Vol. 98, No. 1 Archives – Southern California Law Review

Stringing Along the Songwriter: Successes and Shortcomings of Copyright Policy in Title I of the Music Modernization Act

March 2025March 2025Tara Steinmetz*

Congress’s enactment of the Music Modernization Act (“MMA”) in 2018 introduced the most substantial revision to copyright law in two decades and represented a landmark moment for the music industry. Over the prior decade, the rise of music streaming had taken the music business to the brink of disaster, as many of artists’ most foundational rights within music copyright law proved fundamentally incongruent with a digital future. Amongst the worst hit were songwriters and composers whose royalties had so drastically declined that many throughout the industry felt it was just a matter of time before the profession ceased to exist.

The MMA—particularly its Title I—sought to revive the industry, and five years on from its passage, music industry leaders and lawmakers on Capitol Hill reopened a conversation around the law’s successes and shortcomings given the new era of technological innovation around the corner. Through the lens of music copyright law’s three policy goals—access, incentive, and reduction of transaction costs—this Note closely examines Title I’s implementation to date. It provides a new analysis of the law’s key provisions, including the accomplishments of the Mechanical Licensing Collective and the adoption of a new willing-buyer willing-seller royalty rate setting standard in the most recent Copyright Royalty Board rate proceeding. This Note ultimately argues that Title I of the MMA successfully responded to the most immediate challenges in music copyright law, effectively eliminating untenable transaction costs that left an antiquated licensing system nearly nonfunctional. However, it has achieved relatively marginal results for one of its central aspirations and one of the industry’s most looming concerns: the hollowing out of the songwriting profession, as artists can no longer afford to create for such little return. The plight of the songwriter thus remains one of the music industry’s most fundamental yet unanswered problems in the digital age.

INTRODUCTION

In June 2023, the Recording Industry Association of America, the nation’s leading trade organization for the recording industry, announced that Avicii’s 2013 hit song “Wake Me Up” had become the highest-selling dance track in the organization’s seventy-year history.¹ Just two months later, that August, the chart-topping track joined the ranks of just over forty songs ever to have reached two billion streams on Spotify.² When it was first released over ten years ago, the song quickly rose to become Spotify’s then-most streamed song of all time and Pandora’s thirteenth most streamed song of all time.³ However, in November 2014, Aloe Blacc, the now-famous artist who wrote and sang the song’s ubiquitous chorus, publicly revealed that the song had made only $12,259 in royalties off of 168 million streams on Pandora’s digital streaming platform.⁴ From there, the sum was split amongst the song’s publishers, Aloe Blacc, and his two other co-writers, leaving Blacc with less than $4,000 made off of the United States’ largest music streaming platform at the time.⁵

While Blacc once could have received consistent royalty payments from physical album and digital download sales, an antiquated copyright system controlling songwriters’ work left him and other songwriters defenseless in the new music streaming era. Blacc lamented that the digital age presented a disappointing irony that while more people than ever enjoyed widespread access to music and benefited from songwriters’ creative labor, songwriters were being paid out by streaming services at “abhorrently low rates” and struggling to make ends meet.⁶ Despite incredible demand for their craft, he said that songwriters like him had “no power” to protect the music they created, signaling to songwriters that their work was not valued.⁷

Blacc’s story was a dime a dozen in the 2010s as music streaming services rapidly became the primary mode of music consumption. Kevin Kadish, who wrote Meghan Trainor’s 2014 breakout hit “All About the Base,” told the House Judiciary Committee in a 2015 roundtable that he had made just $5,679 off of 178 million streams of the song,⁸ despite the fact that it had spent eight consecutive weeks at number one on the Billboard Hot 100—the longest of any female artist that year⁹—and was the bestselling song of the decade by any female artist.¹⁰ Kadish was just one of over a hundred testimonies at that roundtable relaying the plight of songwriters in the digital age.¹¹

While digital streaming rapidly transformed the music industry, the legal framework for music copyright fell drastically behind. Ad hoc solutions to monopoly and antitrust concerns from eras of long-dead technologies were cemented into the legal regime, saddling musical composition copyrights with an incoherent patchwork of restrictive copyright laws. Songwriters’ and composers’ most foundational rights thus reflected early twentieth-century problem-solving that was fundamentally incongruent with a digital future.

The shortcomings of the old music copyright landscape became evident throughout the 2010s’ massive decline of songwriter remuneration. Songwriters like Blacc and Kadish penned some of the most recognizable songs of the past decade and wrote hits for some of the country’s greatest pop stars. Yet still, Kadish asked members of Congress how he was to feed his family, making just $5,600 off about “as big a song as a songwriter can have in their career.”¹² Meanwhile, the majority of songwriters, accustomed to a generally middle-class life in their profession before the streaming era,¹³ would never write a song with anywhere near this level of success; for them, the situation was even more distressing.

Concurrently, music publishing companies, which often own and manage songwriters’ copyrights in whole or in part, encountered financial challenges mirroring those of the songwriters they represented. Even digital streaming providers felt the constraints of working within an outdated legal system, which made it nearly impossible to properly license songwriters’ works as required by their business model. Then-fledgling music streaming providers faced a barrage of infringement lawsuits, jeopardizing their ability to grow despite widespread public popularity.

In 2018, Congress enacted the Orrin G. Hatch–Bob Goodlatte Music Modernization Act (“Music Modernization Act” or “MMA”), a major step towards aligning the music industry with modern digital technology.¹⁴ The MMA emerged from a comprehensive negotiation amongst numerous industry players—from publishers and songwriters to record labels and recording artists to fast-growing digital music streaming services. It represented unprecedented consensus across the music industry.

The MMA amended Title 17 of the U.S. Code, also known as the Copyright Act, which, despite two decades of major technological innovation, had not seen any major revision since 1998.¹⁵ Its three titles each aimed to tackle distinct issues in the music copyright landscape. Title I, known more specifically as the Musical Works Modernization Act (“MWMA”), addressed songwriter and publisher concerns emerging from the digital age and is the focus of this Note. Specifically, it redefined the administration of the compulsory mechanical license in response to the incompatibility of interactive streaming services with the old music copyright regime.¹⁶ The remaining two Titles of the MMA addressed copyright concerns for recording artists: Title II, the Classics Protection and Access Act, fixed a gap in prior copyright law by bringing pre-1972 sound recordings under copyright protection, and Title III, the Allocation for Music Producers Act, enables music producers, mixers, and sound engineers to receive royalties on sound recordings.¹⁷

2023 marked the fifth anniversary of the Music Modernization Act, rekindling conversations around its successes and shortcomings, particularly concerning Title I. This Note will discuss the ways in which the development of Title I of the MMA embodies a classic copyright struggle between incentive, access, and transaction costs, arguing that while it aimed to better address deep transaction cost and incentive gap concerns arising from the streaming industry—and succeeds in doing so where an urgent solution was required—there remain important unresolved tensions in music copyright law for songwriters. The songwriters upon which the music industry is built still insist that their profession is dying, and early indicators suggest that the MMA did not go far enough to fully correct failures of traditional copyright policy goals. Ultimately, Title I of the MMA recalibrated messy transaction costs that became glaringly apparent in the pre-MMA regime, but policymakers have yet to strike the right balance between incentive and access that underlies copyright policy.

I. THE MUSIC COPYRIGHT LANDSCAPE

A. Mechanics of Music Copyright

1. Copyright Basics

Each song you listen to on the radio or stream on Spotify contains two separate copyrights: a copyright to the song’s sound recording, known as a master recording or a “master,” and a copyright to the underlying musical composition embodied in the sound recording.¹⁸ Thus, the performance and distribution of a song’s recordings trigger royalties payable to both the holder of the composition copyright and the sound recording copyright.¹⁹ Sound recording copyrights are held by record labels and artists, while copyrights to musical compositions are held by music publishers, songwriters, and composers.²⁰ For example, as depicted in Figure 1, SZA’s 2022 song “Kill Bill” is copyrighted as a master sound recording and as a musical composition.²¹ Use of the sound recording would require royalties to be paid to SZA’s record label, Top Dawg RCA, and SZA as the recording artist. Conversely, the song’s underlying musical composition and lyrics also have a copyright; royalties for its use would be distributed amongst the three rights-holding music publishers and three credited songwriters. While these two types of copyrights, in theory, are granted the same exclusive rights, the licensing regimes arising out of these two copyrights have diverged over time, altering how these two categories of rights holders benefit from their exclusive rights. This Note will focus on the copyright licensing regime for music compositions owned by publishers and songwriters, as Title I of the Music Modernization Act introduced a drastic alteration in this licensing landscape.

Figure 1. Two Copyrights, One Song

Sources: Adapted from Sarah Jeong, A $1.6 Billion Spotify Lawsuit Is Based on a Law Made for Player Pianos, The Verge (Mar. 14, 2018, 9:28 AM), https://www.theverge.com/2018/3/14/17117160/spotify-mechanical-license-copyright-wixen-explainer [https://perma.cc/2QHB-DTPY]. Copyright ownership data is from The MLC Public Work Search, The MLC, https://portal.themlc.com/search#work [https://perma.cc/T6LA-4Z6X] (searched under “Work Title” for the song “Kill Bill,” added writer name criteria, and searched for Rob Bisel).

Under § 106 of the Copyright Act, copyright holders gain six primary exclusive rights as a result of their copyright including (1) the right to reproduce the copyrighted work; (2) the right to prepare derivative works based upon the work; (3) the right to distribute copies of the work; (4) the right to perform the work publicly; and (5) the right to display the work publicly, a right tending to be less prevalent in music copyright beyond displaying lyrics publicly or online.²² Copyright holders thus can license these defined rights to any willing buyer.

The licensing of a musical composition can be broken down into two primary buckets. The right to perform a composition publicly, or a “public performance right,” generates what are known as performance royalties any time someone plays a composition copyright holder’s song publicly.²³ This right, therefore, is most likely to come into play when a song is played on the radio or performed live (and now, when streamed on demand).²⁴ These royalties, which will be of limited discussion in this Note, are administered through collective licensing organizations called performing rights organizations, or “PROs.” PROs are private, non-governmental entities that work on behalf of songwriters and publishers to license performance rights through “blanket licensing”—a licensing model that allows a licensee to use any works in a PRO’s catalog pursuant to the terms of a “blanket” agreement.²⁵ The second license, called a mechanical license, encompasses the right to “reproduce musical works in formats used for mechanical playback.”²⁶ Money paid to use these licenses is known as a mechanical royalty.²⁷ Typically, songwriters and composers “sell or license their composition copyrights to publishing companies, which will administer the copyright in return for 25–50% of the proceeds.”²⁸ In such arrangements, licensees pay royalties to the publishers, who take their cut and distribute the remaining funds to the songwriters they represent.²⁹ As seen in the Figure 1 example, SZA’s “Kill Bill” is co-owned by three publishing companies, each of which receives royalties based on an agreed-upon percentage share of ownership in the song. Subsequently, these publishers distribute royalties to the songwriters they represent after taking their cut from the royalty pool.³⁰

Figure 2. Basic Framework of Copyright Royalties Within a Song

Source: Adapted from Ben Lowe, What Are Music Royalties? The Difference Between Recording and Composition Royalties, Songtrust (Feb. 16, 2023), https://blog.songtrust.com/what-are-music-royalties [https://perma.cc/3E6F-UMA8].

2. Understanding the Compulsory Mechanical License

To understand the mechanical license’s original purpose, it is helpful to briefly look at its origin. In 1908, the Supreme Court decision in White-Smith Music Publishing Co. v. Apollo Co. held that piano rolls—scrolls of perforated paper used in player piano devices to mechanically operate piano keys—did not infringe upon a composition’s copyright because they did not constitute “copies within the meaning of the copyright act.”³¹ Thus, under White-Smith, only traditional copies of sheet music were deemed entitled to protection, and songwriters and composers (or the publishers that represented them) were powerless when others reproduced their music mechanically. While the Court hesitated to expand its understanding of composition copyright protection, Congress responded by establishing the mechanical license in the 1909 Copyright Act, mitigating songwriters’ and composers’ drastic loss of control over their musical compositions.³² Today, “mechanical” music production is largely gone, making the mechanical license’s name and history a cause for confusion when contemplating how such a license might map onto modern technology. Regardless, the mechanical license still refers to the distribution of a piece of music—including on vinyl, CDs, digital downloads, and distribution via streaming services.

While the establishment of a mechanical right brought great relief to music composition copyright holders at the time, it came with one catch—a compulsory licensing requirement.³³ This compulsory license, created under § 115 of the Copyright Act, was the first of its kind in copyright law.³⁴ It converts the copyright holder’s right to license into an obligation to license, allowing interested licensees to use a copyright holder’s composition without permission, provided that they pay a set mechanical royalty fee and comply with a specified list of statutory requirements.³⁵ Thus, if you wish to cover SZA’s song “Kill Bill,” you do not need to ask the “Kill Bill” songwriters or their publishers for permission, though you are obligated to pay a royalty fee and adhere to statutory requirements. Similarly, when interactive music streamers like Spotify³⁶ seek a license to stream “Kill Bill” on their platforms, these companies are not required to ask the songwriters or publishers for permission—they, too, are responsible only for the statutory fee and requirements.³⁷

Beyond imposing a compulsory license regime, the Copyright Act set royalty payments at a frozen statutory rate, which was well below market value at a mere two cents per use.³⁸ This fixed rate’s inability to adjust, even in the face of inflation over the next seventy years, led to its significant criticism as an injustice to composition copyright holders.³⁹ Eventually, the 1976 Copyright Act created a statutory rate-setting entity, the Copyright Royalty Tribunal (“CRT”), to adjust rates at five-year intervals.⁴⁰ The tribunal was the predecessor to the current rate-setting entity, the Copyright Royalty Board (“CRB”), established by Congress in 2004.⁴¹

Before the MMA’s enactment, the CRB (and previously, the CRT) was guided by four copyright policy goals defined in 17 U.S.C. § 801(b) to determine adjusted compulsory license royalty rates.⁴² These “801(b) [policy] factors,” as they came to be known, were intended to “identify the royalty amount that would reward copyright owners and disseminators commensurate to their ‘relative roles’ in providing the public with access to creative works,”⁴³ reflecting the standard utilitarian theory of copyright discussed later in this Note.⁴⁴

Despite the establishment of policy factors claiming to prioritize “fair return” to creators in consideration of “fair income under existing economic conditions,”⁴⁵ the policy-driven statutory rates nonetheless continued to depress mechanical licensing rates by “plac[ing] artificial limits on the free marketplace.”⁴⁶ Notably, one of the MMA’s main goals was to address grievances with the rate-setting process.

B. Complicating the Picture: The Rise of Streaming

For decades, lawmakers dealt with new technologies and the copyright questions they posed by bolstering a “patchwork of laws, industry conventions and private deals.”⁴⁷ Over time, the music copyright landscape has become increasingly fragmented and complex, contorting to fit each new iteration of technology.⁴⁸ This confusing legal scheme ultimately came to a head as digital streaming providers, or “DSPs,” began to dominate the music industry.⁴⁹

Imposing mechanical royalty obligations upon discrete products such as CDs or digital downloads ultimately proved relatively straightforward, despite initial backroom quarrels about which exclusive rights were implicated by these technologies.⁵⁰ For each record manufactured and distributed or each digital copy downloaded, publishers are paid a set mechanical royalty rate—currently set at twelve cents per copy.⁵¹ Because these are privately owned copies of music, each play is not a public performance, and thus, no performance royalty payment is required.

Ironing out royalty calculations for DSPs was much more complicated. When a listener streams a song aloud, is that a public performance, much like terrestrial radio? Or is it a mechanical reproduction and distribution of a song, like a digital download? Here, the traditional line between a performance and a mechanical distribution is not so clear.⁵² To help with this dilemma, DSPs are categorized into two modes of digital music delivery—noninteractive and interactive.⁵³ Noninteractive streaming services, such as Pandora’s traditional genre-based, radio-like streaming platform, “substantially limit user control” of the music they consume, as the platform makes most of the user’s listening choices.⁵⁴ Since this type of streaming was similar to radio, it was eventually determined that songwriters would only get performance royalties for noninteractive streams, mirroring what a songwriter would receive if the radio played their song on air.⁵⁵ In contrast, interactive streaming platforms, such as Spotify, Apple Music, and Amazon Music, allow listeners to stream songs on-demand (“interactively”) by selecting and immediately transmitting the music they wish to listen to.⁵⁶ These platforms fit neither a radio model nor a CD or download model, and existing law failed to provide guidance for resolving this tension. Thus, as interactive DSPs burst into the industry in the late 2000s, legal disputes broke out between publishers and DSPs to decide whether, upon deep analysis of streaming technology, such streams implicated either performance or mechanical rights.⁵⁷

Figure 3. U.S. Recorded Music Revenues by Format

Source: U.S. Music Revenue Database, Recording Indus. Ass’n of Am., https://www.riaa.com/U-S-SALES-DATABASE [https://perma.cc/2X4U-6EG4] (toggled date range from 1998 onwards and opted to adjust for inflation). It should also be noted that Figure 3 displays the source of all revenue in the industry—not just songwriter royalties—to indicate general trends in music consumption. Such revenue streams would be split among composition and sound recording copyright holders, depending on applicable copyrights and contractual agreements.

Publishing companies and songwriters vigorously fought for interactive streaming to encompass both performance and mechanical royalties.⁵⁸ Both royalties were two critical “pillars” of income that the industry had relied on for over a century,⁵⁹ and the industry worried—with good foresight⁶⁰—that music streaming would eat away at these traditional revenue sources from radio, physical sales, and digital downloads, which it so desperately depended upon for adequate income.⁶¹ DSPs inevitably pushed back, arguing that the publishers’ demands that they pay both performance and mechanical royalties for a digital transmission of a single song amounted to “double-dipping.”⁶²

Ultimately, the DSPs’ argument failed, and interactive streaming platforms were deemed to need both a public performance license—reflective of the instantaneous nature of streaming—and a mechanical license—reflective of DSPs’ distribution through streaming—to utilize a musical composition.⁶³ Thus, interactive streaming services, many of which began as small start-ups, had to invest an immense amount of time and money to properly license content from both publishing and sound recording stakeholders. This extraordinarily cumbersome licensing system posed a challenge to the growth of DSPs and, as will be discussed further in Section I.D, created a tension that significantly instigated the enactment of the MMA.⁶⁴

Unlike the simple mechanical rate for CDs and digital downloads, mechanical royalty rates for streams are determined by a complicated rate calculation formula.⁶⁵ The rates are calculated using the greater of (a) a designated percentage of a DSP’s total revenue, accounting for total earnings through both paid subscription and advertising revenue or (b) a designated percentage of the Total Content Cost (“TCC”).⁶⁶ TCC represents the total amount that a DSP spends on sound recording and music composition licensing for interactive streaming.⁶⁷ For instance, the most recent CRB rate proceeding for 2023 to 2027, Phonorecords IV, held that the DSPs’ mechanical royalty rate is the greater of 15.35% of their total revenue or 26.2% of TCC.⁶⁸ The rate measurement that is chosen is known as the “headline rate.” The amount DSPs pay for performance licenses is subtracted from the total money pool determined by the headline rate.⁶⁹ The remaining amount is then compared to a “mechanical floor” rate, calculated by multiplying each paid DSP subscriber by a “per subscriber minimum” mechanical royalty fee.⁷⁰ The final mechanical rate is determined by the greater of the original bucket of money or the mechanical floor calculation. This summary merely offers a high-level overview of a complex calculation, highlighting the music copyright system’s intricate and often inaccessible nature when applied to digital streaming.

Figure 4. Summary of a Musical Composition Copyright

Source: Adapted from Jeong, supra Figure 1.

C. The Underlying Theory: Access, Incentives, and Transaction Costs

The patchwork of laws and licenses that make up the music industry, in theory, serve to honor very fundamental policy goals that lie at the heart of all copyright law. For music, in particular, policy is triangulated around three concerns: incentive, access, and transaction costs.

1. Balancing Incentive and Access

In the American free market, intellectual property presents a rare occasion in which the state grants a monopoly—albeit a limited one—to incentivize innovation. The Constitution outlines that from the outset of the country’s establishment, the United States sought to protect creators via the “promot[ion of] the Progress of Science and useful Arts . . . securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.”⁷¹

Simply put by music attorney Donald Passman, it is entirely understandable that “if you created something and everybody had the right to use it without paying you, not very many people would go through the trouble of creating anything.”⁷² However, while a monopoly over one’s creation drives an incentive to produce and innovate, it equally presents the risk that rights holders may place exorbitantly high prices on their works, blocking the public from accessing the “science and useful arts” the Constitution aimed to protect in the first place. Thus, under the predominant utilitarian theory of copyright, the story of intellectual property law is an “ ‘access versus incentives’ tradeoff”—a balancing game between maximizing incentive for creators while tempering the scope of their exclusive rights so that consumers can still access and benefit from their work.⁷³ This tension repeatedly comes to the forefront during the proliferation of new dissemination technologies—from the advent of the piano roll to digital downloads to modern streaming services. Each technological inflection point has illustrated that the balance between access and incentive—and between control held by users and creators—is in constant flux.⁷⁴

The introduction of the compulsory mechanical license marked the beginning of numerous legislative decisions in the music regulatory regime that aimed to “recalibrate the balance between creators’ financial incentives and public access to expressive works” when rights holders’ exclusive control over their copyright and ability to pursue “free market licensing” would cause a failure of the copyright equilibrium.⁷⁵ When Congress crafted the 1909 Copyright Act in response to the White-Smith decision,⁷⁶ it hesitated to restore music composition copyright holders’ full exclusive rights.⁷⁷ It feared that doing so within the new mechanical licensing regime would allow copyright holders to “monopolize the business of [music] manufacturing” and prevent the growth of the piano player, a new music distribution technology with vast potential to increase the public’s access to music.⁷⁸ Thus, the compulsory license provided a new legal solution to the incentive-access tension; in exchange for continued compensation, composition copyright holders—songwriters, composers, and music publishers—have been forced to sacrifice the control over their copyright and their right to freely negotiate prices on an open market.

