁰¹

The Supreme Court continued to apply strict scrutiny to strike down state laws that discriminated against noncitizens in employment. The Court invalidated a New York law that permitted only U.S. citizens to be eligible for state employment,²⁰² a Connecticut law that permitted only U.S. citizens to become lawyers,²⁰³ and a Texas law that permitted only U.S. citizens to be notary publics.²⁰⁴

However, the Court has also recognized an exception to strict scrutiny in cases where alienage classifications are related to a state’s political function.²⁰⁵ In Bernal v. Fainter, the Court described this as a “narrow exception” that “applies to laws that exclude aliens from positions intimately related to the process of democratic self-government.”²⁰⁶ Under the political function exception, the Court has applied rational basis review to uphold laws that require police officers,²⁰⁷ probation officers,²⁰⁸ and public school teachers²⁰⁹ to be U.S. citizens.

If strict scrutiny applies to an alien land law, then the law must be narrowly tailored to a compelling government interest, a test that is generally difficult to pass. If rational basis applies, the law must merely be related to a legitimate government interest. Determining which level of scrutiny applies is therefore a critical threshold question in assessing the likelihood of prevailing with an equal protection claim.

1. Does Strict Scrutiny Apply?

There are at least three important legal questions that must be answered in order to determine if alien land laws are subject to strict scrutiny. First, do all alienage classifications receive strict scrutiny or only those affecting lawful permanent residents? Second, does the “political function” exception to strict scrutiny for alienage classifications apply to alien land laws? Third, do restrictions that turn on being domiciled (or headquartered, for a corporation) in particular countries discriminate based on national origin?

i. Do All Alienage Classifications Receive Strict Scrutiny, or Only Classifications Affecting Lawful Permanent Residents?

The Supreme Court’s decision in Richardson broadly stated that “[a]liens as a class are a prime example of a ‘discrete and insular’ minority” and that “classifications based on alienage, like those based on nationality or race, are inherently suspect and subject to close judicial scrutiny.”²¹⁰ This language does not distinguish between legal permanent residents and other noncitizens. Subsequently, in Nyquist v. Mauclet, the Supreme Court also applied strict scrutiny in striking down a New York statute that barred a heterogeneous group of noncitizens (not just permanent residents) from state financial aid for higher education, stressing that “[t]he important points are that [the statute] is directed at aliens and that only aliens are harmed by it.”²¹¹

While the Court has never limited the application of strict scrutiny to lawful permanent residents, its use of the term “resident aliens” has created confusion. The term “resident alien” can easily be misconstrued as shorthand for a permanent resident, although it simply refers to an alien residing in the United States.²¹² Asylum applicants, refugees, and noncitizens with a variety of temporary visas, among others, are permitted to reside in the United States, even though they are not lawful permanent residents.

In Toll v. Moreno, the Supreme Court had an opportunity to clarify what level of scrutiny applies to classifications involving temporary immigrants (technically called “nonimmigrants”) when evaluating a University of Maryland policy that prohibited individuals with G-4 visas from receiving in-state tuition.²¹³ But the Court ultimately found that the university policy was preempted and declined to address the equal protection claim.²¹⁴ A circuit split has since emerged regarding what level of scrutiny applies to state classifications involving temporary immigrants.

The Fifth Circuit has held that temporary immigrants are not a suspect class, applying rational basis review in upholding Louisiana laws that prohibit temporary immigrants from taking the bar exam²¹⁵ and obtaining a nursing license.²¹⁶ In explaining why classifications affecting temporary immigrants receive rational basis review, the Fifth Circuit stressed the ways that temporary immigrants are different from permanent residents, noting that “nonimmigrant aliens may not serve in the U.S. military, are subject to strict employment restrictions, incur differential tax treatment, and may be denied federal welfare benefits.”²¹⁷ The Sixth Circuit followed the Fifth Circuit’s rationale, applying rational basis review in upholding a Tennessee statute that conditions issuance of a driver’s license on being a U.S. citizen or permanent resident.²¹⁸

The Second Circuit, on the other hand, has held that temporary immigrants are a suspect class and applied strict scrutiny in striking down a New York statute that prohibited them from a obtaining a pharmacist’s license.²¹⁹ The court refused to create an exception to strict scrutiny for temporary immigrants that the Supreme Court never recognized.²²⁰ Additionally, the court reasoned that the factual similarities between U.S. citizens and permanent residents recognized in Richardson were never intended to be a test for triggering strict scrutiny.²²¹ The court correctly recognized that Richardson’s recognition of aliens as a “discrete and insular minority” was premised on their minority status within the community, not their similarity to citizens.

The only class of noncitizens that the Supreme Court has ever treated differently in terms of the level of scrutiny that applies are undocumented individuals. But even in Plyler v. Doe, in which the Court refused to recognize undocumented children as a suspect class, the Court struck down the Texas statute that denied them a basic education.²²² There, the Court applied a form of intermediate scrutiny by requiring Texas to show that it had a “substantial” interest in excluding undocumented children from public schools.²²³ This heightened scrutiny may have been unique to a case that stressed the importance of education and the innocence of children.²²⁴ Still, if undocumented children received heightened scrutiny, it is difficult to argue that lawfully present noncitizens should receive rational basis review simply because they are not permanent residents.²²⁵

ii. Does the “Political Functions” Exception to Strict Scrutiny for State Alienage Classifications Extend to Ownership of Real Property?

Courts have not yet addressed whether state alien land laws fall under the “political functions” exception to strict scrutiny. If the exception applies, a state’s alienage classifications would receive only rational basis review. In Shen v. Simpson, the case challenging Florida’s 2023 alien land law, Florida argued that the political function exception applies, triggering only rational basis review.²²⁶ Thus far, however, the Supreme Court has only applied the political functions exception to certain state jobs.

The Supreme Court set forth a two-part test for determining “whether a restriction based on alienage fits within the narrow political-function exception.”²²⁷ First, a court examines the specificity of the classification: “[A] classification that is substantially overinclusive or underinclusive tends to undercut the governmental claim that the classification serves legitimate political ends.”²²⁸ As explained further below in the application of the strict scrutiny test, alien land laws are substantially over- and under-inclusive. That alone undercuts the relevance of the political function exception.

Additionally, the second part of the test provides that:

[E]ven if the classification is sufficiently tailored, it may be applied in the particular case only to “persons holding state elective or important nonelective executive, legislative, and judicial positions,” those officers who “participate directly in the formulation, execution, or review of broad public policy” and hence “perform functions that go to the heart of representative government.”²²⁹

The plain language of the second prong indicates that the exception applies only to certain public positions. Owning real property is not a public position. Nor does being a property owner require any involvement in the formulation, execution, or review of public policies. Restricting property ownership is different from “limit[ing] the right to govern to those who are full-fledged members of the political community.”²³⁰

One way to view the issue is to consider whether real property ownership is more closely related to Supreme Court cases protecting noncitizens’ rights to equal economic opportunity,²³¹ or to cases that allow states to limit certain public positions to U.S. citizens.²³² Land is often connected to economic opportunity—agricultural land provides a livelihood through farming and raising livestock; commercial property supports businesses that provide livelihoods; and even residential property is often necessary to work in an area. In fact, in cases striking down state laws that discriminated against noncitizens in employment, the Supreme Court has connected the right to work to the right to “entrance and abode,” stating “they cannot live where they cannot work.”²³³

Furthermore, real property ownership has little in common with the public positions that have fallen under the exception to strict scrutiny. Landowners are not “clothed with authority to exercise an almost infinite variety of discretionary powers,”²³⁴ they do not fulfill “a basic governmental obligation,”²³⁵ and they are not “in a position of direct authority over other individuals.”²³⁶ Under this analysis, if any type of restriction on real property qualifies for the political functions exception, it would only be ownership of state land.

However, if the political functions exception is more broadly construed as encompassing “the process of democratic self-determination” and “the community’s process of political self-definition,” courts may consider land ownership to be relevant.²³⁷ Land can be seen as providing “the basis for political organization.”²³⁸ States’ historical restrictions on foreign land ownership, going back centuries, could also be viewed as reflecting an understanding that such restrictions are somehow inherent to state sovereignty and self-determination.

But choosing who gets to live in a state has not traditionally been part of a state’s right to self-definition. Due to the constitutional right to migrate, the Supreme Court has stressed that “[s]tates . . . do not have any right to select their citizens.”²³⁹ A state law aimed at deterring a particular class of people from migrating to the state is impermissible whether that class consists of welfare applicants, as in Richardson, Japanese immigrants, as in Takahashi, or other noncitizens. Similarly, the Supreme Court has found that a “[s]tate’s objective of reducing population turnover” would “encounter[] insurmountable constitutional difficulties.”²⁴⁰ The political functions exception allows a state to “limit the right to govern to those who are full-fledged members of the political community,”²⁴¹ but it has never allowed a state to limit who lives in the community.

In short, the political functions exception should not apply to alien land laws, and strict scrutiny would be the proper standard of review for their alienage classification.

iii. Do Restrictions Discriminate Based on National Origin if They Draw Distinctions Based on Where a Person or Entity Is Domiciled or Headquartered?

National origin discrimination is distinct from discrimination based on alienage. While alienage discrimination refers to distinctions between citizens and noncitizens,²⁴² national origin discrimination is broadly understood to include discrimination based on an individual’s place of origin, or their ancestors’ place of origin.²⁴³ Laws that place restrictions on citizens or corporations of specific countries ought to trigger strict scrutiny based on national origin.

States may argue, however, that their laws do not discriminate based on national origin but instead draw distinctions based on place of “residence” or “domicile.” For example, Florida’s law restricts only noncitizens who are “domiciled” in certain foreign countries, rather than restricting citizens of those countries outright.²⁴⁴ A federal district court found that the Florida law does not discriminate based on Chinese national origin because Chinese individuals domiciled in the United States are not restricted; only individuals domiciled in China are restricted, and they need not be Chinese.²⁴⁵ Similarly, Montana’s law applies to corporations that are “domiciled or headquartered” in a country identified as a “foreign adversary.”²⁴⁶

A law like Florida’s would clearly have a disparate impact on individuals of Chinese national origin, since over 99% of people living in China are Chinese. But equal protection principles require a showing of intentional discrimination; classifications that merely result in a disparate impact are not subject to strict scrutiny.²⁴⁷ In Village of Arlington Heights v. Metropolitan Housing Development Corp., however, the Court found that discriminatory intent could be evidenced by factors that include “disproportionate impact, the historical background of the challenged decision, the specific antecedent events, departures from normal procedures, and contemporary statements of the decisionmakers.”²⁴⁸ These factors must be assessed cumulatively.²⁴⁹ In Shen, the Florida case, the clearly disproportionate impact on Chinese individuals, along with the legislative history, would support a finding of discriminatory intent under Arlington Heights.

Because the Arlington Heights factors are non-exhaustive, some appellate courts have mentioned other considerations. For example, a “consistent pattern” of actions of decisionmakers that have a much greater harm on minorities than on non-minorities could help establish discriminatory intent.²⁵⁰ In a state like Florida or Texas, where the governors have taken numerous actions to try to prevent immigrants from coming to the state, this may be a relevant consideration.²⁵¹ Courts have also found that applying different, less favorable processes or substantive standards to requests by members of a suspect class may raise an inference of discriminatory intent. Some alien land laws impose special procedures for buyers from certain countries, such as requiring buyers to sign affidavits attesting that they are not principals of China and to register existing properties with the state.²⁵² These types of procedures could further help establish discriminatory intent.²⁵³

2. Analyzing Alien Land Laws Under Strict Scrutiny

In order to survive strict scrutiny, a law must be narrowly tailored to serve a compelling government interest. When strict scrutiny is applied, the government “must show that it cannot achieve its objective through any less discriminatory alternative.”²⁵⁴ The main reasons offered for the new wave of alien land laws are national security, food security, and preventing absentee landownership. As explained below, even assuming these are all compelling government interests, alien land laws are unlikely to survive strict scrutiny because they are not narrowly tailored to achieve these objectives. There are also less restrictive alternatives available.

i. National Security

In explaining the need for Alabama’s newly enacted alien land law, Governor Ivey said, “Across the United States, we have seen alarming instances of foreign entities purchasing large tracts of land, which could have severe consequences for our country’s national defense and economy, if no action is taken.” ²⁵⁵ As discussed above, many of the proposed and enacted laws forbid foreign ownership of land within a certain distance of military installations or critical infrastructure. Such restrictions are highly unlikely to prevent espionage or other national security attacks. The Chinese balloon that hovered over Montana did not need to be launched from land near a military base.²⁵⁶ Neither do drones or cyberattacks gathering U.S. data.²⁵⁷

Furthermore, the “safe” distances from military installations or critical infrastructure are arbitrary in this new wave of alien land laws. As noted above, these distances range from two to fifty miles. The best illustration of this arbitrariness is Florida’s law, which bans ownership by “foreign principals” within ten miles of military installations even though the legislation was triggered by a Chinese company’s purchase of land twelve miles from a military based in North Dakota.²⁵⁸

Nationality-based restrictions on ownership of real property are also easily circumvented. Straw men can be used to purchase the land. A Chinese tycoon can easily have someone purchase it for him, as Mr. Sun did in Texas.²⁵⁹ As commentators have previously noted, alien land laws do not really pose an impediment to acquiring real property.²⁶⁰

A less restrictive alternative would be for states to establish or expand existing reporting requirements for foreign investment in land. Several states have already implemented reporting requirements for foreign investments in agricultural land.²⁶¹ Extending the reporting requirements to all real estate and subjecting those transactions to a review process to identify risky transactions would be less restrictive and potentially more effective than a blanket ban. Once the state has information about a potential transaction, it can decide if the transaction can go forward or if it involves too many risks from a national security perspective. This process imposes less of a restriction on individuals who want to sell their land and is less likely to be perceived as aggressive by foreign countries. It is an approach similar to the one used at the federal level by the Committee on Foreign Investment in the United States (“CFIUS”). However, this approach, like the current one banning transactions, may be preempted by the Foreign Investment Risk Review Modernization Act of 2018 (“FIRRMA”).²⁶² Another way to discourage transactions involving noncitizens abroad is taxation. A tax would increase the cost of real estate transactions, ensuring that only those bringing a large benefit move forward. Taxation, though, could violate the Dormant Foreign Commerce Clause.²⁶³

In short, banning land ownership within certain distances of military installations or critical infrastructure is not going to bring large gains in national security. It will, however, impose significant costs by barring potential good faith purchasers from accessing land, introducing tensions in the United States’ relationship with certain countries, and perpetuating negative sentiments towards people from countries like China.

ii. Food Security

The idea of food security has had a central role in farmland regulation for a long time.²⁶⁴ There is a fear that foreign companies will control U.S. food production and either let Americans suffer if certain products are unavailable or make them pay a higher cost by importing them. The fear is not new: for decades, foreign owners of agricultural land have been required to report to the U.S. Department of Agriculture.²⁶⁵

But this fear is misplaced. The United States has a surplus of agricultural products.²⁶⁶ Furthermore, the bills deal with land ownership as a proxy for agricultural production, but the current structure of agricultural markets may make that an inadequate proxy. Eight of the twenty largest food and beverage companies in the United States are foreign companies, but none are from the countries deemed foreign countries of concern in the new wave of alien land laws.²⁶⁷ Control of agricultural land neither results in automatic control of the food supply, nor does it lead to control of agricultural production. In Iowa, for example, where roughly all non-family corporations are prevented from owning agricultural land, large agribusinesses simply lease the land from several owners, subverting the goal of the ownership prohibition.²⁶⁸ A similar subterfuge could be used by foreign companies in response to state alien land laws.