This policy objectively appears to honor a balance between incentive and access; however, the strict pre-MMA rate-setting mechanisms rarely resulted in adjustments to royalty rates over time,⁷⁹ raising questions as to whether the “incentive” provided by these statutory rates was able to respond to songwriters’ and composers’ shifting real-world incentives (or lack thereof). Notably, the prevailing utilitarian theory of copyright policy does not center artists and their incentives to create but rather views supporting artists with an incentive as a means to an end, with the goal of stimulating creation and idea production.⁸⁰ However, if copyright policy ignores artists and fails to truly consider whether it grants them adequate incentives, the access-incentive equilibrium will ultimately fail.

To properly gauge the effectiveness of the access-incentive equilibrium for artists, artists’ incentives must be examined within the industry structure in which they operate. In constructing the copyright system, Congress contemplated that creators would contract with corporate intermediaries such as record labels and publishers, transferring their copyrights in exchange for commercial distribution of their work.⁸¹ Thus, copyright incentive theory generally accounts for providing incentives to both economic actors—creators and intermediaries.⁸² Incentive theory “assumes a chain of value” in which money trickles down from listeners and music distributors to intermediaries and finally to the creators.⁸³ While creators sit furthest in the chain from any source of financial remuneration, they are the source of the creative work that copyright policy hopes to incentivize. Ultimately, one will not know if this “quid pro quo” between incentive and access is working unless one “know[s] how much the creators are getting from the bargain.”⁸⁴ Therefore, even though artist incentive is not the sole purpose of copyright policy, this Note frequently homes in on the growing disincentive to create in the songwriting field, both pre- and post-MMA, which jeopardizes the songwriting profession and the basic copyright equilibrium.

2. Reducing Transaction Costs

For nearly all copyright interests in the American “free market licensing” system, “creators of . . . copyrighted works . . . are free to choose their licensees and negotiate royalties.”⁸⁵ This free-market commitment falls away in a limited number of exceptions; only five categories of compulsory licenses exist—three of which are implemented within the music copyright regime, with the most obtrusive and well-known being the compulsory mechanical license.⁸⁶ What makes music different? Given that the incentive-access trade-off underlies all areas of copyright law, it cannot be the sole explanation for singling out songwriters to be subjected to a compulsory license.

Theorists widely believe that the strongest explanation for the compulsory license is that it not only balances incentive and access but also serves to “remedy market failures related to transaction costs by allowing licensees to bypass costly or unfeasible negotiations.”⁸⁷ When the cost of a licensing regime to the parties is “substantial and logistically difficult,” transaction costs associated with licensing become prohibitively high and deter dissemination of copyrighted work.⁸⁸ This concern of prohibitively high transaction costs is specific to music due to how music is inherently accessed and consumed by the public; unlike other forms of copyrightable works, it is comprised of numerous little pieces of content, all of which must be individually negotiated and licensed in order to be distributed as a larger catalog.⁸⁹ The compulsory license thus aims to eliminate, or at least reduce, the cost of licensing negotiations. Although transaction costs have become the predominant explanation for the persistence of the mechanical license today, some caution that this justification has also been overstated.⁹⁰ For example, on the public performance side of music composition copyright, PROs issue blanket licenses⁹¹ that similarly function to reduce transaction costs that arise from licensing each discrete piece of music.

Access, incentive, and transaction costs cannot, on their own, explain the logic of copyright policy. However, together, this triad of policy considerations is an essential framework to consider when assessing the legal landscape before the MMA’s enactment to explain its aims and determine the overall efficacy of the legislation post-MMA.

D. Framing the Issue: Where Are We Before the MMA?

The challenge underlying copyright law is finding an equilibrium at which the three competing policy goals can coexist. The Music Modernization Act emerged from a rising tension between what publishers and songwriters felt was a desperate imbalance between the incentives provided to them by the music copyright system and the public’s unprecedented and increased access to their work via modern music streaming services. If more people were listening to music than ever before thanks to the mass proliferation of DSPs, why weren’t creators and their intermediaries sharing in that success?⁹² This tension was most acutely felt by songwriters, who sit at the bottom of the copyright “chain of value” and whose livelihoods were increasingly at risk due to the diminishing return for their creative works.⁹³ Publishers, as corporate entities, suffered less financially overall than individual songwriters but were still concerned about their decline in revenue at the hands of the music streaming landscape. Publishers, represented by the National Music Publishers Association (“NMPA”), became a major lobbying force for change and ultimately were key negotiators in drafting the MMA.

Recording artists and record labels had similar grievances with DSPs, although to a lesser degree, due to less government interference and noticeably higher royalties granted to sound recording copyright holders. Further, recording artists could earn revenue through concert ticket sales to help buffer some of the DSP-era royalty losses.⁹⁴ Concerts were potentially a substantial source of income for successful artists—and a source of income not available to songwriters. Nonetheless, mounting concerns around the evolving music industry also drew in the Recording Industry Association of America (“RIAA”) as another key player in crafting the MMA.

1. Compulsory Mechanical License Causes Drastic Disincentive

Musical composition copyright holders were particularly at a detriment prior to the MMA, given the limitations imposed on their copyrights by extensive governmental regulation, including compulsory mechanical licenses and restrictive consent decrees imposed on PROs.⁹⁵ Further, DSPs and composition copyright holders alike found that the compulsory mechanical license did not sufficiently reduce transaction costs, as the mechanical licensing regime prior to the MMA proved far too cumbersome to allow for streamlined payments to rights holders. In 2015, in response to mounting dissatisfaction with the convoluted music copyright regime, the U.S. Copyright Office released a comprehensive 245-page report (hereinafter referred to as the Copyright Office Music Marketplace Report), studying the landscape and proposing changes that would “promote greater fairness, efficiency, consistency, and transparency.”⁹⁶

The Copyright Office Music Marketplace Report admitted that “[v]iewed in the abstract, it is almost hard to believe that the U.S. government sets prices for music. In today’s world, there is virtually no equivalent for this type of federal intervention.”⁹⁷ For composition copyright holders, the singling out of their works to be placed under a restrictive, involuntary licensing regime has caused deep frustration, as the mechanical compulsory license left them subject to rates far below free-market compensation. To be clear, songwriters and publishers are free to negotiate with mechanical licensees—and they often do. However, as mentioned in Section I.A, the pre-MMA statutory rates set by the Copyright Royalty Board (and its prior equivalent, the Copyright Royalty Tribunal) prevented free market rates from being achieved; even though licensees rarely paid the bare minimum statutory rate, the low statutory rate tethered negotiations to a lower price window.⁹⁸ CRB has openly acknowledged that their rates “have considerable impact on the private agreements that enable copyright users to clear the rights for reproduction and distribution of musical works.”⁹⁹

As such, publishers and songwriters were left with no real leverage when attempting to negotiate against licensees who knew that if publishers or songwriters defected from a negotiation, the licensee could just go get the license using the compulsory rate. To illustrate the extent to which the compulsory license depresses profits earned by songwriters, one can look at the overall split of Spotify revenue in 2016: for every dollar of revenue earned on Spotify, 58.5 cents went to the owner of a song’s sound recording (a record label or the artist), 29.38 cents stayed with Spotify as profit, 6.12 cents went to the composition rights holder to cover performance royalties, and 6 cents went to the composition rights holder to cover mechanical royalties.¹⁰⁰ Apple Music’s revenue distribution did not look all that different, though it paid about 6.75 cents on mechanicals and 6.75 cents on performance rights per dollar of revenue.¹⁰¹ While this is framed in terms of revenue split per dollar, each stream produces just a fraction of a cent for a songwriter, requiring approximately 1,000 streams for composition copyright holders to earn a single dollar.¹⁰² Each royalty paid for the use of a musical composition is even further divided amongst the publisher and the songwriters who worked on the composition.

Figure 5. What Happened to a Dollar of Revenue on Spotify and Apple Music (in 2016)

Source: Adapted from Manatt, Phelps & Phillips, supra note 100.

Overall, streaming represented a shift toward listening behaviors that do not “generate significant income for artists.”¹⁰³ Streaming ushered in an overall increase in performance royalties earned; however, this slight increase in revenue—at royalty rates already persistently lower than publishers or songwriters would like—was unable to make up for the significant “downward spiral of record sales and therefore artist and mechanical royalties.”¹⁰⁴ Many songwriters at the time reported “ ‘a reduction of 60 to 70% or more’ in mechanical royalties” since the introduction of streaming services onto the music market, noting that their amount of mechanical royalties only continued to decrease rapidly.¹⁰⁵

The below-market statutory rate for mechanical royalties and the shift in revenue sources gradually diminished the overall incentive for songwriters to write music in the pre-MMA market.¹⁰⁶ By the time the MMA was enacted in 2018, it had been years since songwriters felt they could make a living off songwriting as they had before the era of streaming services. In 2014, Nashville Songwriters Association International reported that Nashville, a well-known national hub for songwriters, saw an 80% decrease between 2001 and 2014 in individuals claiming songwriting as a full-time occupation, lamenting that many songwriters have been “forced out of the profession” in the wake of digital streaming.¹⁰⁷ In 2016, acclaimed music attorney Dina LaPolt told one songwriter client, who was shocked by her measly royalty check after achieving a hit song, that “unless streaming rates [] changed and the music-licensing system [was] overhauled for the digital age, the profession of songwriting was on its way to extinction.”¹⁰⁸

By applying the incentive theory of copyright to examine the music industry’s structure, one can see the creator and intermediary dynamic at play, such that even though publishers decried their diminishing incentives to administer the copyrights they held, even less remuneration trickled down to the creative mind whose work copyright policy seeks to encourage.¹⁰⁹ While publishers’ services are an essential part of the music industry, the incentive problem was exacerbated by the degree to which the publisher’s share of royalties also diminishes a songwriter’s return for their efforts, as they stand last in line for compensation.

It has often been asserted that when the compulsory license forces rights holders to give up control over otherwise exclusive intellectual property rights, the expected trade-off is, at the very least, an assurance of payment to copyright owners—this is what makes the license work within the balance of copyright policy goals.¹¹⁰ In reality, this guarantee of payment meant little in the way of incentives once DSPs became the primary mode of music consumption, and the royalty system no longer matched the predominant music technology of the day. While testifying in front of the Senate Judiciary Committee in favor of passing the MMA, Grammy-award-winning songwriter Josh Kear said of his community of songwriters in Nashville, “we have lost our entire middle class of songwriters—they’re gone.”¹¹¹ Grammy-nominated artist Will Sheff affirmed this sentiment, telling the Washington Post that “ ‘[a]round 2011, the bottom fell out,’ . . . [marking] the beginning of the end of a ‘musical middle class.’ ”¹¹²

While copyright policy does not center incentives to artists, the Supreme Court has acknowledged that, at a minimum, “[t]he rights conferred by copyright are designed to assure contributors to the store of knowledge a fair return for their labors.”¹¹³ The bounds of such a “fair return” may be debated; however, when incentive dissipated and songwriters left the profession en masse, something was surely off in the copyright access-incentive equilibrium. Ultimately, in the context of a new digital age, the pre-MMA legal regime failed to maintain an adequate incentivization system as had been a pillar of copyright law.¹¹⁴

2. Compulsory Mechanical License Overflows with Transaction Costs

In addition to songwriter grievances, the pre-MMA system also strained the growth and capabilities of DSPs. Streaming companies were immediately overwhelmed by the licensing demands of their business model as they attempted to manage a catalog of tens of millions of songs and an untenable amount of data.¹¹⁵ Further, many of the song files that record labels shared with DSPs for upload onto their services had missing songwriter identification in their metadata, meaning that DSPs often had no idea to whom they owed money on the publishing and songwriter side.¹¹⁶ Due to a simultaneous inundation of information and a lack of information, unpaid mechanical royalties began to pile up.

In the pre-MMA legal regime, section 115 of the Copyright Act allowed DSPs to file a “Notice of Intent” (“NOI”) with the Copyright Office for each song it intended to use.¹¹⁷ By declaring their intent to use a copyrighted musical composition via an NOI, DSPs were granted legal immunity from copyright infringement “until and if the songwriter or their publisher could be identified.”¹¹⁸ DSPs filed NOIs by the tens of thousands, but without the infrastructure to locate rights holders, NOIs essentially enabled DSPs to use copyrighted material without being held accountable for finding and paying rights holders. At the same time, the Copyright Office did not review notices for “legal sufficiency” or errors¹¹⁹ and was overwhelmed by the massive volume of notices submitted by DSPs.¹²⁰

Further, rights holders bore the burden of claiming their royalties yet had no way of knowing when they went missing nor an understanding of how to claim them in a convoluted and complex system.¹²¹ This burden created yet another layer of transaction costs on the rights holders’ side; not only would it take immense time and effort on the part of the rights holder to navigate an impossibly complex web of mechanical licenses, but it also required cultivating a copyright licensing expertise that many creatives in the music industry simply did not have. Very frequently, this hurt independent songwriters the most, as they had little business or legal savviness to understand how and why they needed to assert their copyright ownership.

The mechanical licensing system quickly proved inefficient, burdensome, and unaccountable to rights holders in the context of the streaming business model. As NMPA President and CEO David Israelite pointed out in the MMA’s Senate Judiciary Committee hearing, the music industry was relying on “a [licensing] system that was built for licensing individual albums—ten at a time, twelve at a time—and using it to try to empower digital companies to put up 40 million songs overnight”; he asserted bluntly, “It doesn’t work.”¹²² By 2015, an estimated $50 to $75 million dollars sat unmatched and unpaid by DSPs.¹²³ The music industry thus dubbed these unmatched royalties “black box royalties.” This problem came to a head as publishing companies began launching lawsuits against the DSPs for, in essence, operating their businesses under a model that relied on copyright infringement, even if unintentionally.¹²⁴ In interviews, Israelite estimated that for roughly twenty-five percent of all streams, composition copyright owners were not properly identified, and “as much as 25 percent of royalty payments [were] not being paid to publishers, or [were] being distributed to the wrong entities.”¹²⁵ DSPs, daunted by lawsuits and claiming they wanted to “pay every penny,” began calling on the industry to work with them to develop a transparent and accountable way to manage publishing rights.¹²⁶

Here, the problem pre-MMA was twofold. First, the compulsory mechanical license was no longer serving its purpose of reducing the burden of transaction costs because while licensing negotiations were feasible due to the compulsory license, the DSPs’ task of complying with licensing obligations was still near logistically impossible—DSPs simply could not figure out who to pay. Rather than prohibiting the DSPs from operating, the pre-MMA licensing regime compelled DSPs to function on an “infringement model,” which was damning to the legitimacy of music’s copyright laws. While the compulsory mechanical license may have once been an effective mechanism to constrain transaction costs, when imposed upon streaming technology, the transaction costs of mechanical licensing had shifted and were ultimately too high. Secondly, this issue further harmed the already-diminished incentive for songwriters to create because even as they produced work, it took ages to see that money return to their pockets—if it ever did.¹²⁷

Amongst the triad of copyright policy goals, it at least seemed that access to music was at an all-time high. Subscribers to DSPs only continued to increase, and Americans were listening to more music than ever. However, increased access cannot be considered without its tether to copyright’s countervailing need to incentivize creation. Rapidly decreasing songwriter incentive to create, coupled with the public’s disproportionate new-found access to music, likely signaled that the pre-MMA access-incentive trade-off was sorely out of balance.

Additionally, as publishers and songwriters fought to increase royalty rates, the DSPs had their own impending “access” problem looming over them. Since its invention in the mid-2000s, music streaming had never been profitable; even the major DSPs consistently operated at a loss since their founding.¹²⁸ Though the DSPs’ business models ushered in a new level of public access to music, it remained a concern that these tides could turn. For how long could DSPs continue providing such a high level of access if their business model did not turn a profit? This long-term concern made DSPs fearful that royalty rate increases would stifle company revenue and further challenge the DSPs’ business model.

Table 1. Pre-MMA Policy Concerns
Policy Goal	Score
Incentives	Incentive for songwriters is LOW, as attributable to a minimal statutory royalty rate, governmental restrictions on both types of songwriters’ royalties, and industry structure providing equal payout between publishers and songwriters.
Transaction Costs	Transaction costs are HIGH. While the compulsory license reduces transaction costs of licensing negotiation, matching royalties to songwriters imposes high transaction costs for DSPs and songwriters alike.
Access	Access is HIGH in the short term with concern that this variable may shift if DSPs continue operating at a deficit.

II. SHIFTS UNDER TITLE I OF THE MUSIC MODERNIZATION ACT

The MMA passed the House and Senate unanimously in 2018 and received unprecedented support from a multitude of players in the music industry. It signified the culmination of years of negotiation and bargaining between music publishers, songwriters, and DSPs, which legislators facilitated in a moment when the parties’ interests were at last relatively aligned.¹²⁹ DSPs, faced with a wave of crushing infringement lawsuits, had a major incentive to support the passage of an act that would accommodate their business model and protect them from future liability.¹³⁰ Songwriters and publishers, on the other hand, were desperate for a legal solution that would restore the value of their mechanical rights to something comparable to those before the streaming age or that, at the very least, resulted in a fairer system than the one they had been operating in for nearly a decade.¹³¹ Signed into law on October 11, 2018, the Music Modernization Act was the most substantial revision to copyright law in two decades.¹³² Title I of the MMA, known as the Musical Works Modernization Act, not only endeavored to adapt the Copyright Act’s musical composition provisions to better accommodate streaming technology, but it also tweaked aspects of the musical composition copyright in ways that bettered the position of publishers and songwriters overall to counter issues raised by the streaming model.

At a high level, Title I of the MMA implemented three primary changes to improve how songwriters received their royalties and the standard for calculating them. First, it allowed DSPs to obtain a blanket mechanical compulsory license specifically for streaming mechanicals that would cover all musical compositions held in a licensor’s catalog, much like the blanket license used for granting public performance rights.¹³³ In conjunction, Title I also established the Mechanical Licensing Collective (“MLC”), a non-governmental, non-profit entity authorized to “offer and administer blanket licenses” to DSPs, “[c]ollect and distribute royalties from digital music providers,” and maintain a “musical works database” to track information about licensing and ownership of mechanical rights.¹³⁴ These first two major changes enacted in Title I were designed to resolve “long-festering pain points” between publishers and DSPs due to the DSPs’ inability to pay out licensing obligations under the pre-MMA legal regime.¹³⁵ As discussed further in the next subsection, these two additions to music copyright law aimed to address head-on the mechanical license’s growing transaction cost problem that emerged from DSPs’ massive streaming catalogs. Lastly, Title I imposed alterations to the royalty tribunal procedures, “direct[ing] the Copyright Royalty Board (CRB) . . . to employ a ‘willing buyer, willing seller’ standard to reduce the potential for compulsory licenses to undervalue the mechanical right.”¹³⁶ Ultimately, Title I of the MMA represents an acknowledgment that the fundamental balance of interests demanded by copyright policy was no longer in check.

A. Addressing the Black Box Royalty Problem

1. The MMA’s Solution: Creating the MLC

Three years prior to the passage of the MMA, the Copyright Office Music Marketplace Report had concluded that a centralized, comprehensive database of mechanical copyright ownership “would substantially enhance transparency in the music licensing system, reduce transaction costs, and facilitate direct licensing through private negotiation in the open market.”¹³⁷ Other music licensing regimes, such as public performance licensing, successfully used the combination of a collective licensing organization and a blanket license to streamline similar transaction cost issues, providing a potential model for success.¹³⁸ Thus, throughout discussions about the MMA, the concept of a responsible collective licensing body was highly supported across the industry. The entity that was created, the MLC, serves the primary role of administering the new blanket licenses and royalties and maintaining a comprehensive database of mechanical copyright holders.

Under the MMA, DSPs also agreed to pay all costs associated with this new licensing entity.¹³⁹ Unlike any other existing licensing collective, the MLC does not take a commission from the royalty pool before distributing royalties, ensuring that composition rights holders retain one hundred percent of all royalties passed through the MLC and that their royalty money is not further diminished through the new collective licensing agency’s establishment.¹⁴⁰ In exchange, when DSPs obtain a blanket mechanical license from the MLC, they are broadly shielded from extremely costly infringement liability and burdensome statutory damages.¹⁴¹

At the outset of drafting the MMA, the growing multi-million-dollar pool of unpaid black box royalties presented one of the most pressing challenges to restoring order in the mechanical copyright system. Exorbitant transaction costs from individually licensing mass amounts of content—not to mention the ex-post transaction costs of litigating licensing failures—directly cut against the compulsory license’s original purpose and placed both copyright holders and DSP licensees in untenable positions. Thus, a blanket compulsory license under the MWMA, covering all songs under a single lump-sum royalty fee, represented a major solution to one of the most glaring problems in the pre-MMA digital copyright licensing space. It was to be administered by the newly formed MLC entity, which became officially designated to carry out the MMA’s statutory responsibilities by the Copyright Office in July 2019.¹⁴²

In January 2021, the MLC launched its full operation and began administering blanket licenses.¹⁴³ Only qualified DSPs are eligible to receive this blanket license,¹⁴⁴ and at the time the MLC became operational, it reported that over forty eligible DSP services had started using its blanket license.¹⁴⁵ DSPs now no longer have to individually license songs by the tens of thousands and, as a tradeoff for this newfound legal security, the DSP licensees must pay out the full cost of mechanical licensing fees to the newly formed MLC entity. On a monthly basis, each DSP sends the MLC data that logs all recordings used on the streaming service that month, as well as the corresponding mechanical royalties owed, which the MLC maintains in its “MLC Portal.”¹⁴⁶ It is then responsible for locating and paying out known rights holders in monthly payments,¹⁴⁷ the first one of which was completed in April 2021.¹⁴⁸ The MLC’s model ensures that DSPs regularly process and pay royalties for all licensed works—even those with missing or incomplete songwriter information—rather than allowing them to hold onto royalty money under the pretext that rights holders information was nonexistent or too hard for DSPs to track down.

After the MLC distributes royalties to known rights holders, it places all unmatched and unclaimed royalties in an interest-earning account for three years, where they await a potential match with their copyright holders.¹⁴⁹ The MLC distinguishes between royalties that may be “unmatched,” such as the black box royalties that have no known association with any musical work in the MLC database, and “unclaimed” royalties, which are matched to a song but cannot be paid out because shares of the song have not yet been claimed by their rights holders.¹⁵⁰ Both types of royalties make up the sum of money pooled into this account. If, after three years, the royalties remain unmatched or unclaimed, they are liquidated to known copyright holders—both publishers and songwriters.¹⁵¹ Ultimately, while some unclaimed royalties may never make it into the hands of their rightful owners, the MLC’s pooling measure strengthens incentives for artists in the copyright system by ensuring that the money is, at the very least, redistributed amongst the rights-holding community. Prior, DSPs financially benefited from songwriters’ failure to locate their unmatched royalties, as the unpaid money sat in their pockets and created no incentive for DSPs to seek a solution to orphaned royalties.¹⁵² This further discouraged music composition rights holders, especially songwriters, who felt DSP’s unpaid use of their work signaled just how little both DSPs and the industry at large valued their contributions to music.

While the DSPs benefited from legal clarity as a result of MLC’s creation, the legislation largely arose from a recognition that songwriters needed to be “properly protected” and that a new mechanical licensing system required creating “sufficient incentives . . . to find the people the money [from royalties] belong[ed] to.”¹⁵³ With songwriters’ concerns placed at the center of the legislation, an independent collective like the MLC primarily protects and advocates for the mechanical rights of composition copyright holders. The MLC thus aims to ensure that, eventually, as few royalties as possible will languish in the ownerless “black box” pool by creating an accessible mechanism for songwriters and publishers to claim ownership of a work.

Though DSPs still escaped the primary burden of locating and paying out rights holders, the MMA took part of this burden off the shoulders of rights holders as the MMA established the MLC to increase transparency in the mechanical licensing process and conduct outreach to artists. Title I of the MMA imbued the MLC with the responsibility of “engag[ing] in efforts to identify musical works . . . and locate the copyright owners of such musical works,” creating an obligation on the part of the MLC to seek out rights holders rather than allow black box royalties to accrue without accountability.¹⁵⁴ A central part of executing this mandate included creating a “musical works database” within the MLC Portal, known as the MLC Public Work Search.¹⁵⁵ The public can access the records in the Public Work Search, while verified “members”—payable publishers or songwriters—can use related portals to register their works or find and claim an ownership share in an already registered work.¹⁵⁶ Each MLC database entry contains comprehensive information about the listed song, including the percentage of ownership share in the work, the identity and location of the composition copyright owners, if known, and ownership information regarding the sound recording that embodies the musical work.¹⁵⁷ For example, Figure 6 shows MLC entries for two versions of SZA’s “Kill Bill.”¹⁵⁸ The entry for the song’s original release indicates that the MLC database has information for 100% of the “total known shares” of songwriter ownership, including songwriter names and publishers. In contrast, the version of “Kill Bill” featuring Doja Cat indicates that only 82.5% of the song’s ownership is accounted for in the MLC Portal, so the song’s publisher or collaborating songwriter may search the song and discover whether their share of the work remains unattributed.

Figure 6. MLC’s Comprehensive Database: The Public Work Search

Source: The MLC Public Work Search, supra note 21 (searched under “Work Title” for the song “Kill Bill,” added writer name criteria, and searched for Rob Bisel).

The MLC’s user-friendly database lowers the barrier to entry for copyright holders to assert ownership of their works, making it feasible for them to monitor their rights in licensing transactions. Additionally, by increasing assurances that songwriters will get paid correctly and reliably for what they are owed, the MLC’s Portal and Public Work Search should resolve gaps in songwriter pay and bolster the incentive for creators.

The MLC also works with another entity established in Title I of the MMA called the Digital Licensee Coordinator (“DLC”), which represents DSPs in the new post-MMA licensing system and helps the MLC carry out its mandate.¹⁵⁹ The DLC is required to “assist the [MLC] in the efforts . . . to locate and identify copyright owners of unmatched musical works” through active outreach to the songwriting community.¹⁶⁰ Together, the MLC and DLC create resources and host webinars oriented toward independent songwriters to educate them on how to assert their rights in the post-MMA copyright regime.¹⁶¹ The accessibility and transparency built into the MMA aim to further lower the transaction cost barrier for DSPs and rights holders alike to fulfill obligations that accompany the compulsory license.

Lastly, Title I of the MMA sought to make music composition copyright holders whole by capturing the millions of dollars of unpaid royalties held by the DSPs prior to the MMA’s passage, bestowing upon the MLC the responsibility of uniting historical unmatched royalties with their proper rights holders.¹⁶² In February 2021, twenty-one DSPs finally transferred over $424 million to the MLC in historical unmatched royalties and provided accompanying usage data to help determine how the money should correspond with eventual royalty payments.¹⁶³ Despite the DSPs’ continued use of songwriters’ content on their platforms over the preceding decade, this money represented the first mechanical royalty payment made for content with “unmatched” copyright owners. The MLC made its first payment of formerly unmatched royalties from the $424 million pool in June 2022.¹⁶⁴ As of June 2023, the MLC has distributed $24 million of the originally unpaid, unmatched royalties.¹⁶⁵ The MLC has not yet paid out all historical unmatched royalties; however, its success in doing so will play an essential role in restoring songwriters’ trust in the music copyright system to protect their work and provide the financial incentives promised by copyright’s underlying policy goals.

2. Unfolding Reality: The MLC in Action

As reasonably expected for the establishment of a new entity, the MLC was not operational until over two years after the MMA was passed, and only a few years have since passed for the MLC to prove its worth as a solution to the publishers’ and songwriters’ concerns.¹⁶⁶ That said, the MLC achieved a robust operation rather quickly, with over 30 million songs registered in its database as of March 2023¹⁶⁷ and having already distributed over $1.5 billion in mechanical royalties as of its recent metric announcement in October 2023.¹⁶⁸

While these metrics contextualize the growth of the MLC, they reveal less about the entity’s success in achieving its founding goals. In this regard, however, it also seems relatively successful in its early stages, though much remains to be seen.¹⁶⁹ As of March 2023, the MLC announced that it had achieved a consistent 89% match rate for all royalty payments passing through the collective,¹⁷⁰ and in October 2023, the collective achieved a just slightly higher match rate at 90%.¹⁷¹ This match rate refers to the rate at which the MLC can properly attribute royalties to an existing song in its database, not the rate at which royalties are actually claimed, which still requires a proactive assertion of ownership by the copyright holder.¹⁷² With approximately 25% of all royalties deemed missing or unmatched before 2018, matched royalties have increased by about 15% between the pre-MMA system and now due to the MLC’s database and matching system.¹⁷³ The MLC’s CEO, Kris Ahrend, suggests that the more data the MLC receives, the more it will be able to close out this gap; the MLC continues to grow its database with the participation of its publisher and songwriter members.¹⁷⁴ Most recently, the MLC has partnered with five music data matching vendors to improve its match rate, though the MLC’s greatest challenge will remain in gathering data on its most difficult songs—“those created by DIY, unsigned songwriters, many of whom are still unaware of The MLC.”¹⁷⁵

Additionally, concerning the historical unmatched royalties accrued before 2018, $200 million of the original $242 million pool paid in February 2021 has been matched with copyright holders, meaning that the MLC has almost fully accounted for the historical void of rights holder information in just under three years after its founding.¹⁷⁶ Though it has paid out only a small fraction of these royalties to date, this is due to circumstances outside the MLC’s control.¹⁷⁷

In sum, the MLC is, on paper, doing what the MMA intended—just perhaps more slowly than songwriters and publishers would like. Though more work remains, initial data suggests that the MLC has absorbed a significant amount of the existing transaction costs that complicated and cluttered the pre-MMA mechanical licensing system. DSPs can no longer get away with licensing content for free, hundreds of thousands of unverified NOIs have been eliminated, and composition rights holders can seek out their work and lay claim to what they are rightfully owed.¹⁷⁸ There are, of course, criticisms of the MLC as well. Songwriters have protested the MLC’s refusal to publicize the sum of money it holds in unmatched royalties, claiming that the MLC has an incentive to refuse transparency, as the majority of its board is comprised of publishing company representatives who disproportionately benefit if the unmatched royalty pool remains orphaned.¹⁷⁹ These grievances indicate songwriters’ ongoing scramble for remuneration at the bottom of the value chain.¹⁸⁰ Thus, even though addressing songwriter incentive was one of the MMA’s core goals, it has become likely that MLC can only do so much to remedy this concern.

B. A New Regime for Royalty Rate Setting

1. The MMA’s Solution: A Willing Buyer–Willing Seller Rate

As previously explored, one of the direst realities prior to the passage of the MMA was the rapidly disappearing songwriter profession. Statistics and widespread testimony illustrated just how much songwriters’ livelihoods had been altered by the rise of DSPs; this reality was a repeated feature of discussion at the Senate’s hearing on the MMA in 2018 and framed a sense of urgency around passing the law.¹⁸¹ While unmatched and unpaid royalties accounted for a fraction of songwriters’ diminished incentive, the bottom line remained that there was simply no longer enough money being generated by mechanical royalties in the digital age. The massive reduction in overall mechanical royalties caused by streaming’s replacement of physical albums or digital download sales and the long-depressed statutory royalty rates produced an untenable situation for songwriters. Though publishers, too, felt the strain on mechanical royalty revenue, the plight of songwriters demanded resolution by the MMA for copyright policy to achieve its stated goals.

Title I of the MMA, therefore, introduced a new standard for the CRB to use when determining mechanical royalty rates—not only for streaming but for all other mechanical royalties as well. The MMA shifted the CRB’s rate-setting standard from the highly criticized “policy-oriented” rate to a “willing buyer–willing seller” rate meant to reflect market demand. The standard, which the music industry had demanded for years, requires that the CRB “establish rates and terms that most clearly represent the rates and terms that would have been negotiated in the marketplace between a willing buyer and a willing seller.”¹⁸² As cited in another CRB rate proceeding that utilized the willing buyer–willing seller rate standard, “the Judges look to . . . a hypothetical marketplace, free of the influence of compulsory, statutory licenses.”¹⁸³

To make rate determinations under this standard, CRB judges are also to “base their decision on economic, competitive, and programming information presented by the parties,” which may include factors like the degree to which a compulsory licensee’s service interferes with or enhances a musical work copyright holder’s other revenue streams and the parties’ “relative creative contribution, technological contribution, capital investment, cost, and risk.”¹⁸⁴ Before the MMA, the CRB followed a “limited-evidence process,” and any market data was excluded from consideration when setting mechanical rates.¹⁸⁵ Thus, while the original 801(b) factors demanded that the CRB consider “fair income under existing economic conditions” as suggested by the incentive theory of copyright policy, pre-MMA rates clearly could not fully adjust for economic realities when the legal regime excluded market data from consideration.¹⁸⁶

Despite the full legislative history showing the initial access-incentive justification for the mechanical compulsory license,¹⁸⁷ in the 2015 Copyright Office Music Marketplace Report, the Office asserted that a compulsory license “should exist only when clearly needed to address a market failure.”¹⁸⁸ The report explicitly called on Congress to adopt a “uniform market-based ratesetting standard,” stating that “[t]here is no policy justification for a standard that requires music creators to subsidize those who seek profit from their works.”¹⁸⁹ In taking this position, the Copyright Office acknowledged that the supposed access-incentive reasoning behind the compulsory license overtly restrained creators’ sense of incentive by forcing them to underwrite licensees’ royalty costs at their own expense. The willing buyer–willing seller rate-setting standard had also already been implemented in most other CRB rate proceedings.¹⁹⁰ Some critics argue that shifting from a policy-oriented analysis to a willing buyer–willing seller analysis abandons copyright’s underlying policy goals;¹⁹¹ however, the failure of 801(b) factors to adequately support incentive, as shown by the Copyright Office’s findings, suggests that a willing buyer–willing seller rate could better account for some typical copyright policy considerations than the original policy factors themselves.

Additionally, songwriters and publishers advocated for this change in hopes that it would lead to higher compensation, reinforcing this provision’s goal of recalibrating the incentive-access tradeoff. Though the willing buyer–willing seller rate would not fully create free market mechanical rates, as would be possible without any compulsory license regime, theoretically, it at least mimics a free-market reality tempered by CRB review. David Israelite described these rate-setting alterations, stating that if “songwriters are to remain in a compulsory license prison,” the MMA’s rate-setting provisions “at least improve the conditions of their confinement.”¹⁹² As his tone connotes, publishers and songwriters generally would have much preferred the abolition of the compulsory license “prison” entirely, yet in the name of achieving consensus legislation, accepted the shift to the willing buyer–willing seller rate.¹⁹³ One skeptical songwriter described the songwriting community’s celebration of the willing buyer–willing seller rate as an “example of Stockholm Syndrome” in reaction to a “beautification” of the dreaded compulsory license.¹⁹⁴ Those who share this belief doubt that a willing buyer–willing seller rate regime will make any substantive difference in improving the massive incentive problem in the streaming era.

2. Unfolding Reality: Phonorecords IV Under a Willing Buyer–Willing Seller Rate

Since the MMA’s passage in 2018, only one CRB proceeding has been decided under the new rate-setting regime, and this proceeding provided only minimal foreshadowing as to whether the rate standard can bring composition copyright holders closer to fair compensation of decades past. Phonorecords IV, which set the royalty rates for musical compositions between 2023 and 2027, began in 2021.¹⁹⁵ At its outset, the key industry parties involved in the proceedings—the National Music Publishers’ Association, the Nashville Songwriters Association International, and the Digital Media Association (which represents the DSPs)—seemed poised for an extensive legal battle, as all submitted drastically disparate rate proposals to the CRB.¹⁹⁶

Phonorecords IV represented a major moment for many songwriters who eagerly anticipated the first rate proceeding to implement MMA’s new standard.¹⁹⁷ However, in a surprising turn, industry players ultimately settled in August 2022 before the CRB’s final adjudication, agreeing on a royalty headline rate of 15.35% of the DSP’s total annual revenue.¹⁹⁸ When the CRB approved the settlement and published its rate determinations in December 2022, it was rebuked by a number of songwriter groups, as they had held out hope that the new rate-setting standard would enable another needed rate increase.¹⁹⁹ Several opposition comments submitted to the CRB asserted songwriters’ concerns that the settlement “would thwart application of the willing buyer, willing seller rate setting standard,” which would have been applied for the first time in the absence of a settlement.²⁰⁰

In its final publication of the Phonorecords IV rate determination, the CRB did note that although the willing buyer–willing seller standard “was not expressly applied as it would be in a full proceeding, the operable rate standard exist[ed] as a relevant factor” when considering the validity of the settlement.²⁰¹ The settlement, thus, was not devoid of the new standard, though the extent of its impact on the CRB’s decision-making remains unclear. The CRB stated it found no “reasonable basis” to reject the settlement and its suggested reasons may allude to the influence of the new rate standard.²⁰² The CRB asserted that it found no persuasive economic argument for rejecting the settlement, particularly indicating that the rate structure was not “gratuitous.”²⁰³ Instead, the settlement included thoughtful changes to prior rates and was designed through a careful negotiation in which “complex terms in [the] commercial agreement” represented a “market [that is] itself . . . complex.”²⁰⁴ While this does not fully provide an analysis of how rates may have differed if they had been developed by the CRB under a pure willing buyer–willing seller standard, it suggests that the factors used to consider a free and competitive market do not necessarily lend themselves to as noticeably high of a rate change as many in the industry may have anticipated.

Additionally, the negotiation itself is viewed to represent prices set between a willing buyer and a willing seller, and, as a practical matter, this may forebode a stifled rate standard for future Phonorecord rate proceedings. Strong evidence suggests that regardless of settlement, the willing buyer–willing seller rate holds little weight when mechanical royalty rates have no hypothetical market of willing buyers and willing sellers—the mechanical royalty has never once been considered in the context of true market demand.²⁰⁵ In 1995, Congress enacted the Digital Performance Right in Sound Recordings Act, in which a willing buyer–willing seller standard similarly replaced 801(b) policy factors in CRB rate proceedings for noninteractive public performance royalties.²⁰⁶ In that situation, the Copyright Arbitration Royalty Panel, the predecessor to the CRB, was left with “the unenviable task of ascertaining what a willing webcaster and a willing record label would consider to be a fair deal for Internet radio royalties in the face of an almost total absence of real world evidence.”²⁰⁷ In the same vein, the willing buyer–willing seller mechanical license rate “seeks to construct or emulate [a ‘market’ that] does not exist and often has never existed.”²⁰⁸ Under this analysis, it seems possible that the NMPA, in pursuing this settlement over litigating for a desirable rate, may have filled the vacuum of market knowledge with a rate that will lay a foundation for future decisions and is less than desirable for publishers and songwriters in the long run. Notably, in past rate proceedings for a willing buyer–willing seller standard, the CRB found that “[t]here is no a priori reason to conclude that the rates set in [an] earlier proceeding failed to reflect or approximate market forces.”²⁰⁹

The Phonorecords IV settlement may suggest that, at least in the short term, the MMA’s new rate-setting standard will result in no real improvement in incentives for songwriters, as demanded to rectify the current imbalanced copyright policy. The early days of the Phonorecords IV proceeding may also provide insight into challenges posed by converting to this new rate-setting norm. Immediately after the parties to Phonorecords IV submitted their rate proposals, the CRB used the new willing buyer–willing seller standard to deny the NMPA’s request for further documentation regarding an initial lowball DSP proposal.²¹⁰ The CRB rejected their request for information from the DSPs explaining the “historical quantum” of DSP revenue to be deducted from the mechanical royalty pool under the DSPs’ rate proposal and information on the “impact” of those proposed deduction terms on mechanical royalties.²¹¹ The CRB held that such justifications for further documents may have been relevant under two of the former 801(b) standards of fairness and “disruption-avoidance,” but that such considerations of “impact,” “historical quantum” of revenue, and deduction terms no longer appeared to be appropriate when purely assessing market factors relating to the parties’ investment in their products and positions in relation to economic demand.²¹² This further suggests that though the standard may in some ways seem liberatory by providing a comparison to an open market, it may also limit composition copyright holders from making arguments about traditional copyright policy considerations, including fairness.

Despite early signs that the willing buyer–willing seller standard may not yield the rate increases publishers and songwriters sought, any overall rate increases may also raise a concern on the horizon for DSPs. Phonorecords III, which set rates for 2017 to 2022, was the last CRB proceeding conducted under the pre-MMA standard judged by 801(b) factors, and it introduced the most significant rate increase in the CRB’s history when it increased the headline rate from 10.5% of DSP’s revenue to 15.1%.²¹³ The DSPs appealed this major win for publishers and songwriters, and not long after the CRB affirmed Phonorecord III on appeal, Spotify and YouTube music raised their prices just a few days apart from each other, indicating that when incentives to copyright holders improve, the access variable, in turn, is affected as well.²¹⁴ Apple Music and Amazon Music also raised their prices the year before in 2022.²¹⁵ While this price increase occurred in the aftermath of the Phonorecords III, a proceeding that was not the product of MMA legal changes, it raises questions about the cost of the willing buyer–willing seller rate should it achieve its intended outcome of increasing copyright holder compensation. Increased royalty rates are of particular concern for Spotify, which has operated at a loss every year since its founding despite being the United States’ most popular music streaming platform.²¹⁶ As indicated by the price hikes, the burden of licensing costs is ultimately passed on to the consumer and impinges on public access to copyrighted works. Nonetheless, while this concern remains in the background, the balance between access and incentive likely still demands an increase in consumer prices to effectuate the increased incentive necessary to salvage the songwriting profession.

Table 2. Pre- and Post-MMA Policy Concerns
Policy Goal	Pre-MMA Score	Post-MMA Score
Incentives	Incentive for songwriters is LOW; attributable to a minimal statutory royalty rate, governmental restrictions on both types of songwriters’ royalties, and industry structure providing equal payout between publishers and songwriters.	MARGINALLY IMPROVED. It will require several more years to see how much the MMA altered incentives for creators. However, songwriter testimony strongly suggests that creators at the bottom of the copyright chain of value have not seen the MMA substantively move the needle on the incentive variable. This remains an urgent problem in the copyright policy triad, indicating that the equilibrium of policy goals remains unbalanced.
Access	Access is HIGH in the short term, with concern that this variable may shift if DSPs continue operating at a deficit.	HIGH, but the same problem remains as prior to the MMA. Costs will likely be pushed onto the consumer if royalty rates continue to rise in response to the unsolved incentive variable.
Transaction Costs	Transaction costs are HIGH. While the compulsory license reduces transaction costs of licensing negotiation, matching royalties to songwriters imposes high transaction costs for DSPs and songwriters alike.	VASTLY IMPROVED. The MLC has effectively begun to reduce transaction costs by mimicking the collective licensing and blanket license model of other areas of copyright law. It will take several more years to see the full efficiency of the MLC, but it has driven strong results thus far.

III. CONSIDERATIONS FOR THE FUTURE

A. Approaching the First MLC Five-Year Review

Section 102(a)(3)(B)(ii) of Title I of the MMA calls for periodic review of the MLC’s designation by the Copyright Office every five years to review the entity’s functioning and determine whether the MLC shall continue its designation as the MMA’s chosen organization to orchestrate and manage mechanical licensing.²¹⁷ January 2024 marked the beginning of this first review period, though it remains uncertain what the scope of this review will include.²¹⁸ The MMA suggests that the Copyright Office will solicit information to assess only the high-level question of continued designation rather than initiating a review with further recommendations or imperatives for the MLC’s improvement.²¹⁹ Some songwriters hoped the review would include more, having raised concerns on various alleged issues regarding songwriter representation within MLC’s Board and other critiques beyond the scope of this Note.

Currently, no known reason exists to suspect that MLC’s designation will change. The MLC attained substantial achievements thus far in a short time, establishing its infrastructure, creating a robust and growing database, building out resources, and paying out upwards of $1.5 billion in royalties.²²⁰ Its success was publicly affirmed by an array of industry players in a June 2023 House Judiciary Committee hearing on the MMA’s five-year milestone.²²¹ Even with successes, there has been insufficient time to prove the MLC’s full success or to render it ineffective so as to cause a deeper assessment of the MLC’s designation.