If the concern is foreign control of agricultural land and absentee ownership, focusing on the “who” by targeting specific countries’ nationals would be a partial solution if the countries singled out were the ones that most foreign owners come from. If that were the case, then instead of banning China, Iran, North Korea, or Russia, states should ban Canada, Netherlands, Italy, the U.K. and Germany, in that order, because each of them owns far more agricultural land than China.²⁶⁹ Even a measure like Washington’s—a blanket prohibition on foreign investment in agricultural land—is not automatically going to slow down the consolidation of land and reduce land prices because domestic companies may still accumulate large amounts of natural resources.

A less restrictive alternative to address concerns about foreign control of resources is to limit the amount of these resources that foreigners can own. This approach recognizes that size matters and that small investments give foreign actors less leverage against federal, state, and local governments.²⁷⁰ Restricting the amount of land that noncitizens can own would also discourage financial investors seeking market control who need a certain scale for the investment to be profitable.

iii. Absentee Ownership

A third motivation for the new wave of alien land laws is concern over absentee ownership.²⁷¹ Absentee ownership is problematic because property is treated as an investment, and the owner generally lacks interest in what role the property could fulfill in the community,²⁷² civically and economically.²⁷³ This concern applies to both agricultural lands and dwellings. Alien land laws that distinguish between “resident aliens” and “nonresident aliens” reflect a desire to preserve property for residents. But because most state laws usually define a “resident alien” as living anywhere in the United States, limiting property ownership to resident aliens would not necessarily prevent absentee ownership. An owner of agricultural land in the Central Valley living in Shanghai is no different than an owner living in Rhode Island. Both will lack the local knowledge and the community involvement.

A more narrowly tailored alternative to address absentee ownership would be to impose a requirement of occupancy or production, or both, like the requirements for establishing a homestead.²⁷⁴ Alternatively, a state could tax land that is not in production at a higher rate, no matter where the owner resides.

A few alien land laws do impose stricter residency requirements to prevent absentee ownership.²⁷⁵ For example, Oklahoma’s newly enacted law has an exception for noncitizens who “take up bona fide residence in this state,” but if they leave the state, they must dispose of the land within five years.²⁷⁶ These requirements likely violate the Commerce Clause.

iv. Real Estate Market Prices

Although not explicitly mentioned by legislators proposing alien land laws, another motivation is fear of foreign investors driving up the prices of real estate.

In the agricultural sector, the fear is that it may displace American farmers who will not have access to land. Alabama’s Senate Bill (“S.B.”) 14 banning foreign ownership of agricultural land illustrates these concerns.²⁷⁷ Although the problem of access to farmland for small family farmers is real, the culprit is not necessarily foreigners but rather investors and consolidation.²⁷⁸ Furthermore, agribusinesses have been dominating the market.²⁷⁹ These alien land laws focus on the “who,” instead of on the “what”—in other words, they do not tackle the issue of corporate consolidation plundering natural resources.²⁸⁰

In the residential market, even if the overall Chinese investment in land is not large, it may have significant effects in certain local markets. While Chinese investment in land may drive prices up, it is necessary to consider a more nuanced picture. In some areas of the Midwest, Chinese investment has helped revitalize crisis-stricken areas, such as the Stonewater Community in a suburb of Detroit.²⁸¹ Many municipalities have welcomed the new developments targeting Chinese buyers. Corinth, near Dallas, readily approved new developments in its jurisdiction.²⁸² The situation may be different in Manhattan or San Francisco and other big cities where Chinese investments may be driving up home values.²⁸³ However, targeting the demand side will not solve the housing crisis because it is a supply-side problem.

The poor fit between alien land laws and their objectives, combined with the availability of less restrictive alternatives, means such laws are likely to be struck down under strict scrutiny.

3. Rational Basis Analysis

If rational basis review applies instead of strict scrutiny, then a court need only inquire if the law is rationally related to a legitimate government purpose. There is no analysis of less restrictive alternatives for rational basis review.²⁸⁴ While laws generally survive rational basis review, courts have invalidated laws motivated by animus by applying rational basis with bite, a heightened form of scrutiny. Both types of rational basis review are discussed below.

i. Regular Rational Basis Review

The poor means-end fit discussed above arguably fails not only strict scrutiny, but also rational basis review. There is simply no rational relationship between the asserted objectives and the means being used to achieve them, since the restrictions imposed will be completely ineffective in addressing the problems identified. First, the problems of access and prices of real estate are mostly supply problems, not demand. Second, the countries that are singled out in the new wave of alien land laws completely fail to reflect the nationalities of the largest foreign landowners. Third, these laws are argued as ways to ensure food security, but food security is not a problem in the United States. To the extent that food security embodies consolidation in the agricultural sector and absentee ownership, alien land laws do not solve the food security problem because the real culprits are domestic corporations and corporations from countries that are not mentioned in any of the alien land laws. Fourth, from a national security perspective, foreign adversaries who want to spy on the U.S. are likely to use methods that do not require a land base near the target.

The few cases where courts have upheld alien land laws under rational basis review are distinguishable from many of the current laws because those laws were different in scope and did not single out specific nationalities. For example, the Eighth Circuit upheld a Nebraska constitutional provision prohibiting agricultural land ownership by non-family corporations.²⁸⁵ In addition, the Wisconsin Supreme Court upheld a law that limited ownership of land by “nonresident aliens” to 640 acres.²⁸⁶ The court found the law to be rationally related to the legitimate goal of preventing absentee ownership, stating that “limiting the benefits of land ownership to those who share in the responsibilities and interests of residency is not an unreasonable exercise of legislative choice.”²⁸⁷

The fate of the new laws may be different, especially if they single out specific countries. Laws targeting citizens, corporations, and governments of China, Iran, North Korea, Russia, and other countries on various federal lists are much more ineffective (and more insidious) than the laws considered in these prior decisions, which treated all nonresident aliens equally. If the targeted countries’ citizens and corporations own little to no real property in the state, legislators cannot rationally think that prohibiting them from owning real estate will make a dent in the problems they want to tackle. Additionally, absentee ownership is already pervasive in the agricultural sector. Targeting foreign owners as a potential solution would affect only 3% of the land in the United States if all countries were restricted. Legislators are aware that there is little overlap between the problem of absentee ownership and foreign ownership.²⁸⁸ As for statutes that prohibit landownership within a certain distance of military bases or critical infrastructure, this will do nothing to prevent cyberattacks, which pose the main threat to national security, as noted above.²⁸⁹

Another reason for questioning the rationality of the new wave of alien land laws is that availability bias appears to play a major role in legislators’ decisions. Availability bias is the human tendency to use information that comes to mind quickly and easily when making decisions.²⁹⁰ It is an unconscious mental shortcut that circumvents taking all evidence into consideration. Because a few incidents that involved foreign investors made national news, the new wave of alien land laws was spurred.

In addition to the wind farm project planned by Mr. Sun in Texas, there were two other prominent incidents. One involved a Saudi-owned company called Fondomonte that was leasing public land in Arizona and draining the groundwater supply to grow alfalfa for export back to Saudi Arabia, where alfalfa farming was prohibited due to water scarcity. The company paid relatively little to lease the land in Arizona and got the water for free, while Americans in the surrounding area paid extremely high costs for water.

The other case involved a Chinese food manufacturer that tried to purchase 300 acres of agricultural land in North Dakota located twelve miles from the Grand Forks Air Force Base.²⁹¹ The federal government’s CFIUS reviewed this case and determined that it did not have jurisdiction over the transaction because the Grand Forks Air Force Base was not on its list of military installations.²⁹² This led people to believe that the federal government’s process was inadequate and that states needed to take more action. The Grand Forks incident was relied on not only by legislators in North Dakota, but also by other states including Florida.²⁹³ Yet, as previously noted, the law that Florida ultimately passed would not have stopped such an investment, since it prohibited Chinese foreign investment within ten miles of military installations. The arbitrariness and ineffectiveness of the laws suggest that decisions were driven by implicit biases rather than carefully studied facts. Worse yet, they may have been motivated by animus, as discussed below.

ii. Rational Basis with Bite

To the extent that recent alien land laws are motivated by animus toward China or another country, courts may apply “second order” rational review, also known as rational basis “with bite.”²⁹⁴ In such cases, the Supreme Court has found the government’s interest to be illegitimate because it is motivated by prejudice. The Court has considered a poor means-end fit to be a signal that an illegitimate interest may be motivating the law.²⁹⁵

Comments made by politicians around the time that the recent wave of alien land laws started being proposed certainly suggest that anti-Chinese animus played a role. For example, in 2022, a candidate who competed in the Republican primary for a Texas House seat tweeted, “China created a virus that killed hundreds of thousands of Americans.”²⁹⁶ Former President Trump also continued to call COVID-19 the “China virus” throughout 2022.²⁹⁷ Each tweet from Trump that mentioned “China” and “COVID” together resulted in an 8% increase in anti-Asian hate incidents and tweets with racial slurs.²⁹⁸

Politicians further fanned the flames of anti-Chinese animus by presenting China as a threat to the American way of life. A U.S. Representative from Indiana accused President Biden of “turning a blind eye to CCP spies abusing our visa system.”²⁹⁹ A U.S. Senator from Tennessee warned that “[t]he CCP is attempting to take over the USA across all industries—pushing spies into U.S. universities and buying U.S. farmland.”³⁰⁰ Vice President J.D. Vance, a former Senator from Ohio, analogized U.S. economic dependence on China to slavery when he was running for his Senate seat, stating: “When our farmers go bankrupt the Chinese who sell the fertilizer will happily buy up their land. This is the pathway to national slavery.”³⁰¹ The Washington Post and other outlets have also highlighted how “anti-Asian bigotry” is behind the new alien land laws targeting China.³⁰²

In City of Cleburne v. Cleburne Living Center, Inc., a classic case on rational basis with bite, the Court focused on the lack of “fit” between the language of a zoning ordinance and a town’s asserted objectives for denying a special permit to a group home for people with mental disabilities.³⁰³ The town claimed that the purpose of the ordinance and permit process was to avoid congestion and ensure safety in the event of a fire or flood, but the Court pointed out that the permit process did not apply to hospitals, nursing homes, dormitories, and other uses that could be expected to pose greater problems than a group home.³⁰⁴ This poor means-end fit supported the Court’s conclusion that the ordinance had an illegitimate purpose based on animus.

Similarly, the underinclusive nature of alien land laws that target countries with minimal investments in U.S. land, while omitting the countries with the largest investments, demonstrates a poor means-end fit if the asserted objectives are to protect food security and prevent absentee landownership. These laws also generally “grandfather” in ownership of existing properties, which some commentators have identified as another signal of underinclusiveness that can trigger heightened “rational basis” review.³⁰⁵

In sum, regardless of whether strict scrutiny or rational basis review applies, alien land laws targeting specific countries should be struck down. They are not rationally related to a legitimate government interest, much less narrowly tailored to a compelling government purpose, and they appear to be motivated, at least in part, by impermissible animus.

C. Preemption Concerns

Whether alien land laws are preempted by federal law is another important constitutional question. This Section explores whether alien land laws are preempted by federal immigration laws, the federal government’s national security and foreign affairs powers, and the CFIUS and USDA reporting regimes.

1. Immigration Preemption

The Immigration Act of 1952 established “a comprehensive federal statutory scheme for regulation of immigration and naturalization” and set “the terms and conditions of admission to the country and the subsequent treatment of aliens lawfully in the country.”³⁰⁶ Supreme Court precedents indicate that alien land laws restricting noncitizens who have already been admitted to the U.S. may be preempted by federal immigration law. In Takahashi, the Supreme Court explained that “[s]tate laws which impose discriminatory burdens upon the entrance or residence of aliens lawfully within the United States conflict with [the] federal power to regulate immigration, and have accordingly been held invalid.”³⁰⁷ Both Takahashi, and an earlier case, Truax v. Raich, struck down state laws limiting the employment of lawfully present noncitizens by reasoning that federal immigration law granted a “privilege to enter and abide in ‘any state in the Union,’ ” and that denying the right to work would be “tantamount to . . . deny[ing] them entrance and abode.”³⁰⁸

In Richardson, the Supreme Court confirmed that states may not impose an “auxiliary burden[] upon the entrance or residence of aliens” that Congress had never contemplated.³⁰⁹ Restrictions on ownership of real property impose precisely this type of auxiliary burden. Certainly, Congress never contemplated that lawful permanent residents would be encumbered by ownership restrictions. With respect to temporary immigrants (i.e., “nonimmigrants”), Congress required certain classes, such as tourists, students, and crewman, to maintain a residence abroad that they had no intent of abandoning.³¹⁰ But for other classes of temporary immigrants, Congress did not impose any such requirement.³¹¹ The Supreme Court has interpreted this silence “to mean that Congress . . . was willing to allow nonrestricted nonimmigrant aliens to adopt the United States as their domicile.”³¹² If every state could prohibit temporary immigrants from buying—or potentially even leasing—property, the doors of the United States would effectively be closed to when Congress permitted them to establish domicile here.³¹³ As a federal court in Texas recognized, “[r]estrictions on residence directly impact immigration in a way that restrictions on employment or public benefits do not.”³¹⁴

While lawfully admitted immigrants may have the strongest argument for immigration preemption, courts have also struck down state laws that discriminate against undocumented individuals in housing as preempted by federal immigration law.³¹⁵ For example, in Villas at Parkside Partners v. City of Farmers Branch, the Fifth Circuit found that immigration law preempted a local ordinance that prohibited renting to individuals who are not “lawfully present.”³¹⁶ The court reasoned that Congress contemplated that such individuals would reside in the United States until potential deportation and even required them to provide a reliable address to the federal government.³¹⁷ Additionally, the court noted that deciding whether someone is “lawfully present” requires a complex analysis and should be made only by federal immigration officials.³¹⁸ The same reasoning would support striking down Louisiana’s newly enacted alien land law, which exempts noncitizens who are “lawfully present in the United States” and would therefore require a state official to make a determination about someone’s legal status.³¹⁹

Additionally, in City of Farmers Branch, the court was concerned about the immigration classification in the local ordinance being “at odds” with a much more nuanced federal regime.³²⁰ Some of the proposed and enacted alien land laws raise similar concerns by using terms that conflict with immigration law. For example, Minnesota’s law defines a “permanent resident alien” to include not only someone who is a lawful permanent resident, but also a nonimmigrant treaty investor.³²¹ A bill proposed in West Virginia defined a “nonresident alien” as someone who is neither a U.S. citizen nor a lawful permanent resident. Under that definition, all sorts of noncitizens would be swept into the restriction, even if they live in West Virginia.