B. Reconsidering the Compulsory Mechanical License: Theoretical Shifts Versus Industry Realities

A handful of songwriters have used the pending five-year review as a moment to demand another, more fundamental review of the music licensing system—a review of the mechanical compulsory license.²²² The Copyright Board outright rejected this suggestion, stating that because only two and a half years have passed since the MMA’s licensing changes were implemented, it would be “premature at this time to engage in a new study of the Section 115 license.”²²³

While this demand was more of a pipe dream—and not within the MMA’s review requirements—the early success of the MLC certainly calls into question the basic necessity of a compulsory license. The mechanical compulsory license was once a fix-all solution to calibrating incentive, access, and transaction costs, but now, the MMA’s new collective licensing organization and blanket licensing system appear to pull their weight in correcting transaction costs. Even before the MMA implemented this model, it has been effective in multiple other areas of music copyright law as a solution to mass licensing numerous small, discrete pieces of content. Meanwhile, the compulsory aspect of the mechanical license continues to prevent a recalibration of the access-incentive equilibrium vis-à-vis an unprecedentedly high level of access and demand for music and an unprecedently low level of songwriter incentive to create it.

Theoretically, abolishing the mechanical compulsory license likely makes sense. In addition to being an anomaly in the American system, today, the U.S. remains the only country outside Australia to maintain compulsory licensing for mechanical rights.²²⁴ Its use has also clearly been controversial throughout its history; in 2004, at the earliest outset of streaming in the music industry, then-Register of Copyrights Marybeth Peters called on the House Judiciary Committee to eliminate the compulsory license, recognizing that it violated basic principles of copyright policy and was a deviation from music licensing norms in the rest of the world.²²⁵ Traditional economic assumptions would strongly suggest that, given the music copyright system’s intense reliance on mechanical licenses, high demand from licensees would compel licensing parties to reach a deal reflective of free market demand without obligating composition rights holders to license their work under a compulsory regime. In other areas of copyright law, the free-market forces are continually presumed to engender a successful equilibrium between incentive and access, and both here and in other licensing regimes, the blanket licensing scheme has compensated for other concerns by accounting for transaction costs.

Still, while copyright’s theoretical underpinnings may support such a change, it is widely accepted throughout the music industry that its abolition will never happen. The Copyright Office asserted that review of the compulsory licenses would be “premature” at this juncture and did not rule out the possibility somewhere down the line, should the implications of the MMA be more fully played out and the MLC prove a comprehensive alternative.²²⁶ However, throughout decades past, music industry players have voiced overpowering concerns that eliminating the compulsory license would cause “unnecessary disruptions” in the dynamics of the music industry.²²⁷ Even more critically, a vast majority of the music industry’s convoluted legal patchwork is owed to the outsized influence of vigorous and effective lobbying in Congress at the expense of copyright holders. Far too many players in the copyright landscape depend upon the compulsory mechanical license’s below-market rates, and lobbying forces at record labels, DSPs, radio stations, and beyond will likely ensure that this reality never comes to fruition.²²⁸

C. Songwriters’ Continued Plight Post-MMA

Despite the MLC’s relative successes thus far, the MMA’s fundamental flaw may be that it fixed the glaring problem of untenably burdensome transaction costs without addressing the looming, more existential incentive problem, which lies at the very core of copyright policy. In many ways, the distressing pre-MMA stories and statistics about songwriters explored in this Note have carried into the present day. As Grammy-nominated songwriter Erika Nuri Taylor told the Washington Post in 2023, “I got my real estate license [three years ago] because I thought I’m not going to be able to sustain being a creative person, a songwriter, for the next 10 to 15 years if nothing changes in the music industry.”²²⁹ Taylor, who has written for the likes of Enrique Iglesias, Janelle Monáe, and Meghan Trainor and has been a professional songwriter since 1992, says that her royalty income has been cut in half over the past five years, even after the hopeful enactment of the MMA.²³⁰ Today, she reluctantly considers herself a full-time real estate agent, holding out hope that she can still write songs on the side despite it once being her entire career. Her testimony is just one of many that have persisted after the MMA’s enactment.

Many songwriters will continue to create for the love of it, seeing their art form as a “spiritual calling” for which they have made great sacrifices.²³¹ Perhaps this indicates that despite great financial pressures, the songwriter profession may evolve to become less reliant on copyright’s promise of financial remuneration and will continue to be driven by an intrinsic desire to create. Alternative theories of copyright address this objection, arguing that perhaps copyright ownership and its accompanying “incentives” to create actually play a very minimal role in motivating creative work.²³² However, even amidst these alternative considerations and the possibility that songwriters may still create outside the bounds of adequate copyright incentives, songwriters persistently indicate the impact that low royalty rates across both mechanical and public performance royalties have on songwriting as a profession beyond its persistence as an art form.²³³ Another songwriter cites that modern-day music no longer reflects the caliber of decades past because much talent is lost when many just “can’t afford to do it anymore.”²³⁴ In practice, modern copyright regimes serve as the foundation for entire creative industries; here, the music copyright system’s success dictates the success of the music industry as a whole, making it rather alarming when copyright laws fail to sustain their promise of a “fair return” for the class of creators upon which the music industry is built, resulting in the dying of one of the industry’s most integral professions.²³⁵

Notably, there are also some critical confounding variables at play in the songwriter’s plight beyond anything the drafters of the MMA could have predicted. Given that the CRB did not finalize the DSPs’ appeal of the Phonorecords III rates for 2018 to 2022 until August 2023, the MLC could not pay out royalties for that term without established rates; this ultimately delayed the full payment of an additional $700 to 800 million in streaming royalties over the past five years.²³⁶ From the rights holders’ perspective, even while their royalties may have been increasing on paper since 2018, in reality, they have seen no such improvement in their financial circumstances, as their payments have been frozen at 2017 rates well into 2023. Many in the industry felt the arduous appeal process was disastrous for the publishing and songwriter economy, and Phonorecords IV ended in a settlement in large part because music publishers were adamant about avoiding the impossibly burdensome costs of litigation and further wreckage it could cause for songwriters and publishers.

Until composition rights holders receive full payment from Phonorecords III and routine payments on Phonorecords IV, it is hard to judge the potential adequacy of Phonorecord IV rates that arise from the MMA’s new rate standard or, more generally, how effective the financial incentives are for songwriters within the current copyright licensing system. Further, as suggested by the Copyright Office, it may take several more years for the MMA’s efficacy to come into full view, as both the MLC works toward achieving its envisioned efficiency and the willing buyer–willing seller gets tested in the Phonorecords V CRB proceeding.

CONCLUSION

In June 2023, the U.S. House of Representatives marked the MMA’s impending five-year anniversary with a House Judiciary hearing in Nashville, reopening a conversation with representatives from across the music industry about the law’s realized impact since its implementation in 2018.²³⁷ As discussion resumes on Capitol Hill, there is a renewed opportunity to reflect on the underlying objectives of a music copyright system and whether Title I has brought the current legal regime closer to those aims. Sustained conversation on these issues is more pressing than ever, as many in the music industry fear disruptive impact of artificial intelligence on a system already in slow rehabilitation from the shift to the digital age.

In taking stock of the industry’s current position, Title I of the MMA represents a major success in many ways. Its actualization was made possible by an unprecedented will within the music industry, with often opposing parties rallying behind the conviction that the music copyright system urgently needed to change. Title I of the MMA has pulled the industry from the brink of crisis, effectively eliminating untenable transaction costs that left an antiquated licensing system nearly non-functional.

In other ways, however, Title I also represents a band-aid solution to a crisis that boiled over in a fragmented legal scheme—a consequence of years of balkanized laws and decision-making heavily influenced by lobbying forces rather than fully reasoned policy. The industry’s united front may have emerged not from a desire to affirm value of the copyright system or artists’ creative works, but merely from the recognition that there was no other way out beyond a legal solution. As a result, Title I of the MMA has achieved relatively marginal results for one of its central aspirations and one of the industry’s most looming concerns—the hollowing out of the songwriting profession. The plight of the songwriter thus remains one of the music industry’s most fundamental yet unanswered problems in the digital age.

98 S. Cal. L. Rev. 205

Download

Tara Steinmetz*

* Articles Editor, Southern California Law Review, Volume 98; J.D. 2025, University of Southern California Gould School of Law; B.A. 2020, University of California, Los Angeles. Thank you to my advisors, Professors Jonathan Barnett and Emio Zizza, for their support and guidance, and to the members of the Southern California Law Review for their hard work and thoughtful suggestions on my Note.

Sidestepping Sovereign Trademark Territorialism: The Harmonization and Customization of the Lanham Act

March 2025March 2025Gabriella F. Franco*

Trademark territoriality—the principle that trademarks have a separate legal existence in each national jurisdiction—has been deemed a basic tenet of domestic and international trademark law; however, it is incompatible with a globalized, modern world because borders do not dictate consumer interest in products, and brands do not operate within the confines of sovereign territories. Practically, the principle no longer exists in its originally intended form due to U.S. courts’ circumvention and international treaties’ attempts at harmonization. While current international agreements provide suggested, streamlined standards, they do not establish enforceable rules. The need for an update to the international trademark system is more pressing than ever before due to the Supreme Court’s 2023 decision in Abitron Austria GmbH v. Hetronic International, Inc., which effectively eradicates the Lanham Act’s extraterritorial application. This Note traces the erosion of the territoriality principle through the seventy years of jurisprudence preceding Abitron, the Famous Marks Doctrine circuit split, and the growing practice of stealth trademark filing. The Note then advocates for a dual system that preserves the benefits of a territorial-bound trademark system—customization of trademark laws and international comity—and caters to the demands of a globalized economy in which brands transcend borders. A legally enforceable, self-executing international system would provide trademark owners with a cost-effective, efficient, and reliable approach to trademark protection. This especially benefits small to medium-sized enterprises that are entering multinational markets more easily and may not be equipped with the resources to navigate foreign intellectual property protection. Finally, the Note suggests options for the dual system modeled after the European Union Trademark, which it coins International Community Marks, as well as the Uniform Domain Name Dispute Resolution Policy, whose independent, international anti-cybersquatting framework has served as an effective alternative to traditional sovereign enforcement.

INTRODUCTION

Borders do not dictate consumer interest in products, and brands do not operate within the confines of sovereign territories. Consumers from New York to New Delhi want to purchase the newest iPhone, and brands like Louis Vuitton and Nike want to market their bags and shoes to customers across the globe. Despite the ubiquity of brand names and the dominance of international commerce, trademark law continues to operate on a geographical basis, reminiscent of a time when trade was local and burdened by transportation and information costs. The rights of a brand established in the United States start and end within the domestic borders despite the brand’s influence on consumers in foreign jurisdictions. This is not only problematic for brands, but consumers are harmed by deceivingly similar brand names and inauthentic products. Counterfeit products—$2 trillion of which are sold to Americans annually—injure consumers emotionally and physically, with inauthentic products leading to death and serious injury.¹

The Lanham Act, the predominant U.S. statute governing trademarks, does not extend beyond the territorial jurisdiction of the United States. According to the Supreme Court’s recent decision in Abitron Austria GmbH v. Hetronic International, Inc., courts should apply a “canon of statutory construction known as the presumption against extraterritoriality”² when evaluating the Lanham Act.³ This requires courts to presume that legislation does not extend beyond U.S. borders absent explicit legislative intent to the contrary. Thus, unless Congress adds a specific provision to the Lanham Act permitting its extraterritorial application, courts will continue to limit its reach to acts within U.S. borders, adhering to the territoriality principle.

However, the presumption against extraterritoriality is incongruent with the modern economy, global brand operation, and the last seventy years of U.S. jurisprudence. The rise of e-commerce marketplaces has streamlined the leap from local to international operations.⁴ Not only do they reduce the marketing costs typically associated with cross-border expansion, but they have also reduced transaction costs by working with governments to eliminate tariffs and intermediaries.⁵ As more brands enter the global market, the need for comprehensive intellectual property protection expands. In contending with this increased globalization, U.S. courts have chipped away at the territoriality principle. Thus, a principle that has historically been referred to as “basic to trademark law” no longer appears to serve such a central role.⁶

Due to differences in sovereign rulemaking and the growing global economy, the international community has taken steps to harmonize intellectual property law over the last 140 years. Harmonization occurs when “varying laws of different sovereign entities are changed to more closely reflect a common set of legal principles agreed to by those sovereign entities.”⁷ This has primarily taken the form of international treaties and registration systems, such as the Paris Convention for the Protection of Industrial Property⁸ and the Madrid Protocol.⁹ However, harmonization is not the same as international rulemaking, and it “does nothing to affect the essence of territoriality.”¹⁰ These agreements provide suggested, streamlined standards but do not establish enforceable rules. Thus, even though U.S. brands are not entirely unsupported in expanding their marks into foreign jurisdictions, current international agreements lack the force required for true international cooperation.

While enforceable harmonization appears crucial in the ever-growing international economy, there are benefits to the current territorial-bound system. The justifications for trademark law territoriality extend beyond the notions of sovereignty and international comity—“a judicial expression of one state’s respect for another state’s internal sovereignty because it prevents courts from interfering in the other state’s internal affairs”¹¹—because territoriality also provides the benefit of customization. The “separate legal existence” of trademarks in each national jurisdiction permits sovereigns to customize their trademark law in ways that may benefit trademark owners even if it is inconsistent with the international approach.¹²

This Note provides a review of the erosion of the Lanham Act’s territoriality principle in light of Abitron and presents alternatives to jurisdictional-bound trademark agreements by suggesting a new international model that balances customization with real harmonization. Part I provides a general overview of trademark law in the United States. This Part also explores the costs and benefits of the current approach to trademark territorialism. Part II illustrates the erosion of the territoriality principle. First, this Part explains the impact of Steele v. Bulova Watch Co. on trademark territoriality, the following seventy years of jurisprudence that further eroded the principle, and the Supreme Court’s recent disregard of this precedent in Abitron. Second, it describes the Famous Marks Doctrine as an exception to the territoriality principle. Finally, it discusses how the loopholes to trademark legislation, specifically stealth filing practices, that result from reconciling a global economy with a jurisdictional-bound system further erode the principle. Part III addresses the current approach to harmonization and the way brands practically operate in a territorial-bound trademark system. Part IV explores alternative approaches to U.S. trademark law that would balance the Lanham Act’s territoriality and customization with the harmonization of an international registration system. This Part considers the use of an approach modeled after the European Union Trademark, which this Note coins as International Community Marks, as well as the Uniform Domain Name Dispute Resolution Policy (“UDRP”), whose independent, international anti-cybersquatting framework has served as an effective alternative to traditional sovereign enforcement. This Part acknowledges the potential obstacles and downsides to these solutions, but ultimately concludes that they serve as a viable starting point.

I. TRADEMARK LAW BACKGROUND

Trademarks are the most common form of intellectual property, with 18.1 million trademarks filed in 2021 and only 3.4 million patents filed in the same year.¹³ A trademark consists of “any word, name, symbol, or device, or any combination thereof” used to “identify and distinguish” the source of goods and services.¹⁴ The goals of a trademark include distinguishing an owner’s goods from others, signifying the source and control of goods with a common trademark, indicating a uniform level of quality, advertising and selling the goods and services, and serving as a symbol of goodwill.¹⁵

Because a trademark serves as an indicator of source, similar marks for similar goods have the potential to generate consumer confusion. The law’s goal in granting protection to certain marks is to avoid consumer confusion¹⁶ and safeguard brand owners from misappropriation of their established goodwill.¹⁷ A trademark owner’s right to use a mark is based on priority. An owner has priority, and consequently an exclusionary trademark right, when the mark has been used prior to others. This is the idea of first in time, first in right.¹⁸ However, what constitutes “first in time” varies across sovereign jurisdictions. Thus, while trademarks serve the same purpose and mostly convey the same message across borders, protections and procedures can be unique in each country.

A. Domestic Trademark Law

Trademark law in the United States was historically rooted in common law and left to the states. This configuration lasted until the first federal trademark law was introduced in 1870; however, that law was later deemed unconstitutional.¹⁹ Subsequent laws in 1881²⁰ and 1905²¹ attempted to fill the void of national trademark legislation but still did not meet the demands of the 20th century. In 1946, the Lanham Act substantially revised federal trademark law and established a national registration system and cause of action for federal trademark holders. Since then, the Act has undergone several amendments and is codified as the Lanham Act, 15 U.S.C. §§ 1051–127.²²

The Lanham Act coexists with common law and state-level trademarks; however, federally registered marks grant additional protections to trademark owners.²³ The protections consist of inclusion in a database that provides public notice to anyone searching for similar trademarks; a “legal presumption” of ownership; a basis for filing for foreign trademark protection; an opportunity to sue in federal court; a right to use the trademark registration symbol, ®, to deter infringing use; and a recorded registration with U.S. Customs and Border Protection to stop the importation of infringing goods.²⁴ While the Lanham Act serves as the basis for an infringement claim and confers substantial substantive rights on mark holders, federal law does not create trademarks.²⁵ Instead, the United States operates on a use-based system, requiring that marks be used in commerce to create a right of exclusion. Using the mark in commerce is critical, as trademarks are not directly protected by the Constitution like patents and copyrights,²⁶ but rather by the Commerce Clause.²⁷ However, the Trademark Law Revision Act of 1988 modified the first-to-use system through the introduction of intent-to-use (“ITU”) trademark applications.²⁸ ITU applications create a first-to-file option in the United States, even though the force of an ITU registration is predicated on subsequent actual use.

Due to the Trademark Law Revision Act, there are two ways to register a trademark on the federal register: (1) a use-based filing or (2) an ITU filing. A use-based filing, under Lanham Act section 1(a),²⁹ requires that the applicant use the mark in commerce. “Use in commerce” means the mark is affixed to goods or displayed in the selling or advertising of services and then sold or transported in interstate or foreign commerce.³⁰ This use establishes the applicant as the senior user, conferring an exclusionary right of priority against subsequent junior users. Under section 1(b),³¹ an ITU filing allows an individual or company to apply for a mark’s registration before its use in commerce. ITU filings allow applicants to reserve an earlier priority date, as the date establishing exclusive use is the filing date as opposed to the date of first use in commerce. However, the mark’s registration is not official until an applicant submits proof of a bona fide use in commerce to the United States Patent and Trademark Office (“USPTO”) within six months of the original filing date.³²

United States trademark law is based on the territoriality principle, meaning trademarks have “a separate legal existence” in each sovereign territory.³³ This principle establishes a presumption against extraterritoriality, which indicates that, absent a contrary intent, Congress’s legislation “is meant to apply only within the territorial jurisdiction of the United States.”³⁴ The reasoning behind this presumption is to avoid “international discord,” as “Congress generally legislates with domestic concerns in mind.”³⁵ Therefore, U.S. courts will recognize a trademark if it is registered with the USPTO or has earned common law rights through domestic use of the mark in commerce.

However, this strict adherence to territoriality is only a function of the last century’s jurisprudence. One hundred years ago, the Supreme Court made the pivotal decision in A. Bourjois & Co. v. Katzel³⁶ to move away from the universality principle—which acknowledges trademark rights regardless of the sovereign territory in which they are registered or recognized—to the territoriality principle. The Court has since further clarified its absolute rejection of the universality principle.³⁷ Despite the stated commitment to the territoriality of U.S. trademark law, the Supreme Court has made exceptions to this rule, and lower courts have taken a more nuanced approach when dealing with the reality of global brand names and international commerce.

B. The Costs and Benefits of Territoriality

1. Territoriality’s Costs

Two principal problems arise from trademark territoriality. The first arises from infringement by a third party in a foreign jurisdiction. The second occurs when a foreign trademark is infringed in domestic territory. While these are two sides of the same coin, the problems illustrate the distinct responses required for domestic and foreign conduct.

Infringement by a third party in a foreign jurisdiction falls outside the scope of the United States’ territorial jurisdiction and is therefore governed by the foreign jurisdiction’s law. Thus, an American company’s mark could be subject to unabated infringing use if the company lacks foreign trademark protection. For example, if Starbucks, a company with a registered U.S. trademark, “STARBUCKS,”³⁸ wanted to franchise its coffee shops in Argentina, it could not rely on its U.S. trademark. Thus, if it did not have an Argentinian trademark registration and a third party had registered the mark “STARBUCKS” in Argentina, Starbucks would likely be unable to register its mark.³⁹ Without a recognized Argentinian trademark right, Starbucks could not prevent the foreign company from co-opting its goodwill. In addition to franchising, there are several reasons why a domestic trademark owner may want to protect its trademark abroad: an owner may want to license the trademark, monetize the goods or services through online sales, maintain an air of exclusivity, or protect consumers from the sale of counterfeit goods.

The second problem arises when foreign trademark owners want to protect their marks in the United States. Domestic trademarks that misappropriate a trademark owned by a third party in a foreign jurisdiction, which fall outside the scope of the United States’ jurisdiction, are not liable for infringement. For example, if Louis Vuitton, the famous French fashion house, wanted to expand its operation to the United States but did not have a U.S. registration, it would be subject to any preexisting U.S. trademark registration. Thus, if a domestic company had already been operating with the trademark “LOUIS VUITTON” on handbags, the French fashion house would be unable to stop the domestic company’s use and would likely be prohibited from registration of its established mark. This would likely be true even if the domestic company adopted the mark in bad faith and was willfully trading off the mark’s goodwill.⁴⁰ Similar reasons motivating domestic owners to seek trademark protection abroad apply to foreign owners seeking U.S. protection.

The first problem, namely the Starbucks hypothetical, is evidenced in a string of cases spanning seven decades, in which domestic trademark owners attempted to use the Lanham Act to protect them from foreign infringing activity. These cases generally established that the Lanham Act extends extraterritorially, providing owners with protection, but used various tests to determine when extraterritorial application was appropriate.⁴¹ However, a 2023 Supreme Court decision has effectively eliminated extraterritorial application of the Lanham Act, leaving owners susceptible to unabated foreign infringement.⁴² A circuit split between the Second and Ninth Circuits, which have adopted conflicting approaches to domestic recognition and protection of famous foreign marks, exhibits the second problem.⁴³ These problems, and courts’ attempts to remedy them, have eroded the Lanham Act’s territoriality principle.