Finally, the registration and reporting requirements found in some alien land laws may be preempted by immigration law. In Hines v. Davidowitz, the Supreme Court found that immigration law preempted a Pennsylvania statute requiring adult aliens to register with the state, pay a fee, and carry an ID.³²² Likewise, in Arizona v. United States, the Court stressed that “the Federal Government has occupied the field of alien registration.”³²³ The Court explained that “[t]he federal statutory directives provide a full set of standards governing alien registration, including the punishment for noncompliance.”³²⁴ A state law that requires certain noncitizens to register their property, and penalizes them for failing to do so, is not far afield from one requiring noncitizens to register themselves, especially since the same personal information must be provided.

The arguments presented above all involve noncitizens who are in the United States. The major group omitted from this analysis of preemption by federal immigration laws are noncitizens abroad. But the other bases for preemption, discussed below, would apply to that group.

2. Foreign Affairs Preemption

The Constitution entrusts foreign affairs powers exclusively to the federal government.³²⁵ Foreign affairs preemption serves several purposes: it constrains a state’s ability to offend a foreign country, which could lead to hostilities; it promotes unity in the nation’s external affairs; and it furthers the effective exercise of foreign policy.³²⁶ Yet, as the history of alien land laws shows, states have long engaged with issues that affect foreign nationals.³²⁷

The Supreme Court has provided different versions of the test for determining whether a state law impermissibly interferes in foreign affairs. In American Insurance Ass’n v. Garamendi, which struck down California’s “Holocaust-era” insurance legislation, the Court framed the issue as whether the state law is likely to produce “more than [an] incidental effect in conflict with express foreign policy.”³²⁸ In Zschernig, the Supreme Court invalidated an Oregon probate law that permitted states courts to withhold remittances to nonresident aliens residing in Communist countries.³²⁹ Even though states traditionally have the power to regulate estates and probate, the Court found that the Oregon law “affect[ed] international relations in a persistent and subtle way.”³³⁰ There, the Court framed the test as whether the state law “impair[s] the effective exercise of the Nation’s foreign policy.”³³¹ And in Crosby v. National Foreign Trade Council, which struck down a Massachusetts law that barred state agencies from purchasing goods or services from companies doing business with Burma, the Court considered whether the state law “stands as an obstacle to the accomplishment and execution of the full purposes and objectives of [federal policy].”³³²

Applying these cases to alien land laws, the question is whether, or to what degree, they conflict with U.S. foreign policy or pose an obstacle to the objectives of foreign policy. Do they merely have an incidental impact on foreign affairs, or is the effect more material? While the answer will likely depend on the specifics of a particular law, it is also worth considering the cumulative impact of these alien land laws on foreign affairs. If every state prohibited citizens of China from buying property, the impact on foreign relations would be far more significant than if only a few did so.

State laws that unilaterally identify certain nations as “countries of concern” or “foreign adversaries,” with no reference to a federal law, are particularly likely to raise foreign affairs preemption concerns. Like the Massachusetts law struck down in Crosby, these state laws are making a judgment about the conduct of a foreign country that is “apart from the federal government’s own announced judgment.”³³³ Even if the countries identified by the state law are currently consistent with a federal designation, federal law expressly contemplates those designations changing over time, and state laws may not keep up with them.³³⁴ Some lower courts have already expressed preemption concerns about state laws that are directed at particular nations, noting that they can be perceived as a unilateral declaration of “economic war,”³³⁵ or a “political statement” about the country.³³⁶ As one court recognized, the potential effect on international relations is greater when a state targets a specific country instead of regulating all noncitizens regardless of nationality.³³⁷

Additionally, the countries identified by name in the new wave of alien land laws are already subject to individualized sanctions by the federal government. Several Presidents have issued Executive Orders and Congress has passed laws imposing unique sanctions against China,³³⁸ Iran,³³⁹ North Korea,³⁴⁰ and Russia,³⁴¹ among other countries. Just like the sanctions against Burma discussed in Crosby, the laws addressing sanctions against these countries give the President flexible authority over what sanctions to impose and empower the President to waive any sanctions in the interest of national security. In Crosby, the Court reasoned that Congress would not have “gone to such lengths to empower the President if it had been willing to compromise his effectiveness by deference to every provision of state statute or local ordinance that might, if enforced, blunt the consequences of discretionary Presidential action.”³⁴²

Like the Massachusetts law in Crosby, alien land laws that target countries subject to federal sanctions “impos[e] a different, state system of economic pressure,” “penalize[] some private action that the federal [laws] . . . may allow, and pull[] levers of influence that the federal [law] does not reach.”³⁴³ The restrictions imposed by the alien land laws also make it impossible for the President “to restrain fully the coercive power of the national economy” by lifting or promising to lift sanctions, which leaves the President with “less to offer and less economic and diplomatic leverage as a consequence.”³⁴⁴ These state laws could also conflict with the federal sanctions scheme by flatly prohibiting financial transactions that the OFAC might permit with a license.³⁴⁵

When Congress wanted state and local governments to play a role in sanctioning a country such as Iran, Congress explicitly authorized them to do so.³⁴⁶ The 2010 Comprehensive Iran Sanctions, Accountability, and Divestment Act specified what form such state sanctions could take (divestment from companies that invest $20 million or more in Iran’s energy section), stated that such laws were not preempted, and protected due process by requiring notice and the opportunity for a hearing.³⁴⁷ Without this explicit congressional authorization, however, such sub-federal sanctions would likely be preempted by either the statute or executive action.³⁴⁸

Alien land laws that avoid naming specific countries but rely on various federal designations raise similar preemption concerns. The federal government has already determined the unique purposes and consequences of each of these designations. Adding restrictions involving real property ownership to whatever consequences the federal government has already imposed interferes with the federal scheme. For example, if the Secretary of State designates a country as “of particular concern,” Congress has authorized fifteen specific “Presidential Actions” that may be imposed on such designated countries, as well as any “commensurate action.”³⁴⁹ The President is also authorized to waive the application of any action.³⁵⁰ State laws that restrict real property ownership by citizens or entities of these “countries of particular concern” add consequences that were never contemplated by Congress and that can undermine the President’s decisions.

The variation among the countries included in each federal list underscores the deliberate decisions made by federal actors about how each country should be classified based on specific foreign policy objectives. State laws that use these classifications in a completely different context distort their purpose. This preemption argument is especially strong where the federal law constrains the context in which a particular term may be used. For example, federal regulations specify that the Secretary of Commerce’s classification of certain countries as “foreign adversaries” is “solely for the purposes of ” a particular executive order.³⁵¹

Individually and collectively, alien land laws that target specific countries, either by name or based on a federal list developed for another context, “compromise the very capacity of the President to speak for the Nation with one voice in dealing with other governments.”³⁵² As the Court explained in Crosby, “the President’s maximum power to persuade rests on his capacity to bargain for the benefits of access to the entire national economy without exception for enclaves fenced off willy-nilly by inconsistent political tactics.”³⁵³

Although the argument for foreign affairs preemptions seems strong based on these Supreme Court precedents, the U.S. Department of Justice surprisingly did not assert preemption in a Statement of Interest that it submitted in the case challenging Florida’s alien land law.³⁵⁴ Its failure to do so was noted by the district court in rejecting the plaintiffs’ preemption argument.³⁵⁵ Given the weight that courts give to the federal government’s own position on preemption, the Department of Justice’s position could prove fatal to preemption arguments in other cases as well. However, in the recent Shen case, even with the silence of the federal government, the Eleventh Circuit Court of Appeals granted a preliminary injunction based on CFIUS regulation of real estate transactions.³⁵⁶

3. The CFIUS and USDA Regimes

Concerns about foreign interests in real property are not unique to States. At the federal level, there are two avenues to rein in foreign investment: data collection on foreign interests in agricultural lands by the USDA and the review of certain transactions via CFIUS. These federal regimes may preempt state restrictions on foreign investment.

i. Reporting to USDA

The Agricultural Foreign Investment Disclosure Act of 1978 (“AFIDA”) established a framework to collect reported data on foreign ownership of agricultural land.³⁵⁷ Unfortunately, the system has not been properly implemented. Inaccuracies and underreporting have been pointed out.³⁵⁸ These critiques of the incompleteness and lack of transparency of the USDA reporting system have prompted Congress to include in the Consolidated Appropriations Act for the 2023 Fiscal Year (“FY”) a mandate to USDA to report on the impact that foreign investment has on family farms, rural communities, and the domestic food supply.³⁵⁹ The Government Accountability Office is expected to issue a report on the AFIDA and USDA reporting frameworks. There are several bills being discussed in the 2023–2024 congressional term seeking to ensure compliance with AFIDA. The Not One More Inch or Acre Act would ensure higher penalties for not complying with AFIDA.³⁶⁰ Under current law, persons who have violated AFIDA are subject to a fine of up to twenty-five percent of the foreign person’s interest in the agricultural land. This bill would make the minimum fine to be ten percent. House Resolution (“H.R.”) 1789 would require the penalty to be “at least [fifty] percent” of the market value of the land.³⁶¹ S.B. 2060 (Foreign Agricultural Restrictions to Maintain Local Agriculture and National Defense Act)³⁶² would require USDA to investigate efforts to steal agricultural knowledge and technology and to disrupt the U.S. agricultural sector. S.B. 2060 would also made the Secretary of Agriculture a member of CFIUS.

ii. CFIUS

CFIUS is a system for monitoring and, if necessary, blocking foreign investments that threaten national security.³⁶³ Established by President Ford in 1975, CFIUS is an interagency committee, chaired by the U.S. Department of Treasury.³⁶⁴ If CFIUS determines that an investment poses a threat to national security, the President can block or unwind the transaction. National security is not defined for CFIUS’s purposes, leaving it open to discretion.³⁶⁵

CFIUS originally focused only on foreign investment in U.S. businesses, without reviewing any real estate transactions. But in 2018, the Foreign Investment Risk Review Modernization Act (“FIRRMA”) expanded CFIUS and the President’s authority to review and block “certain types of real estate transactions involving the purchase or lease by, or a concession to, a foreign person.”³⁶⁶ CFIUS only has authority over real estate transactions that are in or around airports and maritime ports, or that are close to certain designated military installations. FIRRMA recognized that the President may want to consider factors such as “the relationship of [the investor’s] country with the United States” and “the adherence of the subject country to nonproliferation control regimes” in deciding whether to block a transaction.³⁶⁷

CFIUS’s jurisdiction also excludes transactions involving a single housing unit or real estate in urbanized areas.³⁶⁸ Small real estate investments are not expected to have a significant impact on national security and may not encourage large investments. Certain transactions must be reported, such as those involving a foreign government or any other transaction that CFIUS’s regulation mandates, while others fall under voluntary reporting. Real estate transactions so far have not been subject to mandatory reporting, suggesting that Congress did not consider them a national security threat. Control of critical infrastructure does trigger an investigation by CFIUS,³⁶⁹ but agriculture and food systems are not specifically identified as critical infrastructure. Bills that Congress considered but did not pass would have made that connection clear.³⁷⁰ In 2022, President Biden instructed CFIUS to consider the implications of foreign investment for food security.³⁷¹

For transactions under the purview of CFIUS, CFIUS is a ceiling and states cannot strengthen the regime by imposing additional obstacles. Hence, the provisions of new alien land laws overlapping with CFIUS are preempted because they could constitute an obstacle for federal enforcement.³⁷² Because of its limited jurisdiction, CFIUS would not have the authority to review many of the individual real estate transactions prohibited by state alien land laws. For example, as noted above, CFIUS found that it did not have jurisdiction to review a Chinese food manufacturing company’s purchase of 370 acres located twelve miles from the Grand Forks Air Force Base in North Dakota.³⁷³ That air force base was not on CFIUS’s list of designated military installations. Additionally, as a practical matter, CFIUS’s review of real estate transactions is negligible. In 2022, CFIUS reviewed 285 notices of non-real estate transactions, and only one notice of a real estate transaction.³⁷⁴ But still CFIUS may operate as a deterrent.

One could argue that Congress steered clear of ordinary real estate transactions in order to allow states to exercise their traditional control over land and property.³⁷⁵ On the other hand, Congress’s decision to include certain transactions while omitting others may reflect a carefully calibrated consideration of national security and economic interests, in which case states should not be allowed to disturb the delicate balance struck by Congress.³⁷⁶ Of course, if Congress had perceived alien land laws as conflicting with federal law (the CFIUS regime), it could have taken some action. So far, however, Congress has done nothing to impede states from implementing such laws. When Congress amended FIRRMA in 2018, at least fifteen states had alien land laws,³⁷⁷ and Congress did not indicate any intent to displace those laws in the amended Act. However, in past years, bills were introduced at the federal level that would have expanded CFIUS’s jurisdiction over real estate transactions,³⁷⁸ or outright prohibited citizens of China, Russia, North Korea, or Iran from purchasing land.³⁷⁹ While the CFIUS regime is limited, states’ unilateral actions singling out certain countries threaten the unified position that CFIUS enshrines with respect to both adversaries and allies.³⁸⁰

The Eleventh Circuit Court of Appeals in the case challenging Florida’s S.B. 264 granted a preliminary injunction in favor of two of the plaintiffs and based the “likelihood of success” on the merits on the potential preemption of S.B. 264 by the carefully crafted balance of CFIUS review under FIRRMA for real estate transactions, including those near military installations.³⁸¹

D. Dormant Commerce Clause

1. Interstate Commerce

While the Commerce Clause gives power to the federal government to regulate commerce between the states, it has also been interpreted as a limit on state action. Unlike preemption doctrine, which asks whether a state law conflicts with a federal law or whether Congress has occupied the field, the Dormant Commerce Clause prohibits state or local action that restricts interstate commerce even in the absence of congressional action. The goal of the Dormant Commerce Clause doctrine is to prevent “differential treatment of in-state and out-of-state economic interests that benefits the former and burdens the latter.”³⁸²

Some alien land laws violate the Dormant Commerce Clause by treating out-of-state actors differently than in-state actors in ways that have a negative impact on interstate commerce. The disparate treatment between in-state and out-of-state residents in North Dakota’s new law is the clearest example. North Dakota’s law disadvantages noncitizens who are either abroad or in another state. It requires noncitizens who are not permanent residents or otherwise exempted to reside in the state for ten months a year. It also exempts those who actively participate in the management of the agricultural operation, which could allow someone to comply with the restrictions without being present in the state.³⁸³ A noncitizen who stops fulfilling these requirements must dispose of the property. A foreign person who moves to another state then cannot hold land while a similarly situated foreign person in North Dakota can.

Another example is the initial version of an Oklahoma bill, which exempted “any alien who is or shall become a bona fide resident of the State of Oklahoma” from the restrictions on ownership.³⁸⁴ Such a provision explicitly treats noncitizens living in another state differently than noncitizens residing in Oklahoma, which would trigger strict scrutiny under the Dormant Commerce Clause.³⁸⁵ Oklahoma likely recognized the Dormant Commerce Clause issue, because the final version of its rule pronounced that “the requirements of this subsection shall not apply to a business entity that is engaged in regulated interstate commerce in accordance with federal law.”³⁸⁶

Courts have struck down similar restrictions on landownership that favor in-state residents. For example, in Jones v. Gale, the Eighth Circuit invalidated a Nebraska initiative that amended the state constitution to ban corporations from owning farmland, with an exception for family farm businesses in which at least one family member resided or worked on the farm.³⁸⁷ The court found that this amendment favored Nebraska residents in violation of the Dormant Commerce Clause.³⁸⁸ Alien land laws that apply restrictions without differentiating based on residence in the state are much more likely to survive a Dormant Commerce Clause analysis.

2. Foreign Commerce

Restrictions on foreign ownership of land have a more obvious effect on international trade than they do on interstate commerce because noncitizens abroad are clearly targeted.³⁸⁹ North Dakota’s law, for example, allows noncitizens to buy agricultural land only if they reside in the state, while U.S. citizens and permanent residents can own agricultural land there regardless of where they live. While no country is singled out in North Dakota’s law, those countries without a treaty of friendship with the United States will be the ones whose citizens will be most affected.³⁹⁰

The Dormant Foreign Commerce Clause operates similarly to the interstate Dormant Commerce Clause, but state laws burdening foreign commerce are subjected to more demanding scrutiny.³⁹¹ When it comes to regulating foreign commerce, the Supreme Court has stressed that state laws should not “prevent this Nation from ‘speaking with one voice.’ ”³⁹² In the seminal case Japan Line, Ltd. v. County of Los Angeles, the Court highlighted the “acute” risk of retaliation by Japan for California’s imposition of a tax rule that deviated from international practice, observing that such retaliation “would be felt by the Nation as a whole,” not just by California.³⁹³

In subsequent cases, however, the Court has acknowledged the difficulty in determining “precisely when foreign nations will be offended by [a] particular act[]” or whether they might retaliate.³⁹⁴ The Court has also upheld state taxes on foreign entities by reasoning that no coherent federal policy exists.³⁹⁵

Under the Dormant Foreign Commerce Clause, it may be hard to demonstrate a uniform federal policy on foreign land ownership, since the federal government has traditionally entered into bilateral treaties with specific countries when it wanted to override state restrictions on foreign ownership of land.³⁹⁶ Additionally, in Barclays, the Supreme Court suggested that congressional inaction indicates acquiescence to differing state laws.³⁹⁷

Alternatively, courts may rely on the Supreme Court’s position in South-Central Timber Development, Inc. v. Wunnicke,³⁹⁸ which allows states to escape scrutiny under the Dormant Commerce Clause only if they are market participants themselves. For example, states could be acting as market participants when they are regulating state public lands, or when Congress has expressly excluded a state law from Dormant Commerce Clause scrutiny.³⁹⁹ In some cases, the Court has not considered references to state power over a resource, like water, in federal laws⁴⁰⁰ or in treaties⁴⁰¹ enough to conclude that Congress has excluded the application of the Dormant Commerce Clause to states. Accordingly, acknowledgement of state power to regulate property is an inadequate basis for refusing to apply the Dormant Foreign Commerce Clause when state laws discriminate against noncitizens abroad.

CONCLUSION

While each wave of alien land laws has responded to unique historical events, xenophobia of some kind undergirds them all. The current wave is no different. The dominant narratives that have fueled such bills involve members of the Chinese Community Party buying land to either spy on U.S. military bases or to “undermine American agriculture and control the global food supply.”⁴⁰² These narratives reflect a few salient examples of Chinese investments near military bases, but they have nothing to do with most foreign investment in the U.S. This Article has argued that one of the most significant weaknesses of these new laws is the complete lack of fit between the objectives asserted and the means being used to achieve them. This lack of means-end fit, combined with the availability of less restrictive alternatives, is highly relevant to both the equal protection analysis and the Fair Housing Act disparate impact analysis.

As legal cases challenging these new laws start percolating through the court system, the Supreme Court may eventually need to decide whether it will stand by hundred-year-old precedents upholding alien land laws that were based on explicitly racist naturalization eligibility criteria—rules that prohibited Asians from becoming U.S. citizens. The time has come for those cases to be overturned. But overturning them will likely require the Court to clarify certain unanswered questions in equal protection doctrine regarding alienage discrimination, such as whether strict scrutiny applies to all classes of noncitizens and whether the political functions exception to strict scrutiny can be extended to landownership.

Courts may also decide to avoid the thorny equal protection questions by striking down alien land laws on preemption grounds instead. However, the various arguments for preemption discussed here involve their own hurdles. Preemption under immigration law would likely be limited to noncitizens who have already been admitted to the U.S. Foreign affairs preemption seems particularly promising, but the federal government’s decision not to argue preemption in the recent Florida case to date may undermine that claim. A Dormant Foreign Commerce Clause argument is also strong, but courts may still be reluctant to invalidate a law related to traditional state powers over property based on interference with commerce.

There is also a chance that Congress will enact new laws in the near future addressing foreign ownership of land, as several such bills have already been proposed.⁴⁰³ Depending on the substance of a federal law, this could either make it harder or easier to challenge property restrictions related to national origin. In FY 2024, the House proposed a bill that would “prohibit the purchase of agricultural land located in the United States by nonresident aliens, foreign businesses, or any agent, trustee, or fiduciary associated with Russia, North Korea, Iran, or the Communist Party of China.”⁴⁰⁴ If the federal government decides to pass a law like this that singles out certain countries, it would be harder to challenge than a similar state law, as rational basis review, rather than strict scrutiny, applies to alienage classifications by the federal government.⁴⁰⁵ Additionally, the Dormant Commerce Clause and preemption arguments would disappear, since they only constrain states.

The enactment of federal legislation would, however, bolster arguments that state laws are preempted. A federal law that did not single out specific countries and instead set some general limits on foreign land ownership across the board, such as a limit on the amount of U.S. land that a noncitizen abroad or foreign business may own, could have a positive effect by displacing state laws that impose much more discriminatory restrictions.

A third possibility is that Congress could explicitly embrace a cooperative approach, specifying that the newly enacted federal legislation does not prohibit states from passing their own laws on foreign ownership of real property. Because this approach would potentially permit discriminatory state laws, it would be wise for Congress to at least set some constraints regarding what types of state restrictions would be permissible to prevent a race to the bottom.

At the end of the day, states and the federal government should be wary of the harm that exclusionary laws inflict. Laws that draw distinctions based on national origin or citizenship are likely to lead to racial or ethnic profiling by realtors, lenders, and others involved in real property transactions, as well as to subordinate minorities more generally. These laws are especially apt to exacerbate discrimination against Asian Americans, Iranians, and others who are already subject to discrimination. Long ago, the Supreme Court recognized that if states were allowed to deny immigrants the right or live and work in their borders, immigrants “would be segregated in such of those States as chose to offer hospitality.”⁴⁰⁶ Alien land laws open the door to this type of segregation.

Property ownership is a crucial means of achieving both financial and social mobility; it provides access to schools, jobs, culture, and community. Restricting property rights has therefore been used as a tool throughout history to disempower certain groups, including women and racial minorities. When we deprive noncitizens of property rights, we prevent hardworking immigrants from achieving the American Dream.

98 S. Cal. L. Rev. 305

Download

Fatma Marouf*

* Professor of Law, Texas A&M University School of Law. I would like to thank the participants of the Asian American Pacific Islander (AAPI) and Middle Eastern North African (MENA) Women in the Legal Academy Workshop, hosted by CUNY School of Law, as well as the participants of the Clinical Writers’ Workshop hosted by NYU School of Law, for their valuable feedback on a draft of this piece.

Vanessa Casado Pérez†

† Professor of Law, Texas A&M University School of Law, Research Professor, Texas A&M Department of Agricultural Economics. I would like to thank the participants at the Rural West Workshop and Grayson Ford for his research assistance. We are grateful for the hard work of the Southern California Law Review editors.

Eighth Amendment Stare Decisis

April 2025April 2025William W. Berry III*

In 2008, the United States Supreme Court decided Kennedy v. Louisiana, holding that the Eighth Amendment barred death sentences for the crime of child rape because such punishments were cruel and unusual. In 2023, Florida passed a statute that directly contravenes this constitutional rule. Under the Florida statute, committing sexual battery against a child is a capital offense.

The Court’s description of the scope of stare decisis stems from its abortion cases. The Court initially explained stare decisis in Planned Parenthood of Southeastern Pennsylvania v. Casey but arguably loosened its meaning in its decision in Dobbs v. Jackson Women’s Health Organization. Indeed, the Court’s decision in Dobbs, in which it reversed the fifty-year-old precedent of Roe v. Wade and its successor Casey, suggests that the Kennedy case could face a similar fate.

But the Eighth Amendment contains substantive doctrinal characteristics that suggest it is unique with respect to stare decisis. In particular, the Eighth Amendment’s relationship to stare decisis is unusual because the premise of the underlying doctrine is that the meaning of the Amendment will change over time. Pursuant to “the evolving standards of decency that mark the progress of a maturing society,” the Eighth Amendment expands over time to bar punishments formerly constitutional but now determined to be draconian.

As such, there become two possibilities with respect to applying stare decisis under the Eighth Amendment. First, stare decisis could mean what it means in other contexts—deferring to precedent and refusing to overrule a prior decision unless it rises to the level of the test previously set forth in Casey and now articulated in Dobbs. Alternatively, stare decisis could mean following the evolving standards of decency doctrine. This approach contemplates that the Amendment would change over time, such that stare decisis would require the overruling of precedent, moving the case law in a progressive, less punitive direction.

This Article argues for the latter reading. Specifically, the Article makes the novel claim that the Eighth Amendment has its own unique stare decisis doctrine, the doctrine moves in one direction, and such a reading of the Eighth Amendment is consistent with the Court’s decision in Dobbs.

In Part I, the Article explores the origins of the unique doctrine of Eighth Amendment stare decisis. Part II examines past and future applications of this doctrine. Finally, in Part III, the Article explains why the Court’s decision in Dobbs supports Eighth Amendment Stare Decisis.

All bad precedents have originated from good measures.

—Julius Caesar¹

INTRODUCTION

In 2008, the United States Supreme Court decided Kennedy v. Louisiana, holding that the Eighth Amendment barred death sentences for the crime of child rape because such punishments were cruel and unusual.² In 2023, Florida passed a statute that directly contravenes this constitutional rule.³ Under the Florida statute, raping⁴ a child is a capital offense.⁵

In a vacuum, one might expect the Court to strike down Florida’s statute as clearly unconstitutional in violation of the Eighth Amendment based on the principle of stare decisis. Traditionally, the concept of stare decisis has referred to the obligation of the Court to follow prior precedent.⁶ A concept central to the rule of law, stare decisis presumes the binding nature of a prior decision, except under certain circumstances that allow for the reversing of the precedent to remedy an incorrect decision.⁷

The Court’s description of the scope of stare decisis stems from its abortion cases. The Court initially explained stare decisis in Planned Parenthood of Southeastern Pennsylvania v. Casey,⁸ but arguably loosened its meaning in its decision in Dobbs v. Jackson Women’s Health Organization.⁹

In Casey, the Court explained that while stare decisis is “not an ‘inexorable command,’ ”¹⁰ its application relates to “a series of prudential and pragmatic considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of law.”¹¹ Specifically, the Court examined (1) whether the central rule has become unworkable;¹² (2) whether the Court could remove the rule’s limitation on state power without serious inequity to those who have relied upon it or significant damage to the stability of the society governed by it;¹³ (3) whether the law’s growth in the intervening years has left the precedent’s central rule a doctrinal anachronism discounted by society;¹⁴ and (4) whether the precedent’s premises of fact have so far changed as to render its central holding somehow irrelevant or unjustifiable in dealing with the issue it addressed.¹⁵

But in Dobbs, the Court adjusted the stare decisis test, using a five-factor inquiry in deciding to overrule Roe v. Wade and Casey.¹⁶ Specifically, the Court examined (1) the nature of the court’s error, (2) the quality of its prior reasoning, (3) the workability of the current standard, (4) the effect on other areas of law, and (5) the reliance interests in the precedent.¹⁷ One way to read this shift is as a means of freeing the Court to reverse precedents it thinks are normatively incorrect.

Indeed, the Court’s decision in Dobbs,¹⁸ in which it reversed the fifty-year-old precedents of Roe v. Wade¹⁹ and its successor Casey,²⁰ suggests that the Kennedy case could face a similar fate if the Court normatively disagrees with the outcome in that case.²¹ A more open-ended view of stare decisis, in which the Court places more weight on getting the “right” answer as opposed to following its precedent, could incentivize the Court to focus on policy over precedent.²²

Likewise, a cursory glance at the Supreme Court’s Eighth Amendment cases suggests that the principle of stare decisis may carry less weight in this context.²³ For instance, the Court reversed its decisions in Penry v. Lynaugh²⁴ and Stanford v. Kentucky²⁵ a mere thirteen and sixteen years later in Atkins v. Virginia²⁶ and Roper v. Simmons, ²⁷ respectively.