2. Territoriality’s Benefits

The territoriality principle’s rationale is typically rooted in international comity and jurisdictional limitations.⁴⁴ However, territoriality also allows sovereigns to customize trademark law to their citizens’ needs and to changing economic and industry demands. Changes to a nation’s law may vary depending on its citizenry’s needs and its willingness or ability to respond to these demands. Further, adapting to modern-day needs is challenging at scale, as evidenced by the century-wide gaps in treaty amendments,⁴⁵ but more manageable at a country level. Therefore, even in a global economy, sovereign customization likely benefits trademark owners.

For example, the United States’ first-to-use system stands in contrast to other countries’ first-to-file system. Bona fide use of a mark in commerce grants protection to U.S. mark owners, even for unregistered marks. This benefits small, unsophisticated mark holders, who may not have considered registration or understood the benefits conferred by a federally registered mark. In contrast, China operates on a strict first-to-file basis, withholding protection from individuals who have used the mark in commerce but have not received a trademark registration.⁴⁶ With typical waiting periods of nine to twelve months, Chinese applicants need the foresight to apply nearly a year before any use of a mark will be protected.⁴⁷ China’s system is less desirable for first-time mark owners who may jeopardize their marks’ right to exclusivity.

Thus, national-level customization of trademark law can prove to be beneficial to trademark owners notwithstanding the challenges it presents when owners operate transnationally.

II. EROSION OF THE TERRITORIALITY PRINCIPLE

Despite the territoriality principle’s pervasiveness, it is in tension with the real-world operation of trademarks. The global economy and flow of commerce have significantly transformed since the conclusion of the Second World War, with the ratio of exports to world gross domestic product (“GDP”) increasing from 5.5% in l950 to 17.2% in 1998.⁴⁸ Further, the turn of the century assisted the rise of international trade through the widespread adoption of the internet, reduction of transport costs, and elimination of policy-related tariff and non-tariff barriers.⁴⁹ The United States International Trade Commission (“ITC”) estimates that digital trade has increased U.S. GDP by 3.4% to 4.8% and created up to 2.4 million jobs.⁵⁰ In 1994, the North American Free Trade Association (“NAFTA”) lifted tariffs and increased international trade between the United States, Mexico, and Canada from roughly $290 billion in 1993 to more than $1.1 trillion in 2016.⁵¹

As trade has become increasingly global, international travel has become more accessible, and the flow of information over the internet has become instantaneous, the world has seemingly become smaller and more interconnected than ever.⁵² Trademarks are present in the flow of international products and are adorned to goods and services in digital commerce, conveying meaning to individuals in varying sovereign states and establishing goodwill that transcends borders. Therefore, “[d]ogmatic territoriality . . . ignores basic reality because informational products cannot be located at a particular spot on the globe.”⁵³

In addition to territoriality’s incongruence with real-world trademark operation and the current economy, obedience to the principle has diminished. Courts have sidestepped the principle by carving out exceptions when its application is impractical or counter to public policy. This Part will first address the judicial weakening of the principle that began in the 1952 case Steele v. Bulova Watch Co.⁵⁴ and continued through circuit court decisions until the 2023 Supreme Court decision in Abitron Austria GmbH v. Hetronic International, Inc.⁵⁵ These cases center around the infringing use of domestic marks in foreign territories and are in the first trademark territoriality problem category—like the Starbucks hypothetical, in which the brand wants to stop infringing use in Argentina, the owners of domestic trademarks want to use the Lanham Act to prevent use of their marks abroad. Mark owners seek protection from the Lanham Act instead of foreign law governing the territory where the infringement occurred because U.S. trademark laws are often more protective, utilize broad discovery rules, and award high damages.⁵⁶ This Part then addresses the judiciary’s weakening of the territoriality principle in cases involving the presence of foreign marks in domestic courts, as evidenced by the Second and Ninth Circuit split over the Famous Marks Doctrine. This falls into the second category—like Louis Vuitton, these foreign brands either want to enter the domestic market or prevent others from using their mark in the domestic market. Finally, this Part turns to reciprocal priority, a legislative exception to territoriality, in which the United States recognizes foreign acts as the basis for domestic exclusionary trademark rights. Reciprocal priority attempts to harmonize the international trademark system, but it creates loopholes that further fracture the territoriality principle.

A. From Steele to Abitron

The Supreme Court first addressed the extraterritoriality of the Lanham Act in the 1952 case Steele v. Bulova Watch Co., in which the Court applied the Act to infringement in a foreign jurisdiction. However, the Steele decision did not establish a clear framework for extraterritorial application. In the seventy years that followed, lower courts were left to contend with an unclear set of principles used to override the presumption against extraterritoriality. This ambiguity led to a fragmented approach taken by circuit courts in addressing the issue, with seemingly every circuit developing a unique test.⁵⁷ By providing mechanisms to override the territoriality principle, most tests confirmed the incompatibility of the principle with the trademark system. However, in 2023, the Supreme Court addressed the extraterritoriality of the Lanham Act for the second time. In Abitron Austria GmbH v. Hetronic International, Inc., the Court ignored Steele’s precedent and the following seventy years of case law; instead, it developed a test that, in practice, prevents the Lanham Act’s extraterritorial application.

1. Setting the Standard

In 1952, the Supreme Court carved out an exception to the territoriality principle in the case Steele v. Bulova Watch Co.⁵⁸ Steele, a U.S. citizen, stamped “BULOVA”—a registered trademark owned by the prominent New York watch manufacturer, Bulova Watch Company—on watches he sold in Mexico. Steele strategically moved his watch operation from San Antonio to Mexico City, where the mark was not registered, to trade on the manufacturer’s international goodwill.⁵⁹ Steele purchased watch parts in the United States but assembled and sold all watches in Mexico under the trademark “BULOVA,” which he registered in Mexico. The counterfeit Mexican watches made their way into the United States, and the New York manufacturer received complaints about the Mexican watches from consumers who believed the watches to be its products. Bulova sued Steele, alleging that he violated Lanham Act section 32(1).⁶⁰ It also sued Steele in Mexico, and the Mexican Supreme Court nullified his trademark registration.⁶¹

The Supreme Court affirmed the Fifth Circuit’s injunction of Steele’s behavior in Mexico by “exercising its equity powers [to] command persons properly before it to cease or perform acts outside its territorial jurisdiction.”⁶² Due to Steele’s U.S. citizenship, the Court claimed that it was “not debarred by any rule of international law from governing the conduct of [its] own citizens . . . in foreign countries.”⁶³ Despite the watches being assembled and sold in a foreign nation and the district court’s decision to dismiss the case for lack of jurisdiction, the Court determined that the Lanham Act’s power, rooted in its ability to regulate commerce, extends to U.S. commerce in foreign jurisdictions. Additionally, Steele’s actions “were not confined within the territorial limits of a foreign nation.”⁶⁴ Thus, the Court seemed to rest its decision on three factors: (1) Steele’s U.S. citizenship, (2) the trickling of the counterfeit watches into the United States and their impact on domestic commerce, and (3) the lack of conflict with Mexican law due to the cancellation of Steele’s Mexican mark.⁶⁵

The Court’s decision in Steele contorted the rule of territoriality around the Court’s goal of enjoining the citizen’s behavior. By disregarding this traditional doctrine, the Court chose to interpret the Lanham Act in a way that is inconsistent with the deference typically given to Congress to determine the extraterritorial scope of its legislation.⁶⁶ Justice Reed’s dissent illuminates the majority’s departure from the territoriality principle. He noted that the Court’s disapproval of the citizen’s foreign actions does not warrant the application of the Lanham Act to acts outside the sovereignty of the United States.⁶⁷ Justice Reed harkened back to territoriality’s origins in international comity, respecting other nations’ sovereignty, and nations’ “capab[ility] of punishing infractions of their own laws.”⁶⁸ Despite this backlash or flawed reasoning in the Steele decision, it set the stage for the Lanham Act’s next seventy years of extraterritoriality.

Since this decision, lower courts have grappled with Steele’s lack of direction. Steele did not establish a clear test for determining the extraterritoriality of the Lanham Act; however, courts have interpreted Steele’s decision as establishing a three-part test. The test considers (1) the defendant’s nationality, (2) the conduct’s effects on domestic commerce, and (3) conflicts or potential conflicts with foreign law.⁶⁹ Circuit courts have created variations of this test, with varying degrees of leniency toward overcoming the presumption against extraterritoriality.

2. Reeling in Extraterritorial Application

Shortly after the Steele decision, the Second Circuit was tasked with determining whether section 32(1) should apply extraterritorially in Vanity Fair Mills, Inc. v. T. Eaton Co.⁷⁰ The court took a measured step in not further eroding the Act’s territoriality by limiting the circumstances in which the presumption against extraterritoriality may be overcome.⁷¹

Vanity Fair Mills, a U.S. underwear manufacturer, registered the mark “VANITY FAIR” in the United States and sold its merchandise in the United States and Canada.⁷² The following year, T. Eaton Co., a Canadian manufacturer, registered the mark “VANITY FAIR” for apparel in Canada. Vanity Fair was denied Canadian registration of its mark due to T. Eaton’s registration. Thus, Vanity Fair sued T. Eaton, seeking injunctive relief under the Lanham Act.

In determining whether the Lanham Act applied to T. Eaton’s extraterritorial conduct, the court applied the three main considerations in the Steele decision.⁷³ The court declined extraterritorial application because the first and third factors were not met: T. Eaton was a Canadian citizen, despite having U.S. employees, and it had valid Canadian trademark rights for the mark “VANITY FAIR.”⁷⁴ The court noted that the absence of one factor would not necessarily defeat extraterritorial application but that the absence of two is “certainly fatal.”⁷⁵

The Vanity Fair test also sought to tighten the three-part Steele test by altering the second factor to require a “substantial” effect on commerce.⁷⁶ Steele’s tripartite test did not “discuss a particular degree of effect that was necessary” but only required an adverse effect on U.S. commerce.⁷⁷ However, because the court did not rest its decision on the substantial effect of T. Eaton’s actions on U.S. commerce, the new element was dicta, and the court added it without guidance as to what constitutes a substantial effect.⁷⁸

3. Adapting to Extraterritoriality

While Vanity Fair attempted to limit Steele’s application, later tests generated by the Fifth, First, and Ninth Circuits did the opposite. The following cases demonstrate the way circuit courts liberally interpreted the Steele test to provide an easier path forward for the extraterritorial application of the Lanham Act, further eroding the Act’s territoriality principle.

The Fifth Circuit returned to the less strict version of Steele’s tripartite test in its 1983 case American Rice, Inc. v. Arkansas Rice Growers Cooperative Ass’n.⁷⁹ American Rice, Inc. (“ARI”) and Arkansas Rice Growers Cooperative Association (“Riceland”) were both U.S. farming cooperatives that sold rice to Saudi Arabia.⁸⁰ ARI sold under the registered U.S. trademark “ABU BINT,” which, translated from Arabic, means “of the girl” or “girl brand”; used a logo of a young girl; and used a combination of red, yellow, and black on its packaging.⁸¹ Riceland began selling rice with the trademarks “BINT AL-ARAB,” translating to “daughter of the Arabs” and “Gulf Girl.”⁸² It also used a red, yellow, and black color combination on its bags of rice.⁸³ ARI filed suit against Riceland, alleging trademark infringement under section 32(1)(a) and false designation of origin under section 43(a)(1)(A)⁸⁴ of the Lanham Act.⁸⁵

In determining the Lanham Act’s extraterritorial application, the court confirmed that the relevant factors include the defendant’s citizenship, the effect on U.S. commerce, and conflict with foreign law; however, the court noted that this is a balancing test whose factors serve a primary but nonexclusive role.⁸⁶ Thus, the court held that the Act applies extraterritorially to Riceland’s acts because it is an American citizen, its acts had a “more than an insignificant effect” on U.S. commerce, and it does not have a legal right to use its mark in Saudi Arabia.⁸⁷

The Fifth Circuit’s test returned to the more lenient approach to the commerce factor. The court’s choice to remove the Vanity Fair “substantial” requirement was because the Steele test “contains no such requirement, and that some effect may be sufficient.”⁸⁸ Thus, the court’s test allows the Lanham Act to extend more easily beyond U.S. borders—compared to the Second Circuit’s test—in conflict with the territoriality principle.

In 2005, the First Circuit reiterated that Steele “settled that the Lanham Act can, in appropriate cases, be applied extraterritorially.”⁸⁹ Nevertheless, the court crafted its own approach to determining the Act’s extraterritorial application in McBee v. Delica Co.

Delica, a Japanese corporation, began using the mark “CECIL MCBEE” to denote its children’s clothing line.⁹⁰ Delica did not ship goods into the United States but had a website for the clothing brand. Cecil McBee, a well-known American jazz singer who had toured throughout Japan but never registered a trademark there, claimed Delica’s unauthorized use created a misleading perception of endorsement. The jazz singer filed suit against the company for unfair competition under section 43(a)(1)(A) of the Lanham Act.⁹¹

The court transformed the Vanity Fair factors test by disaggregating its elements.⁹² In this new analysis, the court first looks to the defendant’s citizenship. If the defendant is a U.S. citizen, the test ends at step one and the Lanham Act extends extraterritorially.⁹³ If not, the court considers if there is

a substantial effect on U.S. commerce. If there is a substantial effect, the court may consider the effect of intentional comity, depending on the circumstances.⁹⁴

In applying this test to the facts, McBee did not overcome the presumption against extraterritoriality.⁹⁵ Delica’s lack of U.S. citizenship required the court to proceed to step two. The court determined that Delica’s use of “CECIL MCBEE” did not have a substantial effect on U.S. commerce because U.S. consumers were not exposed to Delica’s products, as they were not shipped into the country, and there was “virtually no evidence that American consumers [were] actually seeing Delica’s products” online.⁹⁶

While the court followed Vanity Fair’s stricter “substantial” effects test, it lessened the hurdles required for extraterritorial application of the Lanham Act for defendants with U.S. citizenship. In applying the McBee test to Steele, Steele’s U.S. citizenship alone would have been enough to reach his foreign conduct, and the Court would not have been required to examine his acts’ effect on U.S. commerce. Thus, the First Circuit’s test further erodes the Lanham Act’s territoriality because, unlike Steele and American Rice, a defendant’s U.S. citizenship alone is sufficient to apply the Lanham Act to acts in a foreign jurisdiction.

Finally, in 2016, the Ninth Circuit generated a rule of reason balancing test for determining the Lanham Act’s extraterritorial application.⁹⁷ Under this test, the Lanham Act applies extraterritorially if

(1) the alleged violations . . . create some effect on American foreign commerce; (2) the effect [is] sufficiently great to present a cognizable injury to the plaintiffs . . . ; and (3) the interests of and links to American foreign commerce [are] sufficiently strong in relation to those of other nations to justify an assertion of extraterritorial authority.⁹⁸

The effect on American commerce does not have to be “substantial or even significant.”⁹⁹ In fact, the court noted that reputational harm alone may constitute “ ‘some effect’ on American commerce.”¹⁰⁰ Prong three concerns international comity and weighs seven factors.¹⁰¹

In Hallatt, Michael Norman Hallatt, a U.S. Lawful Permanent Resident (“LPR”), purchased goods from Trader Joe’s, a well-known American grocery store, and drove them across the border to sell at inflated prices at his Canadian storefront, “Pirate Joe’s.”¹⁰² Hallatt designed his store with Trader Joe’s classic theme and colors and filled it with the grocery store’s products. Trader Joe’s banned Hallatt from its storefronts, but he continued to frequent the stores in disguises and hired third parties to purchase goods on his behalf.¹⁰³ Trader Joe’s sued Hallatt, alleging infringement, unfair competition, and false advertising under the Lanham Act.¹⁰⁴

Hallatt’s actions satisfied prongs one and two. He engaged in commercial activity in the United States when he sourced the grocery products from domestic stores, and his Canadian activity could harm Trader Joe’s reputation due to a lack of quality control and inflated prices.¹⁰⁵ Prong three permitted extraterritorial application because of the lack of ongoing proceedings in Canada, Hallatt’s subjugation to U.S. laws due to his status as an LPR, the ease of enforcement, domestic consumer confusion, and the intentional harm inflicted.¹⁰⁶

Hallatt’s test permits an attenuated “nexus” between the infringing activity and domestic commerce that guts the Lanham Act of its territorial principle.¹⁰⁷ The court does not stop at dismissing Vanity Fair’s “substantial” requirement; instead, it shifts the standard to the other end of the spectrum by indicating that “some” effect is sufficient.¹⁰⁸ In a global market with instantaneous online communication and advertising, nearly anything can have a nexus with domestic commerce. This is especially true when reputational harm is sufficient.

4. Disregarding Steele

On the second to last day of the Supreme Court’s 2022 term, it disregarded almost a century of trademark law precedent. While the same Court, seventy years prior, rebutted the presumption against extraterritoriality, the current Court decided to quickly “put aside” Steele in its analysis.¹⁰⁹ Instead, it relied on its recent decisions in Morrison v. National Australia Bank Ltd. and RJR Nabisco, Inc. v. European Community to reinforce the presumption.¹¹⁰

In RJR Nabisco, the Court established a two-step framework to determine whether a statute has extraterritorial reach:

At the first step, we ask whether the presumption against extraterritoriality has been rebutted—that is, whether the statute gives a clear, affirmative indication that it applies extraterritorially. We must ask this question regardless of whether the statute in question regulates conduct, affords relief, or merely confers jurisdiction. If the statute is not extraterritorial, then at the second step we determine whether the case involves a domestic application of the statute, and we do this by looking to the statute’s “focus.” If the conduct relevant to the statute’s focus occurred in the United States, then the case involves a permissible domestic application even if other conduct occurred abroad; but if the conduct relevant to the focus occurred in a foreign country, then the case involves an impermissible extraterritorial application regardless of any other conduct that occurred in U.S. territory.¹¹¹

This framework, applied initially to the Racketeer Influenced and Corrupt Organizations Act,¹¹² serves as a canon of statutory construction across legal disciplines.

In Abitron, Hetronic, a U.S. remote control manufacturer, used Abitron, a collection of foreign companies, as a licensed foreign distributor of its remote controls.¹¹³ Abitron later claimed ownership of the manufacturer’s intellectual property rights, placed Hetronic’s trademarks on its products, and sold the products in foreign jurisdictions and the United States.¹¹⁴ Hetronic alleged infringement of its trademark under sections 32(1)(a) and 43(a)(1)(A) of the Lanham Act and sought damages for Abitron’s infringing use. The Tenth Circuit, affirming the district court, awarded damages to Hetronic for the domestic and foreign infringement.¹¹⁵ The Tenth Circuit relied on Steele and RJR Nabisco’s two-step framework to determine whether the Lanham Act extended to Abitron’s foreign conduct.¹¹⁶ The court answered RJR Nabisco’s first step affirmatively by pointing to Steele’s rebuttal of the presumption against extraterritoriality as a clear indication that the Lanham Act applies extraterritorially.¹¹⁷

The Supreme Court’s application of the RJR Nabisco framework did not stop at step one, but instead focused on step two, vacating the Tenth Circuit’s decision.¹¹⁸ The Court instructs that step two does not stop at identifying the statute’s focus but must proceed to determine if the “conduct relevant to the statute’s focus” occurred domestically or internationally.¹¹⁹ According to the Court, the Lanham Act’s focus is unauthorized “use in commerce.”¹²⁰ Thus, step two’s analysis concerned the jurisdiction in which Abitron used Hetronic’s trademark in commerce.¹²¹ Because the Tenth Circuit’s decision did not adhere to this understanding of extraterritoriality—instead, it relied on Steele and seventy years of trademark jurisprudence—the Court vacated the decision and remanded the case.¹²² While the Court took careful steps to not overturn Steele and instead dismiss it as “narrow and factbound,” it appears to no longer be good law.¹²³

The Court’s decision incorrectly determined the Lanham Act’s focus and effectively ended its extraterritorial application. Rather than “put[ting] aside” precedent,¹²⁴ Justice Sotomayor’s concurrence engages with Steele and better understands the Lanham Act’s focus:

A proper application of that framework, however, leads to a result consistent with Steele: Although there is no clear indication that the Lanham Act provisions at issue rebut the presumption against extraterritoriality at step one, a domestic application of the statute can implicate foreign conduct at step two, so long as the plaintiff proves a likelihood of consumer confusion domestically.¹²⁵

The focus of sections 32(1)(a) and 43(a)(1)(A) is “consumer confusion,” not “use in commerce.”¹²⁶ While the Act prohibits infringing use in commerce, this “is ‘merely the means by which the statute achieves its end’ of protecting consumers from confusion.”¹²⁷ The Lanham Act’s “extraterritorial coverage . . . should be gauged not so much by the locus of the activity sought to be reached . . . as by the nature of its effect on that commerce which Congress may regulate.”¹²⁸ A determination that the Lanham Act’s focus is “use in commerce” represents a lack of understanding of trademark law, as “likelihood of confusion is the keystone of trademark infringement.”¹²⁹ Instead, Justice Sotomayor’s determination is in line with the government’s position, as communicated in its amicus brief;¹³⁰ the approach taken in Steele; and tests generated in Vanity Fair, American Rice, McBee, and Hallatt that all require an “effect” on U.S. commerce, not actual U.S. conduct. In applying the Lanham Act extraterritorially, the Steele Court was not forestalled by the watches’ foreign manufacturing and sale; instead, it focused on the watches’ entrance into the United States and the ensuing domestic consumer confusion. Therefore, the presumption against extraterritoriality should be rebutted if foreign trademark infringement results in domestic consumer confusion.¹³¹

While Abitron seems to contradict U.S. trademark jurisprudence, it conforms to the international prioritization of sovereignty and independent trademark laws. In reaching Abitron’s wholly foreign conduct, the Court would seemingly violate international treaty obligations that do not only bind foreign signatories. These international agreements “only work[] if all participating states respect their obligations, including the limits on their power.”¹³² Thus, in restricting its reach, the United States is preserving its right to customization.

However, the Court’s adherence to treaty obligations ignores the economic and practical implications of a strictly territorial-bound trademark system. Infringing products sold abroad can and do have an impact on American consumers. By focusing on use in commerce, the Court is “absolv[ing] from liability” the exact defendants whose actions result in the consumer confusion the Lanham Act seeks to prevent.¹³³ Further, the impact of this decision will be felt more deeply than it would have in prior generations because of omnipresent brand awareness in a shrinking international community.

In sum, the eroded territoriality principle reveals its incompatibility with an international trademark system, and the returned commitment to the principle demands a balanced international solution to adequately protect domestic trademark owners.

B. Famous Marks Doctrine Circuit Split

The Famous Marks Doctrine diminishes the Lanham Act’s territoriality by recognizing acts outside the United States as eligible to establish domestic trademark rights. In contrast to Steele, which extended U.S. trademark rights to prevent foreign infringement, the Famous Marks Doctrine extends foreign trademark rights to prevent domestic infringement. Cases involving the doctrine fall into the second trademark territoriality problem category—like the Louis Vuitton hypothetical, in which, despite a lack of registration, the brand wants to stop the infringing use in the United States. The doctrine has roots in the Paris Convention for the Protection of Industrial Property’s “Well-Known Marks” Doctrine.¹³⁴

Enacted in 1883, the Paris Convention was the first attempt to harmonize international intellectual property laws. It is the principal international trademark treaty, with over 110 signatories. While the treaty does not provide member states with a cohesive registration or enforcement system, it provides citizens of member states with equal treatment across jurisdictions and an expectation of their level of protection.¹³⁵

The treaty’s National Treatment Principle requires member states to provide the same treatment to foreigners as they provide their citizens.¹³⁶ Rather than creating a set of international trademark principles, the Convention defers to each signatory’s set of trademark laws. Thus, signatories must provide the same protections to foreign nationals and their citizens when applying laws.¹³⁷ Additionally, the Paris Convention establishes that each country of the Union will have its own trademark legal system. Article 6 specifies that the filing and registration of trademarks “shall be determined in each country of the Union by its domestic legislation” and that a registered mark “shall be regarded as independent of marks registered in the other countries of the Union.”¹³⁸ Accordingly, the Paris Convention embeds the territoriality principle throughout its guidelines and encourages each nation’s customization of the law. However, the treaty recognizes exceptions to territoriality, including the Well-Known Marks Doctrine.

The Well-Known Marks Doctrine, under article 6bis, establishes that states shall “refuse or . . . cancel the registration, and to prohibit the use, of a trademark which constitutes a reproduction, an imitation, or a translation, liable to create confusion, of a mark considered . . . to be well known in that country.”¹³⁹ Therefore, even without registration, a well-known mark from one member state will be protected in a fellow member state. For example, if Alphabet Inc. failed to register its mark “GOOGLE” for its Internet search engine in Algeria, a contracting party of the Paris Convention, the mark “GOOGLE” would still be protected in Algeria so long as the mark was sufficiently famous in the jurisdiction. Because the “GOOGLE” mark’s fame is ubiquitous, the Well-Known Marks Doctrine would likely protect it in any Paris Convention member state, including Algeria, even if Alphabet Inc. had not registered the mark there. However, the treaty does not provide a clear process for establishing what qualifies as a well-known trademark. Instead, the treaty notes that it will be determined by a “competent authority” in the member state.¹⁴⁰ Despite this ambiguity, the doctrine serves as an exception to territoriality in the Convention, as a famous mark should be recognized in a jurisdiction absent registration.¹⁴¹

The Paris Convention’s Well-Known Marks Doctrine was designed to protect famous, unregistered marks in one member state from infringement in another; however, the United States’ determination that the treaty is non-self-executing guts the doctrine of any domestic legal effect.¹⁴² A non-self-executing treaty requires the implementation of legislation for it to be judicially enforceable;¹⁴³ however, Congress has not affirmatively included the Well-Known Marks Doctrine in the Lanham Act. Thus, courts are free to decide its domestic applicability.

Despite a lack of federal guidance, the Famous Marks Doctrine—the name attributed to the U.S. doctrine—has served as the basis for trademark infringement claims for nearly a century. The Trademark Trial and Appeal Board (“TTAB”)—a USPTO board that handles trials and appeals of trademark applications and registrations¹⁴⁴—has invoked the doctrine in its decisions.¹⁴⁵ The Board acknowledged that foreign use alone does not establish priority in the United States “unless it can be shown that the foreign party’s mark was, at the time of the adoption and first use of a similar mark by the first user in the United States, a ‘famous’ mark.”¹⁴⁶ While TTAB decisions are not binding, they are “to be accorded great weight.”¹⁴⁷

New York state courts have also relied on the doctrine in common law trademark unfair competition claims.¹⁴⁸ For example, in the 1936 New York Supreme Court case Maison Prunier v. Prunier’s Restaurant & Cafe, a New York restaurant opened with the same name, tagline, and type of food as an internationally acclaimed French restaurant.¹⁴⁹ Despite using the mark in distinct jurisdictions, the New York restaurant was enjoined because of its bad faith actions. The foreign mark’s fame partially determined this bad faith exception to territoriality.¹⁵⁰ Later, in Vaudable v. Montmartre, Inc., a different New York court enjoined a New York restaurant from using a French restaurant’s name, decor, and distinctive script style because the misappropriation was of a continuously used and famous mark.¹⁵¹ Notwithstanding Congress’s failure to adopt the Famous Marks Doctrine, this circumvention of territoriality has been occurring for almost a century. The doctrine has become increasingly necessary as commerce becomes more global.

Despite the long history of domestic use of the doctrine, lack of legislation has led to a circuit split between the Ninth and Second Circuits. The Famous Marks Doctrine was adopted by the Ninth Circuit in Grupo Gigante S.A. de C.V. v. Dallo & Co.¹⁵² but rejected in the Second Circuit in ITC Ltd. v. Punchgini, Inc.¹⁵³ In 2004, the Ninth Circuit established the enforceability of the Famous Marks Doctrine in the U.S.’s largest court circuit.¹⁵⁴ The court recognized the complexity added to the principle of first-in-time, first-in-right when “one user is first in time in one place while another is first in time in a different place.”¹⁵⁵ The territorial confines of U.S. trademark law further complicate this. In Grupo Gigante, a small, Southern Californian grocery store adopted the name of a large, well-established Mexican grocery chain.¹⁵⁶ Due to the store’s proximity to the Mexican border, the geographic market consisted of many individuals familiar with the foreign chain’s mark. Despite the store’s argument for the case’s disposal under the territoriality principle, the Ninth Circuit affirmed that the territoriality principle is not absolute and contains an exception for well-known marks, for which the Mexican mark qualified.¹⁵⁷

The Ninth Circuit’s opinion recognizes the danger of strict territoriality because “consumer confusion and fraud” would be promoted without an exception to the principle.¹⁵⁸ The court justified its reliance on the doctrine by returning to the foundational goals of trademark law, which now require heightened scrutiny in a global marketplace in which both commerce and people cross borders. Thus, the Ninth Circuit ultimately grounded its decision in public policy rather than the Well-Known Marks Doctrine or the Lanham Act.

However, in 2008, the Second Circuit declined to adopt the Famous Marks Doctrine. In ITC Ltd. v. Punchgini, Inc., a U.S. restaurant opened with an identical name, “logos, decor, staff uniforms, wood-slab menus, and red-checkered customer bibs” to that of an internationally renowned foreign restaurant.¹⁵⁹ The Second Circuit permitted the restaurant’s behavior because the foreign mark’s priority did not extend into the United States, notwithstanding its accrued fame. Unlike the Ninth Circuit, which relied on a policy rationale, the Second Circuit held that the doctrine’s sound policy “is not a sufficient ground for its judicial recognition, particularly in an area regulated by statute.”¹⁶⁰ Instead, the court noted that the Paris Convention creates no substantive rights and Congress’s deliberate choice to exclude a Famous Marks Doctrine in the Lanham Act must be honored.¹⁶¹ Thus, in the Second Circuit, a foreign mark holder is unable to protect itself through federal legislation against infringement absent U.S. registration. This decision was made despite a lower court noting that the “doctrine is particularly desirable in a world where international travel is commonplace and where the Internet and other media facilitate the rapid creation of business goodwill that transcends borders.”¹⁶²

The circuit split highlights the tension between harmonizing and customizing trademark laws. The Ninth Circuit’s—as well as the TTAB and district courts’—recognition of foreign marks’ status reveals the territoriality principle’s incompatibility with the global economy. The Famous Marks Doctrine represents a practical understanding of trademarks’ modern existence. Even in 1936, a restaurant’s famous mark could transcend borders and establish goodwill in a city thousands of miles away.¹⁶³ Present-day reality enhances the ability to establish recognition and goodwill in faraway jurisdictions.¹⁶⁴ This interconnectedness exemplifies the Ninth Circuit’s practical approach to the Paris Convention’s standards and the need for greater harmonization.

However, even the Grupo Gigante court recognizes that the Famous Marks Doctrine abrogates the territoriality principle, and its extension must be limited.¹⁶⁵ In line with the Second Circuit, the Ninth Circuit acknowledges that territoriality is “basic to trademark law,” and that the United States is “arguably required by the Paris Convention . . . to preserve the territoriality principle in some form.”¹⁶⁶ Thus, the courts are not split on the value of sovereign-drafted trademark law.

Therefore, a solution to the fragmented approach to recognizing foreign marks lies not in eliminating the territoriality principle but instead in creating a dual system that recognizes international registration and enforcement of marks alongside current domestic systems.

C. Stealth Filing Tactics

Attempts to harmonize international trademark law within a strictly territorial system have diminished the strength of the territoriality principle and led to negative externalities. Particularly, reciprocal priority, a form of international cooperation granted under the Paris Convention, has been utilized by large corporations to circumvent traditional trademark protocols. Typically, a trademark owner’s exclusive right to use a mark begins on the day of filing or first use in commerce in a sovereign jurisdiction; however, actions taken in a foreign jurisdiction qualify under a reciprocal priority system.

The Paris Convention sought to incentivize international cooperation through its transferable priority filing date. Applicants who file an ITU application in one treaty country can use the priority filing date received from that application to claim a priority date in other treaty countries, so long as the application is within six months of the original ITU application.¹⁶⁷ Unlike the Well-Known Marks Doctrine, Congress codified the Paris Convention’s transferable priority date in the Lanham Act. A claim of priority based on a foreign filing falls under section 44(d) of the Lanham Act and has become increasingly popular.¹⁶⁸ Reciprocal priority benefits applicants who apply for registration in several jurisdictions and safeguards them from third parties or bad actors who apply for identical marks following news of an initial registration.

However, Congress’s attempt to harmonize foreign and domestic trademark protection contradicts the stated commitment to the Lanham Act’s territoriality. Under Abitron’s logic, foreign acts do not warrant recognition by the Lanham Act. Yet, by granting reciprocal priority, the United States recognizes and extends domestic rights to marks in which the applicant acted in a foreign jurisdiction. Not only is this in conflict with territoriality, but it also opens the door for stealth filing tactics antithetical to trademark law’s goal of limiting consumer confusion.

Stealth filing, also referred to as submarine filing,¹⁶⁹ occurs when an applicant files an ITU application in a treaty country that lacks an accessible online application database. Thus, the applicant secures a priority filing date that will be honored in other treaty countries without the public learning about the trademark. Individuals or corporations who want to secure a mark in advance of its use in commerce but do not want to reveal this planned use may employ stealth filing tactics as a workaround to avoid disclosure. Most trademark applications, such as those filed with the USPTO, are made available to the public and include the applied-for mark, a description of the mark’s intended goods or services, and the mark’s owner.¹⁷⁰ This public database system allows third parties to object to registration as well as informs future applicants of preexisting marks that are unavailable for their use.¹⁷¹ However, it also publicly reveals an individual or company’s future business plans. For example, an ITU application for the mark “APPLE WATCH” would directly reveal the company’s next product launch to Apple’s competitors, given its descriptive name and goods and services category.¹⁷² While this might not be a problem for the average trademark applicant, high-profile individuals or businesses may want to shield the public from their confidential business plans.¹⁷³ And it has become clear that businesses do want to shield the public from their plans.¹⁷⁴

Stealth filing has become increasingly popular since 2006, with filings increasing four-fold between 2006 and 2016.¹⁷⁵ In 2006, only a couple of companies were engaging in stealth filing. By 2016, that number jumped to over forty companies, as indicated in Figure 2. Despite the increased use, stealth filings remain rare compared to overall trademark filings, with stealth filings from 2006 to 2016 representing 0.03% of total applications filed.¹⁷⁶ A 2018 World Intellectual Property Organization (“WIPO”) Working Paper identified eight jurisdictions that account for most stealth filings.¹⁷⁷ While other jurisdictions provide the infrastructure needed for stealth filings, they account for a small share of overall stealth filings. The eight jurisdictions, mostly island countries with developing economies, are Honduras, Jamaica, Liechtenstein, Mauritius, Saint Lucia, Swaziland, Tonga, and Trinidad and Tobago.¹⁷⁸ The Working Paper’s Figure A-1 (Figure 1, infra) illustrates the rise of stealth filings in these jurisdictions, with the eight countries providing a venue for 1,136 stealth trademark filings over the documented ten-year period.

Figure 1. Submarine Filings by Submarine Jurisdiction

Source: Fink et al., supra note 168, app. at i fig. A-1.

Figure 2. Number of Companies Filing Submarine Trademarks

Source: Fink et al., supra note 168, app. at i fig. A-2.

Despite benefits to well-known brands, stealth filings can harm good faith trademark applicants. Applicants who file a legitimate mark and invest time and capital into their business and intellectual property can be deprived of their trademark rights. For example, company Alpha may file for a mark in Tonga on January 1, 2023. Alpha then has until July 1, 2023, to file in the United States while retaining the January 1 priority date. Thus, if company Beta files for the same mark in the United States on February 1, 2023, and Alpha files its section 44(d) application in the United States on June 1, 2023, Beta will lose its exclusionary rights to the mark. Despite Alpha’s later application date, the USPTO will honor the priority filing date of Alpha’s Tonga application. Even if Beta, in good faith, completed a thorough search of domestic and foreign registrations and pending applications, it would still be blind to the existence of Alpha’s mark.¹⁷⁹

Section 44(d) carves out an exception to the territoriality of the Lanham Act. Not only may priority, an essential aspect of securing and protecting trademark rights, be established through actions taken in a separate jurisdiction, but a section 44(d) applicant need not ever use the mark in a foreign jurisdiction. An ITU applicant can be rejected in a foreign jurisdiction and the priority date of that rejected application can still provide the basis for priority in the United States. This not only calls into question the territoriality principle’s function, but also whether section 44(d) is rooted in the reality of any jurisdiction.

Section 44(d) serves as a temporary solution to the problems generated by a global marketplace operating within the confines of territorialism; however, this attempt to broaden the jurisdiction-based structure in the name of harmonization generates negative externalities. Thus, to lessen the tension between territoriality and the global marketplace, international solutions should work alongside the territorial system, rather than trying to fit within them. International cooperation through adoption and enforcement of a legally binding international registration and enforcement system would better effectuate the Paris Convention’s goals.

III. CURRENT INTERNATIONAL TRADEMARK OPERATION

While the territoriality principle limits the effectiveness and efficiency of globally operating brands, trademark owners have not been left alone to navigate international trademark protection. In attempts to harmonize intellectual property law, the international community has adopted treaties and systems such as the Paris Convention and the Madrid Protocol.¹⁸⁰ These treaties attempt to make foreign trademark protection more predictable and accessible. However, these agreements’ focus on procedural trademark law leaves trademark owners without cohesive substantive law to effectuate their rights.

The Madrid Protocol¹⁸¹ is an international trademark registration treaty that allows applicants to register a mark in up to 130 countries via a single application.¹⁸² The registration system is administered by WIPO, an independent United Nations agency,¹⁸³ and is a cost-effective and efficient alternative to independent registration in each country.¹⁸⁴ Further, a Protocol registration is equivalent to a bundle of national registrations, which eases the burden of management. Any renewals or changes to the registration can be done through a single step rather than independently renewing or changing each registration. As a trademark owner’s global presence increases, so too can its Madrid Protocol coverage.¹⁸⁵

The Protocol’s popularity has consistently increased since its creation, with registration tripling from 20,000 in 1998 to over 60,000 in 2018.¹⁸⁶ This increase coincides with the rise of international trade and the Internet. Additionally, the need for international protection of intellectual property is no longer limited to sophisticated, prosperous trademark owners; instead, small and medium enterprises can establish an international presence, as barriers to transnational trade and communication have been lowered. The Madrid Protocol services these less sophisticated applicants, who may lack trademark counsel, by providing a user-friendly system that allows for simple, widespread application in a single language and currency.¹⁸⁷

However, the Madrid Protocol’s benefits are limited by its operation within the confines of each signatory’s territorial trademark system. An international registration granted by WIPO is still subject to each country’s examination and national laws. Thus, territoriality remains a central aspect of the Protocol because each country independently determines whether it will provide registration and legal protection.¹⁸⁸ The Protocol provides a streamlined procedure but lacks substantive solutions to discordant trademark laws. Therefore, the Protocol’s registration process delegates decision-making to sovereigns and thus allows for retained trademark law customization but side-steps proper harmonization.

IV. ALTERNATIVES & POLICY IMPLICATIONS

The inability to constrain trademarks within sovereign borders calls for a set of harmonized trademark laws with legal force; however, a sovereign state’s ability to customize its trademark law to its citizens’ needs and the ever-changing demands of commerce weigh in favor of maintaining a territorial system. Thus, the ideal outcome is a trademark scheme that allows for harmonizing and customizing of trademark law. Introducing a legally enforceable, self-executing international system that coexists alongside sovereign nations’ statutory trademark systems, rather than attempting to squeeze harmonization into a misfitting territorial-bound system, would help achieve this balance. Viable solutions include the International Community Trademark, modeled after the European Union Trademark, or a system that mirrors that of the Uniform Domain Name Dispute Resolution Policy (“UDRP”).

A. International Community Trademark

In 1993, the European Union Intellectual Property Office (“EUIPO”) established the European Union Trademark (“EUT”),¹⁸⁹ a mark that, once filed, is enforceable throughout all twenty-eight EU member states.¹⁹⁰ Like the Madrid Protocol, a single registration establishes its legal presence in all other states; however, the EUT system is unique, as registration provides automatic legal effect and is not subject to each nation’s independent determination. The mark becomes enforceable within each member state’s EU trademark court, provided it withstands any challenge. EU member states have designated courts to enforce EUTs under EU Trademark regulations. Thus, both registration and enforcement are uniform across member states. This single registration process reduces applicants’ filing costs and gives trademark owners a sense of security not provided by the Madrid Protocol. This trademark system has proven to be widely popular. In its first year of operation, more than 40,000 applications were submitted, greatly exceeding the 15,000 to 20,000 applications the Office anticipated receiving.¹⁹¹ In 2021, the number of applications reached nearly 200,000.¹⁹² While there was concern that it would be “almost impossible” to register a new EUT due to the large number of preexisting marks, only around 16% of applied for EUTs have been opposed by prior mark holders.¹⁹³ Thus, even in a cross-border trademark system, there are plenty of marks to go around.

An EUT does not replace a nation’s trademarks but coexists with them.¹⁹⁴ Each mark has equal rank. Therefore, in the case of a conflict between an EUT and national trademark, priority controls. This approach extends to the extension of rights in new EU Member States. For example, in 2004, when ten Member States joined the EU, the rights of all previous EUTs were extended into these jurisdictions, with the limiting principle that preexisting national marks in these Member States prevailed over European Union Trademarks.¹⁹⁵

The complementary system allows applicants to register the same mark at the national level, EU level, or both.¹⁹⁶ This system caters to the differing needs of applicants, as applicants’ filing choices may depend on their size, market, or geographical presence. This system is not completely unlike the way state and federal trademarks coexist in the United States.

An international system modeled after the EUT would allow for cooperation, consistency, and confidence in international trademark operation and protection. This Note refers to this system as an International Community Trademark (“ICT”). Unlike the Madrid Protocol—the only current model that allows for widespread application—an ICT’s registration would grant automatic legal effect and would not be subject to further examination by each jurisdiction’s legal system. Instead, a neutral, international administrative body, such as WIPO, would administer the application approval process.¹⁹⁷ This would mirror the way EUTs are automatically enforceable and administered by EUIPO.

In addition to community registration, the ICT system would involve enforcement via infringement prosecution. An enforcement system would provide necessary consistency, as marks would be subject to the same regulations internationally, rather than the current ad hoc approach to trademark enforcement across different jurisdictions. Further, unlike EUTs, this system would not impose the burden of enforcement on nations’ preexisting court systems. Instead, ICTs would be enforced by a nongovernmental neutral body, administering decisions in a manner similar to the TTAB’s operation in the United States or the Internet Corporation for Assigned Names and Numbers (“ICANN”) in domain name disputes, as discussed below.

Most importantly, ICTs would not replace each nation’s trademark system but would coexist with it. The dual system circumvents sovereignty issues by allowing for a continuation of territorial-bound marks, while introducing borderless marks. Thus, the ICT would serve as a balanced response to the growing international demands of trademarks by allowing for actual harmonization and continued customization.

An example helps demonstrate the practical manner in which these trademark systems would coexist. Suppose a Canadian company has trademarked “ROOTS” for outerwear clothing at the national level. Upon establishment of an ICT system, the Canadian company would have an opportunity to register an ICT for “ROOTS”; however, if the company did not register the ICT and an American company wanted to register “ROOTS” for clothing as an ICT, the Canadian mark would serve as an obstacle for protection of the new mark. In this case, the Canadian brand could prevent use of the American “ROOTS” ICT in its territory. This structure resembles the treatment of state and federal trademarks in the United States. Federal registration, like an ICT registration, does not override registration at the state level. Instead, a state mark can continue to be used within the geographical confines of that jurisdiction even after the registration of an identical third-party mark at the federal level.