But the Eighth Amendment contains substantive doctrinal characteristics that suggest it is unique with respect to stare decisis. In particular, the Eighth Amendment’s relationship to stare decisis is unusual because the premise of the underlying doctrine is that the meaning of the Amendment will change over time.²⁸ Pursuant to “the evolving standards of decency that mark the progress of a maturing society,” the Eighth Amendment expands over time to bar punishments formerly constitutional but now determined to be draconian.²⁹

As such, two possibilities exist for applying stare decisis to Eighth Amendment decisions. First, stare decisis could mean what it means in other contexts—deferring to precedent and refusing to overrule a prior decision unless it rises to the level of the test previously set forth in Casey and now articulated in Dobbs. Alternatively, stare decisis could mean following the evolving standards of decency doctrine. This approach contemplates that the Amendment would change over time, such that stare decisis would require overruling of precedent, moving the case law in a progressive,³⁰ less punitive direction.³¹

This Article argues for the latter reading. Specifically, the Article advances the novel claim that the Eighth Amendment has its own unique stare decisis doctrine, the doctrine moves in one direction, and such a reading of the Eighth Amendment is consistent with the Court’s decision in Dobbs.

I. ORIGINS OF EIGHTH AMENDMENT STARE DECISIS

Stare decisis, at its core, reflects a commitment to the rule of law.³² A vestige of the common law, the idea relates to honoring past decisions for the sake of predictability and consistency.³³ Cases with difficult factual situations challenge this paradigm.³⁴ When a rule of law generates unfair or inequitable outcomes, courts often elect to change the rule or distinguish the case such that the rule becomes inapplicable.³⁵

A more consequential decision, however, relates to a decision to reject the rule itself and replace the rule with a new one.³⁶ Courts seem hesitant to engage in such a rejection of stare decisis without a strong normative reason for doing so.³⁷

Interpreting constitutional language adds an additional wrinkle to the stare decisis calculation.³⁸ The Court has noted that stare decisis should carry less weight in the constitutional context.³⁹ This is precisely because the Court is responsible for defining the scope and meaning of the Constitution, which often includes open-ended language.⁴⁰ The inability to easily amend the federal Constitution means that the Court’s interpretation is not subject to review and will change only when the members of the Court change.⁴¹ When such decisions include placing limits on the power of state legislatures or Congress, the countermajoritarian difficulty arises.⁴²

And yet, in Marbury v. Madison, the Court made clear that its constitutional role is to engage in such judicial review, deciding who decides the scope and meaning of the Constitution.⁴³ The Court usually decides that it is its role to determine the meaning of the Constitution.⁴⁴ The Court has further explained that when it has made such determinations incorrectly, it has the responsibility to push aside the mandates of stare decisis and change the applicable constitutional rule.⁴⁵

What happens, though, when the precedent itself envisions that the rule will change over time, is different. The Eighth Amendment contemplates that the line between acceptable and unacceptable punishment will shift as society matures.⁴⁶ As such, the stare decisis tension at the heart of Casey and Dobbs dissipates. Instead, applying stare decisis means changing the rule.

A. The Evolving Standards Test

The evolving standards test originates from the 1910 case of Weems v. United States.⁴⁷ In Weems, the Court considered whether a punishment of cadena temporal—fifteen years of hard labor—for the crime of forgery constituted a cruel and unusual punishment under the Eighth Amendment.⁴⁸

In finding that the cadena temporal punishment was unconstitutional, the Court explained its approach to interpreting the Eighth Amendment:

Legislation, both statutory and constitutional, is enacted, it is true, from an experience of evils, but its general language should not, therefore, be necessarily confined to the form that evil had theretofore taken. Time works changes, brings into existence new conditions and purposes. Therefore a principle to be vital must be capable of wider application than the mischief which gave it birth. This is peculiarly true of constitutions.⁴⁹

The Court added that constitutional provisions “are not ephemeral enactments, designed to meet passing occasions,” but instead seek to “approach immortality as nearly as human institutions can approach it.”⁵⁰

Almost fifty years later, the Court further developed the concept that the Eighth Amendment did not contain a static meaning, but one that would change over time. In Trop v. Dulles, the Court considered the constitutionality of the punishment of loss of citizenship for wartime military desertion.⁵¹ Specifically, the Court considered whether permanently denying Trop a passport constituted a cruel and unusual punishment.⁵²

In finding for Trop, the Court explored the meaning of the Eighth Amendment.⁵³ Citing Weems, the Court echoed the idea that “the words of the [Eighth] Amendment are not precise, and that their scope is not static.”⁵⁴ As a result, “[t]he [Eighth] Amendment must draw its meaning from the evolving standards of decency that mark the progress of a maturing society.”⁵⁵

Having cemented the idea that the Eighth Amendment would evolve over time in a progressive way, the Court later established a test to determine whether a particular punishment violated society’s evolving standards of decency. In Coker v. Georgia, the Court developed this test in assessing whether a punishment of death for the crime of rape was constitutional.⁵⁶

As established in Coker, the Court’s inquiry contains two parts—an objective assessment and a subjective component.⁵⁷ The objective determination seeks “guidance in history and from the objective evidence of the country’s present judgment” concerning the punishment in question.⁵⁸ In Coker, the Court looked to the number of jurisdictions that allowed death sentences for the crime of rape, finding that Georgia was the only state allowing that punishment where the victim was an adult woman.⁵⁹ Its assessment of the objective indicia also included jury verdicts, which revealed that Georgia juries only imposed death sentences in six out of sixty-three cases involving the crime of adult rape.⁶⁰

After finding that the objective evidence revealed that the punishment of death for rape was inconsistent with the societal standards of decency, the Court “brought to bear” its own independent judgment concerning the constitutionality of the punishment.⁶¹ This judgment constituted an assessment of the proportionality of the punishment in light of the crime committed and the characteristics of the perpetrator.⁶² As the Court developed this subjective inquiry in later cases, it increasingly relied on the purposes of punishment—retribution, deterrence, incapacitation, and rehabilitation—to determine whether a punishment was proportionate.⁶³

Following the Court’s precedents in applying the Eighth Amendment, then, means applying the evolving standards of decency test to determine whether a punishment is cruel and unusual. It is worth noting that the Court initially cabined the application of this test to capital cases, because “death is different.”⁶⁴ It subsequently expanded the test to include juvenile life-without-parole sentences, because “children are different too.”⁶⁵

Practically, this means that stare decisis—following prior precedent—contemplates changing the rule to reflect the evolving standards of society. So, overruling a prior precedent would actually be following the doctrine when the move is from a harsher punishment to a less harsh punishment. The doctrine also makes clear, however, that this concept operates only in one direction—from more severe punishment to less severe punishment.

B. Why It Moves in One Direction

The Court’s Eighth Amendment cases demonstrate why the Eighth Amendment only changes in one direction—with increasing limits on the power of state and federal governments to impose draconian punishments. In particular, the Eighth Amendment values of dignity and proportionality underscore this point.⁶⁶

In its decision in Trop, the Court emphasized that “[t]he basic concept underlying the Eighth Amendment is nothing less than the dignity of man.”⁶⁷ In other words, when the Eighth Amendment bars a particular punishment practice, it reflects the conclusion that a particular punishment treats the defendant “as an object”⁶⁸ beyond what society deems as “civilized, decent, and virtuous.”⁶⁹

The Court has made clear that it “look[s] to the evolving standards of decency that mark the progress of a maturing society” to “enforce” this “duty of the government to respect the dignity of all persons.”⁷⁰ If the society matures to find a formerly acceptable form of punishment to violate a person’s dignity, then the punishment cannot, by definition, become constitutional again at some later date. Indeed, an undignified punishment or a punishment that objectifies an inmate cannot, at a later date, magically become dignified or civilized, decent, and virtuous. If the “Eighth Amendment’s protection of dignity reflects the Nation we have been, the Nation we are, and the Nation we aspire to be,” as the Court has explained, that means that over time, the United States will discard more draconian forms of punishment in favor of more humane ones.⁷¹

A change operating in the other way, from less severe punishment to more severe punishment, contravenes the core principle of the evolving standards.⁷² The evolving standards “mark the progress of a maturing society,” and increasing punishment severity undercuts that very progress.⁷³ This is particularly true concerning the punishments at issue—the death penalty and life without parole. Many states⁷⁴ and Western nations⁷⁵ have abandoned the death penalty, and the United States remains the only nation that allows juvenile life-without-parole sentences.⁷⁶ Undoing limits on punishments that most of the rest of the civilized world abolished long ago would reflect a move away from societal maturation and instead embrace societal savagery. Such a move would be the antithesis of promoting human dignity.

A second principle that the Court has linked to the evolving standards of decency—proportionality—similarly demonstrates why the Eighth Amendment only moves in one direction. The Court has explained that the evolving standards test is a tool by which to measure “the requirement of proportionality contained within the Eighth Amendment.”⁷⁷ As with dignity, the Court has long emphasized the concept of proportionality as “central to the Eighth Amendment.”⁷⁸ That means that when the Court bars particular punishments under the Eighth Amendment, it is because the punishment is excessive in light of the characteristics of the offense⁷⁹ or the characteristics of the offender.⁸⁰

Under the evolving standards test, the proportionality inquiry looks at the objective indicia of national consensus in that the sentence is excessive in light of what other jurisdictions permit and impose.⁸¹ And under the subjective indicia, the Court assesses whether the sentence is disproportionate in light of the purposes of retribution, deterrence, incapacitation, and rehabilitation.⁸²

For a barred punishment to again be constitutionally permissible, it would mean that the consensus against the punishment has reversed. Such a scenario is unlikely because it would involve states implementing punishment practices in violation of the Constitution. One or more states, like Florida and Tennessee currently, might engage in a barred punishment practice, but such actions would not be enough to create a consensus to allow that kind of punishment again.⁸³

In addition to a change in national consensus, a reversal would also mean that the concept of proportionality would have a fickle application. When a punishment is excessive, whether in light of retribution or one of the utilitarian purposes of punishment, it cannot magically become proportionate again. The argument would be that the initial determination was incorrect, that the Court defined a proportionate punishment as a disproportionate one.

The cautiousness of the Court’s evolving standards doctrine, though, makes such a claim less persuasive. All of its decisions to find punishments disproportionate under the Eighth Amendment have first found a majoritarian objective consensus⁸⁴ against the punishment in question before also finding the punishment disproportionate in its own subjective judgment.⁸⁵

A view of the Eighth Amendment as moving only in a more progressive direction is also consistent with its original meaning.⁸⁶ As John Stinneford has explained, the concept of “unusual” reflects a notion of longstanding usage.⁸⁷ Drawing on the writings of Edward Coke as well as the common law, this original understanding reflected a proscription against cruel innovation—the adoption of newer methods of harsh punishment.⁸⁸ The idea is that moving in a harsher direction undoes the original Eighth Amendment meaning of contrary to long usage, even if the evolving standards evolved in a more punitive direction.⁸⁹ Under either an evolving standards reading or under an originalist reading, then, it is clear that the Eighth Amendment can change in only one direction—expanding to bar harsh punishments.

II. APPLICATIONS OF EIGHTH AMENDMENT STARE DECISIS

While not describing its application of the Eighth Amendment as a unique form of stare decisis, the Court has nonetheless followed this approach on several occasions. And, as discussed, the national consensus continues to evolve.

A. Past Applications

Arguably, the first application of the concept of evolving stare decisis was outside of the Eighth Amendment, before the Court articulated the details of its test in Coker. But the discussion begins here because the sentiment is the same—promoting a more progressive, humane form of punishment by placing constitutional limits on a draconian one.

1. McGautha and Furman

In 1971, the Court considered the constitutionality of the death penalty in two companion cases, McGautha v. California⁹⁰ and Crampton v. Ohio.⁹¹ These challenges made Fourteenth Amendment claims, specifically that the procedures used to impose the death sentences violated due process.⁹² Both claimed that the lack of guidance given to the jury determining the sentence allowed the imposition of the death sentence without any governing standards.⁹³ Crampton also challenged the unitary trial procedure in which the jury determined guilt and punishment at the same time.⁹⁴

In a 6–3 decision, the McGautha court rejected petitioners’ arguments, finding that the Ohio and California sentencing procedures were constitutional.⁹⁵ Examining the history of the death penalty, the Court surmised that sentencing discretion in capital cases constituted a form of mercy, not the application of a generalizable concept or standard.⁹⁶ While recognizing the force of petitioners’ claim on a general level, the Court nonetheless emphasized the indeterminacy of the task of developing an applicable standard for capital juries.⁹⁷ It explained, “[t]o identify before the fact those characteristics of criminal homicides and their perpetrators which call for the death penalty, and to express these characteristics in language which can be fairly understood and applied by the sentencing authority, appear to be tasks which are beyond present human ability.”⁹⁸

Even though the criteria given to the juries in McGautha and Crampton did not do more than exercise “minimal control” of the jury’s “exercise of discretion,” the Court found it “quite impossible to say that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is offensive to anything in the Constitution.”⁹⁹ This was because “[t]he infinite variety of cases and facets to each case would make general standards either meaningless ‘boiler-plate’ or a statement of the obvious that no jury would need.”¹⁰⁰

Likewise, the Court found that the unitary trial procedure of forcing a defendant to choose between arguing innocence and arguing for mercy did not violate due process because requiring that difficult choice was not a denial of process.¹⁰¹ For similar reasons, the Court likewise concluded that the unitary trial model did not infringe upon Crampton’s Fifth Amendment privilege against self-incrimination.¹⁰²

Just a year later, the Court considered the constitutionality of the death penalty under the Eighth Amendment in Furman v. Georgia.¹⁰³ Based on similar arguments to the ones raised in McGautha, the Court held 5–4 that the death penalty was unconstitutional as applied.¹⁰⁴

Unlike the later examples of Eighth Amendment stare decisis, the decision in Furman turned on the procedure in question, not the substance.¹⁰⁵ But the idea is the same—moving from a more draconian procedure to a less draconian one. The failure to provide juries guidance on how to differentiate between murderers who should receive the death penalty and those who should not resulted in sentencing outcomes that the Court found to be random and arbitrary.¹⁰⁶ Imposing death sentences in an arbitrary and random manner was particularly troubling because “death is different”—the consequence is severe and irrevocable.¹⁰⁷

So, the decision in Furman followed the underlying principle of the evolving standards of decency—protecting the dignity of criminal defendants by preventing states from subjecting them to arbitrary, random sentencing procedures in capital cases.¹⁰⁸ The Court did not find that the death penalty itself was now cruel and unusual; instead, it was the unprincipled ways that Georgia imposed it that made it unconstitutional.¹⁰⁹ Capital punishment without any jury guidance was the prevailing practice, and the Court found that it no longer constituted a constitutional punishment.¹¹⁰