However, the ICT has disadvantages inapplicable to the Madrid Protocol. The Community Mark system creates a high-risk, high-reward dynamic. An ICT modeled after the EUT would grant a “single registration covering several countries, [whereas] a Madrid registration represents a mechanism for obtaining a bundle of national trademark rights.”¹⁹⁸ Under the Madrid Protocol, a rejected application does not prevent an applicant from pursuing registration via the same application in other jurisdictions. However, rejecting an ICT would result in the refusal of registration in all member states. Applicants could receive protection at the national level under the complementary system; however, the goal of creating a consistent, efficient registration system would be frustrated.

Additionally, converging procedural and substantive rights through an ICT implicates international comity. While a neutral body would create and execute the system’s substantive law, harmonizing disparate trademark law presents a significant challenge. This challenge is not insurmountable, as evidenced by the existence of the EUT; however, bridging varying international trademark laws presents challenges beyond those involved in regional harmonization. The manner used to establish trademark rights represents a basic rift in sovereigns’ substantive laws. Would ICT rights be established on a first-to-file basis, on which the United States operates a modified version, or on a first-to-use basis, the way most sovereign jurisdictions grant priority? In considering equitable answers to these questions, it is critical to recognize that “United States law governs domestically but does not rule the world,” as is true for foreign superpowers.¹⁹⁹

Thus, an ICT has the potential to provide more efficient, comprehensive protection to trademark owners but would require heightened applicant attention and care.

B. UDRP Cybersquatting Framework

In 2000, ICANN adopted the UDRP to address the international issue of cybersquatting.²⁰⁰ Cybersquatting involves the bad faith use of a domain name. The UDRP provides four examples of bad faith use of a domain name: (1) holding the domain name for ransom from the trademark owner; (2) registering the domain name to block the trademark owner; (3) registering the domain name to disrupt the business of a competitor; and (4) registering the domain name to confuse users into coming to the website.²⁰¹ ICANN requires global top-level registrars—companies registering domain names and providing IP addresses—to include the UDRP in all domain name contracts. The policy creates a mandatory dispute resolution process for “abusive registrations,” such as cybersquatting.²⁰² Panelists, who consist of a geographically diverse set of experienced attorneys, oversee the process. These panelists serve ICANN-approved dispute resolution providers. Currently, ICANN has six approved providers, including the Asian Domain Name Dispute Resolution Centre and the Arab Center for Dispute Resolution.²⁰³ The providers have geographically diverse bases but operate worldwide with distinct specialties.

The UDRP, which “represents the first example of a truly global body of intellectual property enforcement,”²⁰⁴ has served as a successful dispute resolution system over the past twenty-three years. This system is the byproduct of “global consensus and contract, not . . . governmental enactment and coercion.”²⁰⁵ The UDRP is an effective alternative to traditional jurisdictional legal disputes, and its role in the international community continues to grow. From January 2013 to December 2020, UDRP complaints grew by 6% on average, totaling 38,349 for the entire period.²⁰⁶ Further, these disputes, from filing to decision, took, on average, forty-six days.²⁰⁷ The true expediency of UDRP disputes is revealed when considering “civil cases in the U.S. district courts have a median length of 27 months from filing to trial, and close to 10% of cases have been pending for over three years.”²⁰⁸

The UDRP’s informal approach to dispute resolution would allow for a more expedient and convenient process for international parties. With 18.1 million trademarks filed in a single year, the sheer magnitude of potential disputes requires an international system that can handle the demand.²⁰⁹ By relying on panels—free from the hierarchical and formalistic standards of courts—to make decisions, the UDRP model increases efficiency in dispute resolution. Parties to international disputes also hail from different parts of the globe. Therefore, by eliminating in-person hearings, parties would not be burdened by international travel and panelists could make quicker decisions.²¹⁰ However, panelists would still operate in a way consistent with typical legal proceedings. While the system benefits from efficiency and informal proceedings, there are high standards for the quality of decision-makers and the decision-making process itself. Per the UDRP model, panelists would be “selected on the basis of their well-established reputation for their impartiality, sound judgment and experience as decision-makers, as well as their substantive experience.”²¹¹ Additionally, these proceedings aim to achieve “a high degree of predictability and consistency . . . through consensus or precedent.”²¹² Thus, this system would more likely achieve the Paris Convention’s goal of predictability—currently accomplished via the National Treatment principle—than the current model. Rather than requiring parties to understand the changing ways sovereigns interpret and enforce trademark law, parties could look to the decision-making of a single enforcement body.

A UDRP-like enforcement system would need to work in tandem with a legally enforceable, self-executing international system. An international enforcement system is only effective if it has a uniform set of laws to enforce. Therefore, just as global consensus would be critical in the creation of an ICT, the same would be required for establishing an international dispute resolution system. The UDRP’s structure establishes a replicable framework for adjudicating a more diverse set of trademark disputes.

However, trademark infringement disputes are typically more complex than those involving cybersquatting and thus require a more nuanced decision-making process. The rapid turn-around time for UDRP disputes is a byproduct of their limited scope and lack of in-person hearings. UDRP Panels make decisions upon review of filed Complaints and Responses from the relevant parties.²¹³ This is adequate for cybersquatting cases because the UDRP has outlined a strict set of criteria that constitute an abusive registration; however, multi-jurisdictional trademark infringement cases set forth issues that are not present in cybersquatting disputes. For example, “Nike” may exist in different countries to identify shoes and athletic apparel but there can only be one nike.com. Thus, while there can only be one domain name, trademarks may coexist in different sovereign jurisdictions, presenting nuanced issues that may not be able to be settled within a forty-six-day window.

A comprehensive international trademark dispute resolution system would undoubtably involve unprecedented challenges, yet the UDRP serves as a successful, small-scale test case for what this system could look like in practice.

CONCLUSION

Territoriality has been deemed a basic tenant of domestic and international trademark law; however, it is incompatible with a globalized, modern world. Practically, the principle no longer exists in its originally intended form due to U.S. courts’ circumvention and international treaties’ attempts at harmonization. The system has needed an update for some time, yet the need is more pressing than ever before due to the Supreme Court’s recent decision to effectively eradicate the Lanham Act’s extraterritorial application. While domestic trademark owners may still effectuate their rights at the national level, through foreign registration, or through the Madrid Protocol, this siloed approach to trademark rights is inadequate. It has harmed both well-known brands, whose goodwill is traded on in bad faith, and emerging brands, whose good-faith, responsible brand establishment can be dismantled due to stealth filing. However, this Note does not advocate for a return to the universality principle. Instead, it suggests a dual system.

It is crucial that this updated system does not attempt to fit a square peg into a round hole in the way previous agreements have attempted to fit harmonization into a jurisdictional-bound system. A dual system would preserve the benefits of a territorial-bound trademark system—customization of trademark laws and international comity—and cater to the demands of a globalized economy in which brands transcend borders. A legally enforceable, self-executing international system would provide trademark owners with a cost-effective, efficient, and reliable approach to trademark protection. This especially benefits small to medium-sized enterprises that are entering multinational markets more easily and may not be equipped with the resources to navigate foreign intellectual property protection. Additionally, international enforcement reduces the unpredictability of sovereign decision-making and ensures that agreed-upon principles become a reality and are not shrugged-off as non-self-executing. The ICT and UDRP are not exhaustive solutions to the problem; rather, they provide a useful starting point to guide further discussion of the future of international trademark operation. While scaling either system presents unique challenges, both systems’ current success provides a glimmer of hope.

Finally, globalization is not going away, and the world is more interconnected than ever before. The growth of information technology and communication platforms as well as the decrease in business scaling and trade barriers have facilitated the ubiquity of trademarks in daily life.²¹⁴ While the explosive growth of global trade may slow in the coming years, the need for international trademark protection will not wane.²¹⁵ Therefore, regardless of the form of international solutions to the trademark problems, the Lanham Act must find a way to better reconcile harmonization and customization.

98 S. Cal. L. Rev. 165

Download

Gabriella F. Franco*

* Articles Editor, Southern California Law Review, Volume 98; J.D. Candidate, University of Southern California Gould School of Law, 2025; B.A. Philosophy, Politics, and Economics, University of Michigan, 2021. A special thank you to Professor Scott Altman and Professor Jonathan Barnett for their invaluable guidance, my friends and family for their unwavering support, and the editors of the Southern California Law Review for their hard work and thoughtful suggestions.

Evidence that the President Is an “Officer of the United States” for Purposes of Section 3 of the Fourteenth Amendment

March 2025March 2025James A. Heilpern* & Michael T. Worley†

AUTHORS’ NOTE

This Article was written during the weeks before certiorari was sought in Trump v. Anderson in the aftermath of the Colorado Supreme Court’s decision to remove former President Donald Trump from the 2024 Republican primary ballot. An initial draft was uploaded to SSRN on Jan. 1, 2024, and a revised draft incorporating constructive feedback we received from other scholars was uploaded on Jan. 8, 2024. Because those versions of the Article featured prominently in the Respondents’ merits briefing¹ and were discussed at length by popular media sources² and scholarly circles,³ we have decided to make no substantive changes to the text of the Article, treating it instead as an artifact of a specific and important historical period. We have made grammatical and syntactic revisions, and updated some of the footnotes in order to reference final versions of articles which, at the time, were only available in draft form on SSRN.

INTRODUCTION

In 1868, three years after the conclusion of the Civil War and the assassination of Abraham Lincoln, the Fourteenth Amendment was ratified and became part of the United States Constitution. The Amendment officially overturned the notorious Dred Scott decision and was designed to grant citizenship and ensure equal protection under the law for recently freed slaves. But Section 3 of the Amendment also contained a provision that limited the ability of a small class of former Confederates—those who had previously taken oaths to support the U.S. Constitution—from holding public office in the future:

No person shall be a Senator or Representative in Congress, or elector of President and Vice-President, or hold any office, civil or military, under the United States, or under any State, who, having previously taken an oath, as a member of Congress, or as an officer of the United States, or as a member of any State legislature, or as an executive or judicial officer of any State, to support the Constitution of the United States, shall have engaged in insurrection or rebellion against the same, or given aid or comfort to the enemies thereof. But Congress may by a vote of two-thirds of each House, remove such disability.⁴

Six months prior to the writing of this Article, William Baude and Michael Stokes Paulsen made headlines by publishing an article on SSRN, The Sweep and Force of Section Three, in which they argued that Donald Trump’s actions on January 6, 2021 qualified as an insurrection and that Section 3 therefore disqualified him from being elected President again.⁵ At the time, Trump was the front-runner for the Republican nomination for President in 2024. Baude and Paulsen’s paper inspired lawsuits in twenty-one states, seeking to remove President Trump from the upcoming primary ballots.

Most of the media attention has focused on whether Trump actually “engaged in insurrection.” This paper focuses on a far less titillating question. In order for Section 3 to apply to Donald Trump, he must have been an “officer of the United States” prior to committing the alleged insurrection. Baude and Paulsen argue that, as President of the United States, Trump was an officer of the United States.⁶ In making that argument, Baude and Paulsen disagreed with an earlier piece by Josh Blackman and Seth Tillman, Is the President an “Officer of the United States” for Purposes of Section 3 of the Fourteenth Amendment?⁷ Blackman and Tillman examined the original 1788 Constitution and concluded that the founding generation understood that the President was not an “officer of the United States.”⁸ Their analysis focused on the text of the constitution and subsequent sources. Based on this conclusion, Blackman and Tillman “contend that the phrase ‘officer of the United States’ has the same meaning in Section 3 as it does in the Constitution of 1788.”⁹ This implies that “the elected President is not an ‘officer of the United States.’ ”¹⁰

On December 19, 2023, the Colorado Supreme Court concluded that Donald Trump is ineligible to be on the Colorado Republican primary ballot for President because he is disqualified under Section 3.¹¹ The opinion reversed a trial court judge who had found Trump did commit insurrection but that Section 3 did not apply because Presidents are not officers of the United States.¹² Rejecting Trump’s contention that “officer of the United States” was a term of art, the state supreme court concluded that “[i]f members of the Thirty-Ninth Congress and their contemporaries all used the term ‘officer’ according to its ordinary meaning to refer to the President, we presume this is the same meaning the drafters intended it to have in Section Three.”¹³ The court cited examples of the contemporaries of the Fourteenth Amendment referring to the President as an officer,¹⁴ but only cited limited evidence about the use of the full term “officer of the United States.”¹⁵ Baude and Paulsen similarly cite limited historical evidence, spending under ten pages on this issue, which they spend discussing logical reasoning more than historical evidence.¹⁶

This Article attempts to fill the gap in historical evidence and provide a more detailed theoretical foundation. Part I reviews Blackman and Tillman’s article and other arguments made in the Colorado litigation, including the argument that the President is not an officer because he is “elected” not “appointed,” and the argument that he is not an officer because he does not take an oath to “support” the Constitution as required by Article VI, but instead takes the Article II oath to “preserve, protect, and defend” the Constitution.¹⁷ In Part II, we respond to these arguments as a textual matter, ultimately concluding that the President was an “officer of the United States” at the time of the Founding. Here, we (1) provide corpus linguistic evidence that the full phrase “officer of the United States” was not a term of art in contradiction to the explicit arguments made by President Trump at the Colorado Supreme Court; (2) demonstrate that at the time the Constitution was ratified, the words “appoint” and “elect” were largely used interchangeably; (3) provide founding era cites, including to a 1799 Act regarding the post office, that either explicitly identify the President as an “officer of the United States” or otherwise indicate that he is such an officer; and (4) present evidence that many state officers prior to the Civil War took an oath similar to the President’s and were still unambiguously covered by Section 3 despite not taking an oath that follows the precise language of Article VI of the Constitution. In Part III, we briefly discuss the phenomenon of “linguistic drift.” In Part IV and V, we then turn to the meaning of the phrase at the time of the ratification of the Fourteenth Amendment. In Part IV, we discuss and confirm that historical records including the text, legislative history, and ratification debates of the Fourteenth Amendment, the legislative history of the Fifteenth Amendment, and popular sources such as contemporary newspapers demonstrate that elected officials were often referred to as officers, including “officers of the United States.”¹⁸ Part V then discusses specific evidence that the President is not just an officer, but is an “officer of the United States” as contemporaries of the Fourteenth Amendment would have understood that term. The most probative evidence is perhaps proclamations from President Andrew Johnson—the President at the time the Fourteenth Amendment was ratified—explicitly referring to himself as either the “chief executive officer of the United States” or “chief civil executive officer of the United States.” Other evidence comes from numerous texts, including legislative history, Johnson’s impeachment trial, and newspapers.¹⁹ Part VI reexamines case law that Blackman and Tillman rely on. We then conclude.

In his brief to the Colorado Supreme Court, President Trump argued that “not one authority holds that the President is an officer of the United States[:] no case, no statute, no record of Congressional debate, no common usage, no attorney general opinion. Nothing.”²⁰ We have done our best to collect and catalog extensive evidence to the contrary.

Our conclusion is simple: the President was an officer of the United States as originally understood both at the Founding and the ratification of the Fourteenth Amendment. Numerous sources confirm that “officer of the United States” was not a term of art, which by itself settles the matter. Regardless, founding-era sources also refer to the President as an officer of the United States. This includes the Postal Act of 1799, which lists the President with officers of the United States. Additionally, there is strong probative evidence that in 1868 the President was considered an officer of the United States.

I. SUMMARY OF PAST SCHOLARSHIP AND ITS USE IN THE COLORADO PROCEEDINGS

This Section attempts to fairly familiarize the reader with Blackman and Tillman’s points and walk through how their article informed the proceedings in the Colorado case regarding President Trump’s eligibility to appear on the 2024 Colorado Republican Primary ballot.

A. Summary of the Blackman-Tillman Interpretation

Blackman and Tillman argue that Section 3 of the Fourteenth Amendment cannot bar President Trump from holding future office because the only office he has held is that of President, and the President is not an officer of the United States. They compare the text of the Fourteenth Amendment to the text of the original Constitution and infer that (1) “Section 3’s ‘officer of the United States’-language was imported from the Oath or Affirmation Clause,”²¹ and (2) “[i]n 1788, the President was not an ‘officer of the United States.’ ”²² They also tentatively state a third conclusion: “[W]e do not think linguistic drift occurred with respect to the phrase ‘officer of the United States’ ” between the founding in 1788 and the enactment of the Fourteenth Amendment in 1868.²³

Blackman and Tillman first look at the Constitution’s text, specifically the use of the term “officers of the United States” in the Appointments Clause, the Commission Clause, the Impeachment Clause, and the Oath or Affirmation Clause. They claim that none of these clauses suggest the existence of officers who are elected, only officers who are appointed.

First, Blackman and Tillman emphasize that the Appointments Clause states that the President shall appoint “Ambassadors, other public Ministers and Consuls, Judges of the [S]upreme Court, and all other Officers of the United States.”²⁴ Because the President does not appoint himself, they reason that he cannot be an officer of the United States.²⁵

Next, they rely on the Impeachment Clause’s reference to “[t]he President, Vice President and all civil Officers of the United States.”²⁶ From this language, they conclude that

the president and vice president’s [express] enumeration in the Impeachment Clause in addition to ‘all civil Officers of the United States’ shows that the president and vice president are not deemed ‘officers of the United States’ themselves. Otherwise, the Framers would have stated that ‘all other civil officers’ were subject to impeachment.²⁷

Third, they cite the Oath or Affirmation Clause, which requires the “Senators and Representatives before mentioned, and the Members of the several State Legislatures, and all executive and judicial Officers, both of the United States and of the several States” to take an oath to “support the Constitution.”²⁸ But because the President takes a different oath specified at the end of Article II, they conclude he must not be an officer of the United States.²⁹

Finally, they note that Article II, Section 3, states that the President “shall Commission all the Officers of the United States.”³⁰ Here, they argue, “All means all.” This structure explains why appointed executive-branch and judicial-branch officers receive commissions, but there is no record of any elected official, whether a president, vice president or a member of Congress, ever receiving a [presidential] commission.”³¹

Based on their analysis, Blackman and Tillman claim that “Section 3’s ‘officer of the United States’-language was imported from the Oath or Affirmation Clause.”³² They make this claim because both clauses “reference the same four categories of office holders who swore an oath to support the Constitution: [1] Senators and Representatives, [2] members of the state legislatures, [3] executive and judicial officers of the United States, and [4] executive and judicial officers of the states.”³³ Based on the parallel structure of these clauses, they conclude that because the President is not mentioned in the Oath or Affirmation Clause, the parallel language of Section 3 excludes him.³⁴

Blackman and Tillman next argue that “[i]n 1788, the President was not an ‘officer of the United States.’ ”³⁵ To support this conclusion, they first state that “[e]lected officials like the president are not ‘Officers of the United States.’ ”³⁶ Second, they rely on the drafting process surrounding the original Constitution:

For example, in the Succession Clause, the phrase “officer of the United States” was changed to “officer.” In the Impeachment Clause, the phrase “[President, Vice President,] and other Civil officers of the U.S.” was changed to “President, Vice President, and Civil Officers of the U.S.” And in its final form, the Impeachment Clause became: “President, Vice President, and all civil Officers of the United States.” The Framers changed the word that preceded “Civil Officers of the United States” from “other” to “all.”³⁷

From these changes, they conclude:

This and other similar alterations to the draft constitution’s “office”- and “officer”-language were significant. First, these revisions show that this language was not modified indiscriminately. The Framers paid careful attention to the words they chose. Second, the use of “other” in the draft constitution shows that at a preliminary stage, the Framers used language affirmatively stating that the President and Vice President were “Officers of the United States.” But the draft constitution’s use of “other” was, in fact, rejected in favor of “all.” The better inference, arising in connection with the actual Constitution of 1788, is that the President and Vice President are not “Officers of the United States.”³⁸

Finally, Blackman and Tillman tentatively conclude: “[W]e do not think linguistic drift occurred with respect to the phrase ‘officer of the United States’ ” between the founding in 1788 and the enactment of the Fourteenth Amendment in 1868.³⁹ They cabin this conclusion carefully, noting repeatedly that this conclusion was based on the lack of “direct, clear, or compelling evidence.”⁴⁰ They cite two cases—United States v. Mouat (1888) and United States v. Hartwell (1867)—as evidence.⁴¹ They also cite to statements from two individuals who viewed the President as not an officer of the United States:

In 1876, the House of Representatives impeached Secretary of War William Belknap. During the trial, Senator Newton Booth from California observed, “the President is not an officer of the United States.” Instead, Booth stated, the President is “part of the Government.”