Penry and Atkins

The Court’s cases concerning whether it is constitutional to execute an intellectually disabled¹¹¹ offender provide a clear example of the application of Eighth Amendment stare decisis.¹¹² In Atkins v. Virginia, the Court reversed its decision in Penry v. Lynaugh as a matter of stare decisis because the standard of decency had changed.¹¹³

In 1989, the Court decided Penry.¹¹⁴ Penry brutally raped, beat, and stabbed Pamela Carpenter with a pair of scissors, causing her subsequent death a few hours later.¹¹⁵ A Texas jury sentenced Penry to death despite his claims of intellectual disability and insanity.¹¹⁶ As part of his habeas appeal, the Court considered whether the Eighth Amendment barred his execution in light of his intellectual disability and resulting diminished culpability.¹¹⁷

The Court found no evidence of a national consensus against the execution of intellectually disabled offenders.¹¹⁸ Only two states and the federal government barred such death sentences.¹¹⁹ Adding in the fourteen states that barred capital punishment, this meant that sixteen states barred the

execution of intellectually disabled offenders, falling short of establishing a national consensus.¹²⁰

Similarly, the Court concluded that its own subjective judgment did not bar such sentences.¹²¹ Applying the purposes of punishment, the Court held that the execution of some intellectually disabled individuals could serve the purpose of retribution—the variance among such individuals did not mean that such individuals could never act with the culpability required to receive the death penalty.¹²²

Just over a decade later, the Court considered the same question in Atkins.¹²³ It applied its evolving standards of decency test in finding that the Eighth Amendment now prohibited the execution of intellectually disabled offenders.¹²⁴

In its analysis of objective indicia, the Court found a national consensus against executing the intellectually disabled.¹²⁵ The Court noted that state legislatures had reacted to its decision in Penry as well as the execution of a different intellectually disabled inmate.¹²⁶ By 2002, thirty states barred the execution of intellectually disabled offenders, including twelve states that had abolished the death penalty.¹²⁷ This number far surpassed the number of states previously barring the punishment in question—a change from sixteen to thirty.¹²⁸ The Court also emphasized the direction of the change, a consistent move by state legislatures away from allowing the execution of intellectually disabled offenders.¹²⁹ Finally, the Court noted that states had executed only five known offenders with a known IQ under seventy since Penry.¹³⁰

With respect to the subjective indicia, the Court concluded that none of the purposes of punishment justified the execution of intellectually disabled offenders.¹³¹ The reduced culpability of intellectually disabled offenders meant that death sentences for those individuals did not satisfy the purpose of retribution.¹³² With respect to deterrence, the Court also concluded that the execution of intellectually disabled offenders was unlikely to deter other intellectually disabled individuals from committing homicides.¹³³

In overruling its decision in Penry, the Court did not address the concept of stare decisis as a hurdle that it had to overcome.¹³⁴ This is because the Court majority did not view the decision in Atkins as overturning precedent.¹³⁵ Instead, the Atkins decision followed precedent—the precedent of the evolving standards of decency doctrine—in reaching a different outcome. The decision in Atkins did not constitute an abrogation of a prior position; it constituted a foreseeable evolution in the application of a constitutional principle.¹³⁶

3. Stanford and Roper

The Court’s decision in Roper v. Simmons, three years after Atkins, provides another example of the application of Eighth Amendment stare decisis. Roper held that the execution of juveniles—offenders under the age of eighteen at the time of the homicide—violated the Eighth Amendment,¹³⁷ reversing the Court’s decision in Stanford v. Kentucky, which had allowed the execution of seventeen-year-old defendants.¹³⁸

In its first consideration of age and capital sentences, the Court held in Thompson v. Oklahoma that the execution of a fifteen-year-old defendant violated the Eighth Amendment under its evolving standards of decency test.¹³⁹ Under its objective indicia, the Court found that eighteen states set the minimum age for a capital sentence at sixteen years old.¹⁴⁰ When combined with the fourteen states that had abolished capital punishment, the Court counted thirty-two jurisdictions that barred the execution of defendants under the age of sixteen.¹⁴¹ The Court also looked to jury verdicts and found less than twenty instances of executions of individuals who committed capital crimes under age sixteen.¹⁴² And none of those verdicts had been after 1948, in the forty years prior to the case.¹⁴³

With respect to the subjective indicia, the Court highlighted the diminished culpability of juvenile offenders as a basis for finding that retribution did not support the execution of a fifteen-year-old offender.¹⁴⁴ It also found that deterrence did not support executing those under the age of sixteen who committed crimes; offenders over the age of sixteen had committed ninety-eight percent of homicides.¹⁴⁵

In Stanford v. Kentucky, the Court found that the evolving standard of decency that had reached under-sixteen-year-old offenders had not reached sixteen and seventeen-year-olds.¹⁴⁶ The Court held that the Eighth Amendment did not bar the execution of Stanford, who was seventeen when he committed murder.¹⁴⁷

With respect to the objective indicia, the Court found that most states permitted capital punishment for sixteen-year-olds.¹⁴⁸ Fifteen states rejected the death penalty for offenders under seventeen years old and twelve for offenders under eighteen years old.¹⁴⁹ The Court also rejected the evidence that few juries had sentenced under-eighteen-year-old offenders to death because so few under-eighteen-year-old offenders had committed capital crimes.¹⁵⁰

In applying the subjective indicia, the Court found no conclusive evidence supporting a determination with respect to either retribution or deterrence.¹⁵¹ And the Court did not really engage with this idea because it had found that a national consensus against executing sixteen- and seventeen-year-olds did not exist.¹⁵²

Sixteen years later, the Court decided Roper, following the same Eighth Amendment stare decisis approach used in Atkins to find that death sentences for juvenile offenders were cruel and unusual punishments.¹⁵³ As in Atkins, the application of the majoritarian objective indicia commenced with counting the state laws, and like Atkins, thirty states prohibited the execution of juvenile offenders (twelve of which banned the death penalty altogether).¹⁵⁴ Also like Atkins, the Court in Roper was assessing whether the evolving standards of decency provided enough evidence of changed circumstances to reverse its prior decision in Stanford.¹⁵⁵ The Court also noted the presence of objective evidence moving toward ending juvenile executions, although only five states (as compared to sixteen in Atkins) had abandoned the juvenile death penalty since Stanford.¹⁵⁶ Also, no state had reinstated the juvenile death penalty since Stanford.¹⁵⁷

With respect to the subjective standards, the Court developed the idea that juveniles were offenders that, by definition, possessed a diminished level of culpability.¹⁵⁸ Specifically, the Court cited (1) the lack of maturity and undeveloped sense of responsibility, (2) the susceptibility of juveniles to outside pressures and negative influences, and (3) the unformed nature of juveniles’ character as compared to adults.¹⁵⁹

In light of the diminished level of culpability, the purposes of punishment, in the Court’s view, failed to justify the imposition of juvenile death sentences.¹⁶⁰ Such death sentences failed to achieve the purpose of retribution in light of the diminished culpability.¹⁶¹ Likewise, the Court concluded that execution of juveniles did not achieve a deterrent effect—offenders with diminished capacity will be unlikely to be susceptible to deterrence.¹⁶² In addition, the Court found no evidence that a juvenile death sentence would add any deterrent value beyond that achieved by a life-without-parole sentence.¹⁶³

As with Atkins, the decision in Roper is a clear example of the principle of Eighth Amendment stare decisis. The Court followed its precedent—the evolving standards of decency—in finding that the national consensus and its subjective judgment demonstrated that the execution of juveniles constitutes a cruel and unusual punishment. As such, the decision in Roper to overrule Stanford constituted an application of Eighth Amendment stare decisis, reflecting the Court’s interpretation of the Eighth Amendment.

4. Death Is Different and Juveniles Are Different

A final important example of the Court’s application of Eighth Amendment stare decisis relates to its use in the juvenile life-without-parole context in Graham v. Florida.¹⁶⁴ Here, the Court found that a principle underlying its evolving standards of decency—differentness—had evolved to include another category of cases.¹⁶⁵

For over thirty years after Furman, the Court had cabined its application of evolving standards to capital cases.¹⁶⁶ The Court’s reasoning for this bright line focused on the idea that “death is different.”¹⁶⁷ As a punishment, death was unique both in terms of its severity—the most severe punishment available—and its irrevocability—one cannot undo a death sentence after an execution.¹⁶⁸

In Graham v. Florida, the Court considered whether the Eighth Amendment forbid life-without-parole sentences for juvenile offenders in non-homicide cases.¹⁶⁹ Building upon its decision in Roper, the Court applied the evolving standards of decency to cases of juvenile life without parole in barring such sentences in non-homicide cases.¹⁷⁰

The Court further clarified its expansion of the differentness principle to include juvenile life-without-parole cases in Miller v. Alabama, in which it struck down mandatory juvenile life-without-parole sentences under the Eighth Amendment.¹⁷¹ The Court in Miller explained that while death is different, “children are different too.”¹⁷²

As with its other applications of Eighth Amendment stare decisis, the Court in the juvenile life-without-parole cases relied on both objective and subjective understandings of the nature of juvenile offenders. The Court in Graham emphasized that only eleven states allowed life-without-parole sentences for juveniles in non-homicide cases.¹⁷³ Both cases also expanded on the conversation from Roper concerning the reduced culpability of juveniles.¹⁷⁴ Juvenile life-without-parole sentences not only make retribution and deterrence less justifiable, but also implicate incapacitation and rehabilitation, with the age of juveniles making change more possible than with older offenders.¹⁷⁵

The important point here relates to the idea that part of the evolving standards expansion includes punishments other than the death penalty. It is certainly possible that, as society evolves, other kinds of punishment, including life without parole and solitary confinement, might also violate the Eighth Amendment.¹⁷⁶

B. Distinguishable Deviations

The Court’s application of the Eighth Amendment has arguably moved in a more punitive way in a few situations, but careful examination of these cases in context shows that they are distinguishable from the concept of Eighth Amendment stare decisis and do not undermine that concept.

1. Furman and Gregg

The first example where one might argue that the Court moved in a direction favoring harsher punishment occurred when it reinstated the death penalty in Gregg v. Georgia,¹⁷⁷ four years after it had declared it unconstitutional in Furman.¹⁷⁸

The Court in Furman, however, with its per curiam opinion and five concurrences, did not rule out the future use of the death penalty.¹⁷⁹ Rather, the Court’s as-applied decision meant that the states had to remedy the flaw in the death penalty—the random and arbitrary use of it—before using it again.¹⁸⁰

Importantly, a majority of the Court did not find a consensus against the death penalty,¹⁸¹ and the response of the states—an overwhelming number immediately passing new statutes—supports the idea that, at least at that time, the evolving standard did not bar death sentences.¹⁸²

Even so, a modern examination of the Court’s decision in Gregg suggests that it is incorrect. This is because the safeguards it believed remedied the problems identified in Furman actually were insufficient to do so.¹⁸³ The number and diversity of aggravating factors that most states used in their statutes did little to narrow the class of murderers; with felony murder, almost all homicides could still be death-eligible if the prosecutor was so inclined.¹⁸⁴ Equally as important, the comparative proportionality review never occurred as promised, but instead as a diminished form of review that never included cases with life sentences.¹⁸⁵ As a result, the arbitrariness and randomness in jury outcomes persists and is perhaps even worse that it was in 1972.¹⁸⁶

2. Enmund and Tison

Another set of cases that might appear to demonstrate a move from less harsh to more harsh punishment are the Court’s decisions in Enmund v. Florida¹⁸⁷ and Tison v. Arizona.¹⁸⁸ These cases, nonetheless, are similarly distinguishable.

In Enmund, the Court considered whether a death sentence for a felony murder involving a person who did not kill, attempt to kill, or intend to kill violated the Eighth Amendment.¹⁸⁹ Of the thirty-six jurisdictions that permitted the death penalty at the time, the Court noted that only eight jurisdictions authorized the death penalty for accomplices in felony murder robbery cases like Enmund without proof of additional aggravating circumstances.¹⁹⁰ In addition, another nine states allowed death sentences for felony murder accomplices where other aggravating factors were present.¹⁹¹ The Court found that the legislative practice weighed “on the side of rejecting capital punishment for the crime at issue.”¹⁹²

In the second part of the evolving standards test, the Enmund Court brought its own judgment to bear, finding that the death sentence was inappropriate for Enmund.¹⁹³ Specifically, the Court held that his criminal culpability did not rise to the level required by just deserts retribution to warrant a death sentence.¹⁹⁴ The Court similarly dismissed deterrence as a supporting rationale for a death sentence in Enmund’s case.¹⁹⁵

Finally, it is notable that Enmund appeared to focus only on the relevant facts of Enmund’s case.¹⁹⁶ The Court did not explicitly create a categorical rule with respect to death sentences for felony murder convictions.¹⁹⁷

Tison involved the prosecution of two of Gary Tison’s sons after their father and an associate brutally murdered a family after stealing their car.¹⁹⁸ The sons participated both in helping Tison break out of prison and in the carjacking.¹⁹⁹ They were not directly present, however, at the moment when their father killed the family²⁰⁰ and were unaware that he intended to do so.²⁰¹

In assessing the jury’s imposition of death sentences on the sons, the Tison Court considered whether their punishments violated the Eighth Amendment.²⁰² The Tison Court adopted a new rule—that a capital felony murder is constitutional when the individuals in question are (1) major participants in the felony and (2) exhibit a reckless indifference to human life.²⁰³

Using the evolving standards of decency doctrine, the Court applied the same counting of state statutes as in Enmund but combined the jurisdictions that allowed felony murder for any accomplice with those that only allowed felony murder with additional aggravating circumstances.²⁰⁴ The Court reasoned that, unlike Enmund, the Tison sons played an active role in the crime (particularly the prison escape), and as a result both categories of jurisdictions should count, leading to a finding that only eleven jurisdictions did not allow death sentences in felony murder cases like Tison.²⁰⁵

The Court’s subjective judgment likewise found that the death sentences imposed on the Tison sons were not disproportionate.²⁰⁶ Specifically, the Court cited that the Tison sons’ “reckless indifference to human life” provided the intent to justify a death sentence, even though the sons did not participate in the killing itself.²⁰⁷ The distinction, then, between the outcomes in Enmund and Tison was the intent of the felony murder accomplices.²⁰⁸ Unlike in Enmund, the Tison Court made clear that the majority view did not provide a consensus view in favor of eliminating the application of the punishment at issue.²⁰⁹

So, the Court’s decisions here were not a move toward narrowing the Eighth Amendment. Rather, the Court in Tison simply qualified the scope of Enmund, which did not even impose a categorical rule in the first place. Tison did not overrule Enmund but instead reframed the inquiry. Note that the shift with respect to the act requirement moves the inquiry to the relationship of the act of the defendant to the felony, not the homicide.²¹⁰ In addition, the Tison rule keeps the mens rea connected to the homicide and captures all reckless actors.²¹¹

3. Thompson and Stanford

One might perceive that the decisions in Thompson and Stanford, discussed above, constitute a move away from the evolving standards, but like Enmund and Tison, the decisions reached parallel, but not overlapping, conclusions. Thompson barred the execution of fifteen-year-olds and younger; Stanford allowed the execution of sixteen and seventeen-year-olds.²¹² In much of the same way that Tison clarified the scope of Enmund, Stanford clarified the scope of Thompson.²¹³

4. Solem and Harmelin

The final example of the Court arguably narrowing the Eighth Amendment occurs in the cases of Solem v. Helm²¹⁴ and Harmelin v. Michigan.²¹⁵ Both of these cases concern the Eighth Amendment doctrine that the Court applies in non-capital, non-juvenile cases—the gross disproportionality doctrine.²¹⁶ These decisions parallel the opinions in Enmund and Tison, with the Court granting relief under the Eighth Amendment in the first case but using the second case to make sure that the outcome in the first case only had a narrow application.

In Solem v. Helm, the Court found that the life-without-parole sentence imposed for a bad check in the amount of $100 was grossly disproportionate in violation of the Eighth Amendment.²¹⁷ Specifically, the Court explained that the Eighth Amendment required consideration of (1) the gravity of the offense and the harshness of the penalty; (2) the sentences imposed on other criminals in the same jurisdiction; and (3) the sentences imposed for commission of the same crime in other jurisdictions.²¹⁸ Applying these concepts, the Court held that Helm’s sentence violated the Eighth Amendment because it was a far less severe crime than others for which the life-without-parole punishment—the most serious other than death—had been applied.²¹⁹ Even with the recidivist premium, the Court found that the punishment of life without parole for passing a bad check was grossly disproportionate.²²⁰

Less than a decade later, however, the Court clarified its test from Solem. In Harmelin, the Court upheld a mandatory life-without-parole sentence for a first-time offense of possession of 672 grams of cocaine.²²¹ In a 5-4 decision, the Justices in the majority splintered on the reasoning for the decision.²²² In a clear attempt to narrow Solem, Justice Scalia, joined by then-Chief Justice Rehnquist, held that the Eighth Amendment did not contain a proportionality guarantee, and therefore Harmelin’s sentence could not be unconstitutionally disproportionate.²²³ The controlling plurality, however, found that the Eighth Amendment had a proportionality guarantee,²²⁴ but that Harmelin’s sentence was nonetheless proportionate in light of the deference accorded to states in non-capital sentencing.²²⁵ Justice Kennedy determined that the Solem three-part analysis remained useful,²²⁶ but a reviewing court should consider the second and third factors—that is, the intra- and inter-jurisdictional analyses—only if “a threshold comparison of the crime committed and the sentence imposed leads to an inference of gross disproportionality.”²²⁷

Harmelin, then, did not overrule Solem. It simply qualified the gross disproportionality test, specifying that failing to pass the first part, which most cases do not, ends the inquiry.²²⁸

C. Future Applications

A cursory examination of recent trends in state punishment practices suggests that the evolving standards have already evolved to reach other kinds of punishments.²²⁹ The most obvious category of punishments is the categorical areas barred in capital cases, but not juvenile life-without-parole cases.²³⁰

The Court has identified six categories of capital punishment that the Eighth Amendment proscribes: (1) mandatory death sentences;²³¹ (2) executions of juveniles;²³² (3) executions of intellectually disabled defendants;²³³ (4) executions for certain felony murder crimes;²³⁴ (5) executions for the crime of adult rape;²³⁵ and (6) executions for the crime of child rape.²³⁶ The Court has extended some of the categorical punishment bars to juvenile life without parole, covering three of the unconstitutional capital punishment categories—mandatory juvenile life-without-parole sentences,²³⁷ juvenile life-without-parole sentences for adult rape,²³⁸ and juvenile life-without-parole sentences for child rape.²³⁹

The other categories the Court should extend the death penalty evolving standards doctrine to are (1) categorical limits on juvenile life-without parole sentences in felony murder cases like in Enmund and Tison; (2) categorical limits on juvenile life-without-parole sentences for intellectually disabled defendants like in Atkins; and (3) a categorical limit on juvenile life-without-parole altogether, mirroring the Court’s decision in Roper imposing a categorical ban on the death penalty for juveniles.

Beyond these categorical exceptions, three broad categories of punishment seem like future candidates for constitutional bars under Eighth Amendment stare decisis: the death penalty, juvenile life-without-parole sentences, and emerging adult life-without-parole sentences.²⁴⁰

1. Death Penalty

The recent move toward death penalty abolition among the states suggests that it may soon reach the evolving standards threshold of national consensus against it, if it has not already.²⁴¹ At the time of Gregg, thirty-nine states had capital statutes.²⁴² Currently, twenty-seven states allow capital punishment, but six have gubernatorial holds on executions.²⁴³ Of those twenty-seven states, fifteen have not had an execution in the past five years and thirteen have not had an execution in the past decade.²⁴⁴ Indeed, only Alabama, Arizona, Florida, Georgia, Mississippi, Missouri, Oklahoma, South Carolina, Utah, Tennessee, and Texas—eleven states—have executed anyone in the past five years.²⁴⁵

And those states are not conducting many executions.²⁴⁶ For the past five years, fewer than twenty-five executions have occurred each year, with a total of ninety-two in the period from 2019–2023.²⁴⁷ The direction of change is also clear. Since 2007, ten states have abolished the death penalty.²⁴⁸ Finally, the number of new death sentences has dropped drastically²⁴⁹ with the adoption of life without parole in almost every jurisdiction.²⁵⁰

With respect to the objective indicia of national consensus, then, the evidence is close if not already there. While there are twenty-seven capital statutes in place, only twenty-one states allow executions currently, and only twelve states have recently executed an offender.²⁵¹ The pattern of abolition, including five states in the past decade,²⁵² and the decline in death sentences also supports this conclusion.²⁵³ International consensus supports a similar conclusion, with the European Union and most Western nations having abolished the death penalty long ago.²⁵⁴

With respect to the subjective indicia, it would not be a stretch for the Court to conclude that the death penalty does not serve any of the purposes of punishment.²⁵⁵ It is certainly possible to conclude that the death penalty is an excessive punishment for the purpose of retribution.²⁵⁶ And there is strong evidence that the death penalty does not deter.²⁵⁷

2. Juvenile Life Without Parole

If there is evidence that the death penalty has contravened the evolving standards of decency under the Eighth Amendment, there is perhaps even more evidence that juvenile life-without-parole sentences also cross the constitutional line.²⁵⁸ After the Court’s 2012 decision in Miller v. Alabama, states have moved consistently in the direction of abolishing juvenile life without parole.²⁵⁹

As of 2023, thirty-three states and the District of Columbia have no one serving juvenile life-without-parole sentences, with twenty-eight of those states banning juvenile life without parole.²⁶⁰ In addition, the number of juvenile life-without-parole sentences has drastically declined over the past decade in light of the Court’s decisions in Graham,²⁶¹ Miller,²⁶² and Montgomery v. Louisiana.²⁶³ A survey of the Sentencing Project found 1,465 people serving juvenile life-without-parole sentences in January 2020, a 38% decline from 2016 and a 44% decline from 2012.²⁶⁴ With respect to international consensus, the United States remains the only country in the world that permits juvenile life-without-parole sentences.²⁶⁵

In addition to the evidence of national consensus against juvenile life without parole, it is clear that the purposes of punishment do not support these sentences. The diminished culpability of juveniles, as discussed in Roper, Graham, Miller, and Montgomery, makes it unlikely that a juvenile would deserve a life-without-parole sentence.²⁶⁶ The Court has explained this point at length in the context of the juvenile death penalty, juvenile life-without-parole sentences for non-homicide crimes, and mandatory juvenile life-without-parole sentences.²⁶⁷

It likewise seems impossible to determine that a juvenile’s “crime reflects irreparable corruption” at the time of sentencing, meaning that the utilitarian purposes of deterrence, incapacitation, or rehabilitation would not support such a sentence.²⁶⁸ In particular, the Court has emphasized the pronounced potential that juveniles have for rehabilitation.²⁶⁹

3. Emerging Adult Life Without Parole

A similar, but broader category of young offenders has also garnered judicial interest in the context of state constitutions. The Court in its juvenile life-without-parole cases recognized the diminished capacity and culpability of under-eighteen offenders.²⁷⁰ But the science supporting this understanding does not draw a bright line at age eighteen.²⁷¹ If anything, it suggests that brain development is not complete until one reaches their late twenties.²⁷²

As a result, courts have begun to consider emerging adults—offenders aged eighteen to twenty—as similar to juveniles and worthy of the same constitutional protections.²⁷³ A recent case, Commonwealth v. Mattis, demonstrates this trend.²⁷⁴ In Mattis, the Massachusetts Supreme Court struck down all life-without-parole sentences for emerging adults, individuals aged eighteen to twenty, under the state constitution.²⁷⁵

In applying the language of its state constitution,²⁷⁶ the Mattis court relied heavily on the Court’s Eighth Amendment juvenile cases—Roper, Graham, and Miller—in recognizing the “mitigating qualities of youth.”²⁷⁷ In assessing the contemporary standards of decency, the court looked to science, trends in the state, and trends in other jurisdictions.²⁷⁸ After explaining why the science overwhelmingly supports treating twenty-year-old offenders like seventeen-year-old offenders,²⁷⁹ the court looked to examples of how Massachusetts treated emerging adults more like juveniles than adults.²⁸⁰ It then surveyed other jurisdictions in finding that Massachusetts was only one of ten states that currently requires eighteen- to twenty-year-old offenders convicted of murder to receive life-without-parole sentences.²⁸¹

With Michigan and Washington reaching similar conclusions under their state constitutions, it seems possible that the Court could arrive at a similar place.²⁸² The first step would be a conclusion that emerging adults were like juveniles in that they would also be “different” for purposes of the Eighth Amendment. Then, the question would be whether a consensus existed. As the Mattis court found, most states bar mandatory life-without-parole sentences, suggesting a national consensus with respect to mandatory life-without-parole sentences for emerging adults.²⁸³ A broader application could exist if other states follow the lead of Massachusetts, Michigan, and Washington in barring the imposition of life-without-parole sentences on emerging adults.²⁸⁴

The subjective proportionality analysis would be less difficult. The scientific evidence of the similarity between juveniles and emerging adults means that the same arguments from Roper, Graham, and Miller would apply.²⁸⁵ That means that retribution, deterrence, incapacitation, and rehabilitation might not support the imposition of life-without-parole sentences on emerging adults.²⁸⁶

D. The Limit of Evolving Standards

Having mapped out the concept of Eighth Amendment stare decisis and some potential future applications, the next question is whether the doctrine limits the Court, particularly in considering laws that violate the current doctrine, such as the Florida law highlighted at the beginning of the Article. In particular, the issue is whether Eighth Amendment stare decisis would bar the Court from reversing the limits imposed in Kennedy v. Louisiana²⁸⁷ and Graham v. Florida.²⁸⁸

1. Kennedy

As discussed, the Court in Kennedy barred the imposition of death sentences for the crime of child rape.²⁸⁹ Applying the evolving standards demonstrates why the Florida law is unconstitutional. First, the evolving standards only evolve in one direction—from more severe to less severe. If the Eighth Amendment currently limits the harshest punishment for child rapes, the only direction this punishment could move is to less severe—to barring life-without-parole for child rape.

In addition, there is a clear national consensus against the death penalty for child rape as, prior to the Florida and Tennessee laws, no state has sentenced anyone to death for child rape since at least before the Court barred it in 2008.²⁹⁰ Further, as the Court explained in Kennedy, the death penalty for adult rape was rare even before Coker, much less for child rape.²⁹¹

The subjective indicia also counsel against the death penalty as a punishment for rape. The Court has made clear in Coker and Kennedy that death is an excessive punishment in most cases for non-homicide crimes, particularly sex crimes.²⁹² The Court views death as a punishment for a non-death crime as extending beyond just deserts retribution, as well as being insufficient to accomplish the purpose of deterrence.²⁹³

2. Graham

The analysis for a challenge to the rule in Graham would be almost identical. In Graham, the Court barred the imposition of life-without-parole sentences in non-homicide cases.²⁹⁴

To reverse this decision under the Eighth Amendment stare decisis rule would be impossible, as it would require the Court to move from a less harsh punishment to a harsher one in allowing juvenile life without parole for a non-homicide crime when it was previously unconstitutional.

Likewise, there is a national consensus against imposing life-without-parole sentences for non-homicide crimes committed by juveniles.²⁹⁵ If anything the evidence is even stronger than in Graham, with a majority of states having either banned juvenile life without parole or having no person serving such a sentence.²⁹⁶

And the analysis of the subjective indicia would be the same. The diminished culpability of juveniles would mean that juvenile life without parole would be a disproportionate sentence in light of the goals of retribution, deterrence, incapacitation, and rehabilitation.

Under the evolving standards, then, the Court would apply Eighth Amendment stare decisis to strike down any statute, like Florida’s, that contravened Kennedy or any state statute that contravened Graham. The one possible loophole in this analysis would relate to the Court’s decision in Dobbs, which articulated the current stare decisis standard. Section III.B eliminates that possibility by demonstrating that Eighth Amendment stare decisis is consistent with the rule in Dobbs.

III. WHY DOBBS SUPPORTS EIGHTH AMENDMENT STARE DECISIS

In considering whether the Court has latitude to overrule Kennedy, the question involves the application of Dobbs to Eighth Amendment stare decisis. As demonstrated below, the Dobbs approach to stare decisis affirms both the concept of Eighth Amendment stare decisis and the individual decision in Kennedy. The Dobbs case articulated five factors the Court should consider when weighing whether to follow its prior precedents: (1) the nature of the Court’s error, (2) the quality of its prior reasoning, (3) the workability of the current standard, (4) the effect on other areas of law, and (5) reliance interests in the precedent.²⁹⁷

A. The Dobbs Test

It is worth noting that the framework of the evolving standards of decency rests in part on an assessment of majoritarian consensus, despite its purpose of articulating a countermajoritarian right.²⁹⁸ This means that Court decisions in this area are much less likely to be products of judicial activism as they base their decisions on what they perceive to be the majority practice.²⁹⁹

1. The Nature of the Court’s Error

While the Court has often had disagreements concerning the application of the evolving standards of decency test, the test itself has never been a point of contention.³⁰⁰ From the beginning of its Eighth Amendment cases, the Court has been virtually unanimous in its determination that the Eighth Amendment evolves over time, and only moves in one direction—toward less harsh punishments.³⁰¹ A decision ignoring or overruling Eighth Amendment stare decisis as a general principle would constitute a complete disregard of the rule of law.³⁰²

If there is an error in the evolving standards of decency test, it would relate either to the objective determination of the Court concerning the national consensus for a particular punishment or to the subjective determination of the Court with respect to the purposes of punishment.³⁰³

With respect to Kennedy, finding an error with respect to the objective indicia would be almost impossible. At the time of Kennedy, only five states allowed the execution of child rapists.³⁰⁴ And currently, only two states allow the death penalty for child rape.³⁰⁵

To reverse this perception, a national revolution with more than half of the states adopting statutes similar to Florida’s statute would be a prerequisite for even raising the objective indicia question.³⁰⁶ Even then, some additional evidence of state juries sentencing individuals to death for child rape would also be necessary. And as the Court in Kennedy indicated, such prosecutions have been rare.³⁰⁷

This is exactly the point. Where almost every jurisdiction is unwilling to sanction a particular punishment for a particular crime and juries are unwilling to impose such sentences, the rare jurisdiction with the outlier jury that imposes a death sentence for child rape defies the evolving standards of decency that mark the progress of a mature society.³⁰⁸

Moving to the subjective standards, the Court likewise would be unlikely to find that the punishment of the death penalty was a proportionate punishment for child rape. First, the Court has always reached the same conclusion under the subjective standards as it has under the objective standards when it applies the evolving standards of decency test.

Second, the Court has made clear, both in Coker and Kennedy, that it finds imposing death for a non-homicide sexual crime to be disproportionate.³⁰⁹ It has consistently found that despite the brutal and scarring nature of sex crimes, such crimes do not result in physical death.³¹⁰ A punishment of death, then, would be excessive in light of the crime committed.³¹¹ While the anger that many feel toward child sex offenders likely makes that bright line unsatisfying, it is nonetheless the bright line that the Court has chosen twice.³¹²

The purposes of punishment support such a determination. If retribution concerns just deserts, and not revenge, then it requires courts to impose a sentence no more than and no less than what the offender deserves based on the culpability of the offender and the harm caused.³¹³ If the harm caused did not involve death, then it follows that the punishment should not involve death either.³¹⁴ Similarly, deterrence does not support death as a punishment for child rape.³¹⁵ The marginal deterrence between a death sentence as opposed to a life-without-parole sentence is likely insignificant, particularly in light of the two-decade time gap between sentencing and execution.³¹⁶

Finally, as discussed, the deeper problem here would be that remedying that “error” would violate the core principle of Eighth Amendment stare decisis—that the evolving standards only evolve in one direction.³¹⁷ It would involve enabling states to engage in a harsher punishment than before for a particular crime or offender.³¹⁸

2. The Quality of Its Prior Reasoning

The question of the strength of the prior reasoning with respect to Eighth Amendment stare decisis and the evolving standards approach mirrors the question of error. If there is a flaw in the overall structure of the evolving standards paradigm, it is that it relies on majoritarian indicia to inform a countermajoritarian standard.³¹⁹ In the Court’s cases, this has served as a mechanism to reduce judicial activism and the aggressive substitute of the Court’s normative views for those of state legislatures and juries.³²⁰ If anything, it has caused the Court to be entirely too hesitant in permitting states to use the draconian sentencing practices that have contributed to mass incarceration.³²¹

A likely argument against the reasoning of the evolving standards doctrine would be that the standards should evolve in both directions, allowing punishments to become harsher. The Court cannot achieve such a result without repudiating the entire doctrine. As discussed, the evolving standards doctrine serves to protect human dignity and promote proportionality.³²² Moving toward harsher punishments would undermine both.

To allow movement toward harsher punishments would invert the entire Eighth Amendment and its basic meaning. Instead of being a constitutional protection for individuals against cruel and unusual punishment, the Eighth Amendment would protect the ability of outlier states to engage in extreme punishments disallowed by most other jurisdictions. In other words, reading the Eighth Amendment to allow harsher punishments to reemerge would mean that the Eighth Amendment would authorize cruel and unusual punishments—the very thing it proscribes.

As applied to Kennedy, these objections would be even more robust. Attacking the underlying reasoning of the evolving standards would mean ignoring both the dignity of the offender and the concept of proportionality. And undoing the outcome in Kennedy would sanction the imposition of a cruel and unusual punishment.

The imposition of the death penalty for a child rapist in Florida would be cruel as it is disproportionate in two senses. First, as discussed above, it is an excessive punishment for the crime committed.³²³ Second, it is comparatively disproportionate—almost no other child rapist would receive the same punishment.³²⁴

For the same reason, it would be an unusual punishment in several ways. Not only would it be rare, as almost no other child rapists would receive a death sentence, but it would also be contrary to longstanding practice.³²⁵ Even when the Eighth Amendment allowed the death penalty for child rape, almost no states had such a law, and within those states almost no one received a death sentence.³²⁶

3. The Workability of the Current Standard

The concept of the evolving standards of decency remains very workable. It is a simple two-part test that requires the Court to assess readily available information and then make its own determination, applying criminal law theory to criminal sentences.

In reviewing Kennedy, for instance, it will not be difficult to determine how many states authorize the death penalty for child rape. It will similarly be easily ascertainable how many individuals have received death sentences for the crime of child rape.

With respect to the Court’s subjective analysis, it will similarly not have difficulty engaging in the analysis of whether a death sentence satisfies the purposes of retribution and deterrence for the crime of child rape.

4. The Effect on Other Areas of Law

The Court’s Eighth Amendment stare decisis approach will not have a significant effect on other areas of law. While the Eighth Amendment is not unique in its reliance on jurisdiction counting, it also does not bear particularly on other kinds of constitutional interpretation.³²⁷ While having some similar characteristics to the due process doctrine, the Eighth Amendment does not invoke that doctrine, and that doctrine does not invoke it.³²⁸

As such, this part of the Dobbs test would not have much of an impact on its application to Eighth Amendment stare decisis or the evolving standards doctrine. Upholding Kennedy would not create a significant change in other areas; striking it down would not either. The analysis here would pertain simply to the future of the doctrine itself and its application.

Even so, one could imagine tangential effects from overturning the evolving standards doctrine. There are certain parallels with Sixth Amendment jurisprudence in which the doctrines of the Sixth Amendment and Eighth Amendment could inform each other.³²⁹ These relate to the similar constitutional restrictions both amendments have placed on mandatory sentencing schemes.³³⁰

Another possible ripple from abandoning the evolving standards of decency doctrine could relate to juvenile offenders. The concept that juveniles are different from adults extends beyond the Eighth Amendment. In other areas of law, courts and legislatures have chosen to treat juveniles differently from adults. Changing the approach to juveniles under the Eighth Amendment could influence other areas that have adopted similar approaches.

5. The Reliance Interests in the Precedent

Finally, the question becomes whether there are significant reliance interests in the Eighth Amendment stare decisis approach and the evolving standards of decency doctrine. Criminal defendants clearly have an interest in preventing states from subjecting them to draconian punishments. While the Court’s limits on states have been few—far fewer than perhaps the national consensus reflects—rolling back those limits could exacerbate expansive uses of the death penalty by outlier jurisdictions and promote unequal punishment. It could also invite small groups of citizens to engage in human rights abuses with no judicial review.

B. The Dobbs Reasoning

Implicit in the Court’s holding in Dobbs is both a disdain for abortion and the Court’s prior holdings in Roe and Casey. For the majority, the decision clearly reflects a view that the Court “got it wrong” in its earlier cases in a fundamental way. On some level, the Court’s reasoning was beside the point.³³¹

Unlike the culture war terrain of the abortion issue, criminal justice has historically enjoyed a bipartisan consensus of sorts.³³² Liberals and conservatives, for different reasons, both rode the “tough on crime” wave of the 1980s and 1990s to unprecedented levels of mass incarceration.³³³ And since the turn of the century, both groups have worked to slowly and incrementally undo some of this trend.³³⁴ The bipartisan First Step Act provides one example of this consensus.³³⁵

Outside of Furman, the Court’s Eighth Amendment decisions have not generated widespread public response or objection.³³⁶ This is in part because the evolving standards doctrine has served to restrict outliers, not advance broad normative change.

Undoing the decision in Kennedy would encourage states to engage in draconian punishment practices to test the boundaries of the Eighth Amendment. Florida’s statute is unconstitutional on its face. Upholding it would not only undermine the rule of law, but would also encourage state legislatures to disregard the Court’s decisions and the evolving standards. This would be different than ignoring stare decisis. It would constitute a repudiation of over one hundred years of jurisprudence.

Further, a significant part of the Court’s reasoning in Dobbs dealt with its concern with the “disruption of democratic deliberation.”³³⁷ The concern related to the use of the constitution to interfere with legislative authority, particularly on issues of “profound importance to the electorate.”³³⁸ As shown by the response to the Court’s decision in Furman, abolition of the death penalty might constitute a similar kind of issue.³³⁹ But it does not appear that the evolving standards of decency generally or the execution of child rapists specifically would fall into this category.

The difference again relates to the majoritarian anchor of the evolving standards of decency. Overruling an Eighth Amendment limit would not restore the power to the people as a general matter. It would give power to a particular state to violate a national, and in some cases international, consensus against a particular punishment practice. Put differently, it would provide a license to certain jurisdictions to violate the individual rights of defendants when an overwhelming majority of jurisdictions accord defendants those very rights.

CONCLUSION

Florida’s decision to pass a new statute that clearly violates the Eighth Amendment and the Court’s decision in Kennedy does not change the analysis in the case or under the Constitution. The Court’s decision in Dobbs does not open the door to such defiance, and it does not support rejection of the Court’s precedents.

This Article has demonstrated why, even if the Court thinks the normative outcome in Kennedy is wrong, the Court still must strike down the Florida statute if given the opportunity. Specifically, this Article has made the case for a novel reading of the doctrine of stare decisis under the Eighth Amendment. Drawn from the Court’s evolving standards of decency doctrine, this Eighth Amendment stare decisis requires the Court to change the rule in cases in which the national consensus has evolved and the Court finds the sentence to be disproportionate.

The Article first developed this concept by explaining the origins of this doctrine and defending the core principle that the evolving standards only evolve from more severe to less severe punishment. The Article then explored past applications of the doctrine, distinguished deviations from the doctrine, highlighted some future applications of the doctrine, and delineated the limits of the doctrine on state legislatures. Finally, the Article concluded by demonstrating how this reading of the Eighth Amendment is consistent with the Dobbs decision, both as a doctrinal and theoretical matter.

98 S. Cal. L. Rev. 255

Download

William W. Berry III*

* Associate Dean for Research and Montague Professor of Law, University of Mississippi. The author would like to thank Corinna Barrett Lain, Meghan Ryan, Cara Drinan, Kathryn Miller, Eric Berger, Alex Klein, Katie Kronick, Rachel Lopez, and Daniel Harawa for helpful comments on an early draft during the Eighth Amendment Roundtable at Cardozo Law School in April 2023.

INTRODUCTION

I. Fair Use: Law and Theory

A. The Fair Use Doctrine

B. Theoretical Justifications of the Fair Use Doctrine

II. Fair Use as Licensing Leverage

A. The Theory

B. The Model

C. Illustrations

Oracle and the Future of Licensing in Technologies

Documentary Filmmakers

Parodies and Satires

Warhol and the Future of Transformativeness

III. Normative Implications

A. Countervailing Allocative Inefficiencies

B. Redistributing Bargaining Surplus

C. The Role of Damage Awards

D. The Latent Virtue of Ambiguity

Conclusion

98 S. Cal. L. Rev. 761

Introduction

I. The Persistent Nature of Workplace Harassment

A. The Historical Origins of Harassment

B. The Current Landscape of Workplace Harassment

C. Harassment as a Workplace Hazard

II. Legal Constraints Fail to Provide Accountability

A. The Individuation of Harassment Evidence

B. Mandatory Arbitration Provisions

C. Coverage Gaps Constrain Workers’ Access to Courts

D. Retaliation Law and Its Effect on Underreporting

E. Rigid Administrative Exhaustion Requirements Block Access to Justice

F. Severe or Pervasive Requirement Excludes Strong Claims

G. The “Objectively” Hostile or Abusive Standard is Out of Touch

H. The Standard for Vicarious Liability

I. Damage Caps Fail to Remedy Harms

J. Implied Hierarchy of Harassment Claims

III. An Empirical Analysis of State and Federal Gender Equity Legislative Activity, 2016–2022

A. Methodology

B. Volume of Gender Equity Legislation Post #MeToo

Federal Reforms

State Reforms

i. General Trends

C. Political Trends in Gender Equity Legislation Post #MeToo

Political Party

Representation of Women Lawmakers

D. Topical Analysis of Gender Equity Legislation Post #MeToo

IV. Discussion & Implications

A. Did States Fill Any Gaps?

B. Legal Gaps in Need of Greater Attention

Symbolic Action and Individuation of Harassment Continues, and Even Worsens?

Greater Focus on Intersectional Harms

Navigating an Employer-Friendly Judiciary

Dealing with Retaliation is Required for Reforms to Work

Conclusion

Appendix A. Search Terms Used to Identify #MeToo Related Legislation

Appendix B. Search Terms Used on Bill Corpus to Verify Topic Coding

98 S. Cal. L. Rev. 761

Introduction

I. The Developing Practice of Forensic DNA

A. DNA Science

B. DNA and Exonerations

C. Familial DNA

II. The Privacy Argument for Restricting Familial DNA

A. The Constitutional Dimension

B. DNA and the Fourth Amendment

C. The Case for Genetic Privacy post-Carpenter

1. The Scholarly Debate

2. Legislative Responses

D. The Fourth Amendment Argument Against Familial DNA Searches is Weak

III. The Due Process Rights of the Innocent Suspect

A. The Compulsory Process Clause

The Subpoena Power

The Right to Present a Defense and its State-Rule-Based Limitations

Constitutional Limitations

Conclusions on Familial DNA and Compulsory Process

B. The Due Process Right to Exculpatory Evidence

1. Brady and DNA

2. Conclusions on Brady and Familial DNA

C. Due Process Rights to Post-Conviction Proceedings Under State DNA and Innocence Statutes

The Lack of Constitutional Innocence Claim

State Post-Conviction Statutes