Two years later, David McKnight wrote an influential treatise on the American electoral system. He reached a similar conclusion. McKnight wrote that “[i]t is obvious that . . . the President is not regarded as ‘an officer of, or under, the United States,’ but as one branch of ‘the Government.’ ”⁴²

However, Blackman and Tillman admit that they “do not suggest that there is no counter-authority” but ask for evidence “as probative as” their own before they accept the proposition of linguistic drift.⁴³ We do not suggest that, if Blackman and Tillman were right about these core points, they would be wrong in the conclusion they draw: that a President who had not otherwise served as an officer of the United States would not be subject to the Fourteenth Amendment.⁴⁴ Instead, we believe they are wrong on these points.⁴⁵

Blackman and Tillman have written a more recent article, Sweeping and Forcing the President into Section 3, where they respond directly to Baude and Paulsen’s criticism at length.⁴⁶ Much of their substantive discussion is indistinguishable from their earlier discussion as they focus their attention on their existing scholarship on the term “officer of the United States.” As they did in their original article, they cite Senator Booth and David McKnight.⁴⁷ They discuss the other Clauses their first paper discussed and spend time critiquing Baude and Paulsen’s discussion of this issue.⁴⁸

But they continue to convey a commendable level of humility about their position: “Maybe Booth and McKnight were right, or maybe they were wrong. No doubt, there are other competing authorities.”⁴⁹ “We do not say this question has an obvious answer. Rather, we say it does not have an obvious one. If so, ambiguity leans against extending disqualifications.”⁵⁰ In discussing if there was linguistic drift regarding “Office . . . under the United States,” they say, “[w]e would prefer to add to the body of scholarship and be correct, rather than overreach and be wrong.”⁵¹

In their response to Baude and Paulsen, Blackman and Tillman point to the Oath or Affirmation Clause as the source of the meaning of terms in Section 3, suggesting that “[i]f we are correct, it illustrates that constitutional draftspersons, in 1787–1788 and 1866–1868, closely adhered to parliamentary drafting conventions,” and critique the legal academy for not understanding those conventions.⁵² They also repeat an argument made in the Colorado litigation: that because the presidential oath says “preserve, protect and defend” and the Article VI oath says “support,” Section 3 includes the latter and excludes the former.⁵³ Blackman and Tillman also cite contemporary opinions by attorneys general and others that link Article VI and Section 3.⁵⁴

B. Reliance by Lower Colorado Court

In late 2022, Donald Trump announced he would run for President again.⁵⁵ In September 2023, five Colorado residents sued Colorado’s Secretary of State, arguing that Trump was ineligible to be on the ballot in Colorado because he had violated Section 3.⁵⁶ Obviously, Trump intervened.⁵⁷ Trump cited Blackman and Tillman in support of the proposition that the President was not an officer of the United States.⁵⁸ Trump quoted the examples Blackman and Tillman gave: Senator Booth and treatise writer David McKnight stating that the President was not an officer of the United States.⁵⁹ He also cited recent Supreme Court precedent, including the statement by Chief Justice Roberts that “[t]he people do not vote for the ‘Officers of the United States.’ ”⁶⁰ Finally, Trump cited the Appointments Clause, the Impeachment Clause, the Commission Clause, and the Oath or Affirmation Clause, closely tracking Blackman and Tillman in some detail as to why each Clause supports the idea that the President is not an officer of the United States.⁶¹

The trial court ultimately ruled against Trump on every dispositive issue except whether the office of the President was an officer of the United States.⁶² While the court did not cite Blackman and Tillman, it did cite to the same four clauses that Blackman and Tillman and Trump’s briefing rely on—the Appointments Clause, the Commission Clause, the Impeachment Clause, and the Oath or Affirmation Clause.⁶³ The trial court also noted that the President takes a different oath than Article VI officers, suggesting that his oath was not covered by Section 3.⁶⁴

However, part of the reason for the trial court’s decision was the implications of an alternative conclusion. The court stated:

To be clear, part of the Court’s decision is its reluctance to embrace an interpretation which would disqualify a presidential candidate without a clear, unmistakable indication that such is the intent of Section Three. As Attorney General Stanbery again noted when construing the Reconstruction Acts, “those who are expressly brought within its operation cannot be saved from its operation. Where, from the generality of terms of description, or for any other reason, a reasonable doubt arises, that doubt is to be resolved against the operation of the law and in favor of the voter.” The Reconstruction Acts, 12 U.S. Op. Att’y Gen. 141, 160 (1867) (emphasis added). Here, the record demonstrates an appreciable amount of tension between the competing interpretations, and a lack of definitive guidance in the text or historical sources.⁶⁵

C. Rejection by Colorado Supreme Court

Trump’s core argument at the Colorado Supreme Court was that the catch-all phrase “officer of the United States” excludes the President as “no case, no statute, no record of Congressional debate, no common usage, no attorney general opinion” supports the conclusion that the President is an officer and “by contrast, three Constitutional provisions—the Appointments Clause, Impeachment Clause, and Commissions Clause in Article II—all distinguish the President from ‘officers of the United States.’ And ‘officers of the United States’ in Article VI take an oath different from the Presidential oath in Article II.”⁶⁶ Trump also argued that the presidential oath does not qualify as an oath under Section 3’s requirements for disqualification.⁶⁷

The Colorado Supreme Court disagreed. The court concluded: (1) “[T]he normal and ordinary usage of the term ‘officer of the United States’ includes the President”;⁶⁸ and (2) “Section Three’s drafters and their contemporaries understood the President as an officer of the United States.”⁶⁹

Trump was incorrect to argue that “ ‘officer of the United States’ . . . is a constitutional term of art” because the court “perceive[d] no persuasive contemporary evidence demonstrating some other, technical term-of-art meaning.”⁷⁰ In other words, if “officer of the United States” was a term of art, someone would have said so.

Attorney General Stanbery’s opinions on the subject suggest that the term “officer of the United States” was to be broadly understood.⁷¹

However, the Colorado Supreme Court did not cite much historical evidence referring to the President by the term “officer of the United States.” Many of the examples referred to the President simply as an “officer.”

While three Justices dissented from the holding of the majority, none of them argued that the President was not an officer of the United States.⁷²

D. The Colorado GOP’s Petition for Certiorari

The Colorado Republican Party filed for certiorari, represented by Jay Sekulow, a long-time ally of Trump who has argued fourteen or more cases at the Supreme Court. The leading question presented concerns whether the President is an officer of the United States.⁷³ Their brief for certiorari reads like a merits brief.

In arguing that the President is not an officer of the United States, the brief makes the following core points: First, the President is not an officer because officers are commissioned by the President under the Commissions Clause, not elected.⁷⁴ Second, “officers of the United States” is a term of art that is only used in three places in the Constitution: Section 3 of the Fourteenth Amendment, the Commissions Clause, and the Appointments Clause.⁷⁵ The Colorado Republican Party argues, “The Constitution uses a distinct, specific term, ‘officer of the United States.’ Generic references to the term officer do not overcome the more specific meaning evidenced by the constitutional language.”⁷⁶ Third, they rely on the same sources of Senator Booth and David McKnight that were previously explained.⁷⁷ Fourth, they make a distinction between the presidential oath and the Article VI oath, relying on Attorney General Stanbery’s discussion of the Article VI oath.⁷⁸ Finally, they posit that this exclusion of the President from Section 3 makes sense as a policy matter:

Every Senator or Representative represents a geographic area where sympathy for insurrection was (at the time of the post-Civil War era) a real and legitimate concern. Lower federal officers, meanwhile, are not elected and thus do not face national electoral scrutiny. Only the President (and Vice-President) face nationwide electoral accountability. And if an electoral majority of the voters determine that they want a certain individual as Chief Executive, regardless of alleged or even actual past transgressions, that is their national choice under the Constitution.⁷⁹

(This last argument appears poorly reasoned; if a President had previously served as a senator, governor, or general, as many Presidents have, it would not matter that “an electoral majority of the voters determine that they want [that former President] as Chief Executive” if that President committed insurrection; they would be ineligible to run for a second term. This argument does not explain why the Fourteenth Amendment’s drafters would have wanted to exclude only Presidents who had never held offices such as Senator, Governor, or General.)

II. THE PHRASE “OFFICER OF THE UNITED STATES” INCLUDED THE PRESIDENT AT THE TIME OF THE FOUNDING

Despite Blackman and Tillman’s articles being impressive examples of careful scholarship, there are at least four reasons we disagree with the conclusion that the original public meaning of “officers of the United States” did not include the President or Vice President: (1) corpus linguistics evidence—including our own past research—demonstrates that at the time of the Founding, “officer of the United States” was not a term of art but instead referred to any federal official; (2) the assumption that there was a rigid distinction between “elections” and “appointments” is at odds with the historical record, which shows that the words were instead used interchangeably; (3) the text of the Constitution specifically identifies the Presidency as an “office”; and (4) additional contextual considerations complicate Blackman and Tillman’s otherwise straightforward textual analysis. While this topic merits an article of its own, we will address each reason briefly below.

A. Corpus Linguistics Evidence Supports that the President Is an Officer of the United States

First, many who argue that the President is not an officer of the United States—including President Trump and Steven Calabresi—do so based on the assumption that the phrase “officer of the United States” is a term of art⁸⁰—an assumption that past research refutes. In 2018, we wrote an amicus brief on behalf of fifteen scholars of corpus linguistics which was submitted to the U.S. Supreme Court in the case Lucia v. SEC.⁸¹ As part of that brief, a corpus linguistic analysis of the phrase was performed, from which we drew the following conclusions:

The phrase “Officer(s) of the United States” appears in [BYU’s Corpus of Founding Era American English (“COFEA”)] just 109 times between 1789 and 1799, with just over a third of those being direct quotations of the Constitution. This is a tiny minority of the 5,353 times the word “officer” appears in the database overall during this same period―even though 59.8% of the time the word “officer” appears in COFEA it is clearly referencing a federal [official].

While the relative obscurity of the longer phrase does not prove that it was not a legal term of art at the time of the Founding, we perceive no specialized meaning attached to its use [to suggest that it was]. Instead, the appellation was often used simply to clarify that the agent was in the employ of the federal government, as opposed to a private actor or employee of a state or territory.

For example, in a letter to George Washington, General Arthur St. Clair expressed concern that the Attorney General of the new Ohio territory “would be an Officer of the Territory only, whereas he should be an Officer of the United States.” Likewise, Alexander Hamilton wrote to New York merchant William Seton, requesting he purchase public debt on behalf of the federal government since the government had yet to “employ some officer of the United States” for the task.⁸²

We did not see then—and do not see now—any evidence to suggest that the term excluded the President or was limited to some special subclass of federal officials. To the contrary, it applied broadly to all government officials—“civil and military”—exercising any non-trivial federal authority. For instance, in his Eighth Annual Address to Congress at the end of 1796, George Washington called for “legislative revision” of “[t]he compensation to the officers of the United States,” particularly “in respect to the most important stations.”⁸³ Congress responded a little over one year later, raising the salaries of sundry government officials, starting with “the President and Vice President of the United States.”⁸⁴ The fact that Congress did not use the phrase “officers of the United States” in this appropriations bill, but instead referred generally to “officers,” “offices,” and “persons employed,” even when referring to positions such as the Secretary of State, Attorney General, Secretary of the Treasury, Secretary of War, Chief Justice, and Consuls—positions that neither Blackman and Tillman nor President Trump would dispute are “officers of the United States”—further demonstrates that the larger phrase was not considered a term of art.

In fact, a corpus search of BYU’s Corpus of Early Statutes at Large⁸⁵—which the authors of the Lucia brief created and which contains all of the Statutes at Large from the first five Congresses—reveals that Congress almost never used the phrase “officer(s) of the United States” during this time period, despite being an era when Congress was constantly exercising its power to “establish[] by [l]aw”⁸⁶ such positions within the new government. In its first decade, Congress used the phrase just thirteen times, while using the word officer or officers 1,481 times and office or offices 630 times. This would be baffling if “officer of the United States” was a legal term of art but makes perfect sense if the phrase merely designated a federal official—after all it was the Congress of the United States creating the positions, what other type of office would we expect? One for Virginia? Japan? In the absence of clear textual clues to the contrary—such as perhaps territorial officers—the default assumption should be that all of such positions created by Congress are officers of the United States.

In addition, of the thirteen times the full phrase appears, one—a postal bill specifying which “officers of the United States” should be granted a franking privilege—specifically listed both the President and Vice President as officers of the United States.⁸⁷ The Postal Act of 1799 stated:

Sec. 17. And be it further enacted, That letters and packets to and from the following officers of the United States, shall be received and conveyed by post, free of postage. Each postmaster . . . ; each member of the Senate and House of Representatives of the Congress of the of the United States; the Secretary of the Senate and Clerk of the House of Representatives . . . ; the President of the United States; Vice President; the Secretary of the Treasury; Comptroller; Auditor; Register; Treasurer; Commissioner of the Revenue . . . .⁸⁸

The conclusion that the phrase “officer of the United States” was not a term of art at the time of the Founding is further buttressed by the research of Jennifer Mascott, who used aspects of corpus linguistics to demonstrate that the phrase was in use prior to the creation of the Constitution.⁸⁹ Using a corpus of 340,000 issues of early American newspapers, she found twenty uses of the phrase “prior to the signing of the Constitution on September 17, 1789.”⁹⁰ The first reference was in 1780, describing Benedict Arnold as a “general officer of the United States.”⁹¹ It appeared again in 1783 referring simply to continental officers. Other uses included “Judicial Officers of the United States” and “commissaries and other officers of the United States” who gave out certifications of debt under the Constitution.⁹²

Mascott also performed a corpus analysis of the Journals of the Continental Congress, “a highly relevant source for identifying the well-understood meaning of legally relevant terms and phrases in the time period just prior to the drafting and ratification of the Constitution.”⁹³ The Journals contain forty-one references to “officer(s) of the United States.”⁹⁴ Often the phrase was “just another way to describe continental military officers or identify national- as opposed to state-level, officers.”⁹⁵ For example, one letter distinguished between the time a military officer served as an “officer of the United States” and the time he served as a captain for his State.⁹⁶

From both a legal and linguistic perspective, we think the importance of the full phrase “officer of the United States” not being a term of art at the time of the Founding cannot be overstated. It narrows the scope of the linguistic inquiry: the meaning of the word “officer” standing alone becomes the relevant question, with “of the United States” simply being a prepositional phrase functioning as an adjective. A comment by Alexander Hamilton in Federalist 67 supports this conclusion:

The second clause of the second section of the second article empowers the President of the United States “to nominate, and by and with the advice and consent of the Senate, to appoint ambassadors, other public ministers and consuls, judges of the Supreme Court, and all other OFFICERS of United States whose appointments are NOT in the Constitution OTHERWISE PROVIDED FOR, and WHICH SHALL BE ESTABLISHED BY LAW.”⁹⁷

While we will return to the substance of Hamilton’s statement later, what is important for present purposes is his interesting use of capitalization for emphasis. If he had understood the term “officers of the United States” to be a term of art, he would have written “OFFICERS OF THE UNITED STATES.” But he did not. He focused on just the word “OFFICERS.”

The addition of a modifier—whether it is “tall,” “red,” or “of the United States”—can rarely change the meaning of the underlying noun it is modifying. As we observed in another corpus linguistics-based amicus brief we wrote in a different Supreme Court case:

An adjective is a word that modifies nouns and pronouns, primarily by describing a particular quality of the word that it is modifying. . . . In syntactic representations the adjective is a subordinate category, a dependent of the noun . . . that it modifies. In layman’s terms, this means that in the relationship between adjectives and their nouns, the noun is king—an adjective’s meaning and scope is always relative to the noun it is modifying . . . . Put another way, an adjective can hardly ever decontextualize a noun.⁹⁸

We think that this is important for evidentiary purposes: it dramatically expands the universe of helpful examples relevant in determining the original public meaning of the phrase in at least three ways. First, we can assume that references to “federal officers,” “national officers,” “officers of the federal government,” “United States officers” and similar phraseology are also references to “officers of the United States”—unless, of course, there is evidence that one of those phrases is a term of art.

Next, we also think that it makes historical details about state and local officers more relevant. While the fact that a state constitution might allow for a governor or other state officer to be elected does not tell us anything about whether the U.S. Constitution allows for a federal officer to be elected, it does provide evidence that there was nothing inherently wrong about officers generally being elected.

Finally, in his brief to the Colorado Supreme Court, President Trump argued, that “despite the many words and citations that treat the President as an officer[,] not one authority holds that the President is an officer of the United States[:] no case, no statute, no record of Congressional debate, no common usage, no attorney general opinion. Nothing.”⁹⁹ While our discovery of the Postal Act of 1799 has already proven that the former President’s assertion is not true, even if it were, it would not matter. Those many “words and citations that treat the President as an officer” are absolutely relevant unless it can be shown that the full phrase “officer of the United States” means more than the sum of its parts. Which it does not.

B. At the Time of the Founding, “Appoint” and “Elect” Were Used Interchangeably

Second, the argument that the President is not an officer of the United States is built on the assumption that there is a rigid distinction between appointments and elections. Blackman and Tillman have repeatedly stated this:

“In order to reach this . . . conclusion [that the President could be an ‘Officer of the United States’], the interpreter must reject the distinction we drew between elected and appointed federal positions.”¹⁰⁰

“The Constitution hardwired this distinction between appointment and election.”¹⁰¹

“The President is in ‘the service of the government,’ but is not appointed by a President, a court of law, or a department head; indeed, he is not appointed at all . . . the Constitution expressly states that the President ‘shall . . . be elected.’ Under the rule in Mouat, the President is not ‘strictly speaking’ an officer of the United States.’ ”¹⁰²

“Presidents are not appointed; they are elected.”¹⁰³

“Presidents are not ‘appointed’ by the ‘government.’ Rather, Article II describes the President as an ‘elected’ position in several clauses.”¹⁰⁴

And they are not alone: Steven Calabresi¹⁰⁵ and Michael Mukasey¹⁰⁶ have all made similar arguments. But there is one really good reason to “reject the distinction [they all draw] between elected and appointed federal positions”¹⁰⁷: it did not exist at the time of the Founding. It is a linguistic anachronism. The words appear to have been used interchangeably, at least to the extent that an election was considered a mode of appointment. For example, in a speech given during the Constitutional Convention, James Madison discussed different options for selecting the President: “The option before us then lay between an appointment by Electors chosen by the people —and an immediate appointment by the people.”¹⁰⁸ Likewise, during the impeachment trial of Senator William Blount, Congressman Robert Harper of South Carolina—one of the House Impeachment Managers—stated, “[T]he President himself is liable to be impeached, as well as the officers whom he appoints. So also is the Vice President. And yet these two great officers are appointed by the people themselves, in a manner far more direct and immediate than Senators, and removable at shorter periods.”¹⁰⁹

And these comments were not one-offs. As we will show below, the historical record—including the text of the Constitution itself, the Articles of Confederation, early state constitutions, and repeated statements to and by George Washington—overwhelmingly shows that the founding generation used the word “elect” and “appoint” largely as synonyms.

1. Text of the Constitution

To begin with, the text of the Constitution itself demonstrates that the founding generation did not use the words “appoint” and “elect” in the same binary fashion we do today. This is evident in the original Constitution’s discussion of the Electoral College. It is often easy to forget that while the President is “elected” (as it states in the Constitution), he is not elected by the people. He is elected by the Electoral College—a non-standing, multimember body of sorts whose members choose the President and Vice President by vote. But how are these Electors chosen? There is no Constitutionally required method. Article II, Section 1 simply states: “Each state shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors, equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress.”¹¹⁰

Ten states participated in the first Presidential Election of 1788–89.¹¹¹ While the methods for appointing electors varied, none followed the pattern we are accustomed to today with a single individual or council (with or without the approval of a legislative body) unilaterally making the decision. Instead, four states “appointed” these electors through popular elections. Unlike today where the voters generally cast ballots for Presidential candidates (with slates of electors who support the winner of the popular vote in the state being appointed behind the scenes in accordance with state law), in several states in 1789, the citizenry voted for individual electors. For example, “Virginia chose presidential electors by popular vote. The state was divided into electoral districts. The voters in each district selected one presidential elector, who was then certified to the governor by the sheriffs of the counties comprising that district.”¹¹² Delaware followed a similar procedure.¹¹³ And Pennsylvania and Maryland chose their electors through state-wide elections, with the highest vote getters at-large being “appointed.”¹¹⁴

Massachusetts and New Hampshire followed hybrid models. In Massachusetts, a popular election was held in each of eight districts, but then the legislature appointed one of the two highest vote getters.¹¹⁵ In New Hampshire, a state-wide election was held, and then the legislature selected five of the top ten vote getters.¹¹⁶ In three states—Connecticut, Georgia, and South Carolina—the state legislatures appointed their respective electors entirely themselves.¹¹⁷ But how does a multimember legislative body “appoint” someone? As one newspaper article about the legislative component of the Massachusetts process explained, “the Members of the two Houses [of the Legislature], assembled together in one room for the choice of Electors of President and Vice President of the United States, and proceeded to the choice by joint ballot.”¹¹⁸ In other words, they voted.

Thus, in nine of the ten states, the electors were chosen through some sort of election—either by the people, the legislature, or both. But these elections did not mean that the electors were not appointed. Instead, it demonstrates that at the time of the Founding, appointments and elections were not thought to be mutually exclusive.

2. Early State Constitutions

Next, we find evidence that the founding generation used the words “appointed” and “elected” interchangeably in the first constitutions of the thirteen original states. We collect the relevant text of these Constitutions in Appendices A and B. At the time, popular elections for chief executives and judges were almost unheard of.¹¹⁹ Instead, executive and judicial officers were typically chosen by a state’s general assembly. But while some of the states used the word elect to describe this process,¹²⁰ others used the word appoint.¹²¹ Often the same Constitution would use both words to describe the same process, often in the same sentence. For example, under the Georgia Constitution of 1776, the Governor was “cho[sen] . . . by ballot”¹²² by the legislature—that the drafters of the constitution considered this process to be both an election and an appointment is demonstrated by the governor’s constitutionally-mandated oath of office: “ ‘I, A B, elected governor of the State of Georgia, by the representatives thereof, do solemnly promise and swear that I will, during the term of my appointment, to the best of my skill and judgment, execute the said office faithfully and conscientiously’ according to law . . . .”¹²³ Likewise, the Maryland constitution states that “a person of wisdom, experience, and virtue, shall be chosen as Governor . . . by the joint ballot of both houses.”¹²⁴ The ballots were to be counted by “a joint committee of both Houses” and the results then reported to the rest of the Assembly so “that the appointment may be entered.”¹²⁵ However, the Constitution then specified that if after two rounds of voting, two or more candidates received the same number of votes, “then the election of the Governor shall be determined by lot.”¹²⁶

3. Articles of Confederation

Similarly, there is ample evidence that the Continental Congress under the Articles of Confederation used the two words interchangeably as well. The Articles of Confederation were adopted by the Continental Congress on November 15, 1777, as the new nation’s first constitution. It used the word “appoint” (or one of its variations) eighteen times, but for present purposes we will focus on just one. Article IX—the predecessor to the Constitution’s Appointments Clause—gave the “united states in congress assembled . . . [the] authority . . . to appoint . . . civil officers as may be necessary for managing the general affairs of the united states under their direction.”¹²⁷ But despite this verbiage, the Journals of the Continental Congress often used the word “elect” for such actions. We have reproduced just a handful of the many examples below:¹²⁸

Pursuant to the resolution of the 7th of March last, Congress proceeded to the election of a minister plenipotentiary, to succeed Mr. J. Adams at the court of the United Netherlands, and, the ballots being taken, William Livingston, esq. was elected having been previously nominated by Mr. Stewart.¹²⁹

Congress proceeded to the election of a secretary and paymaster to the navy board in the middle district, and the ballots being taken, Mr. Joseph Pennel was elected.¹³⁰

According to the order of the day, Congress proceeded to the election of a secretary for foreign affairs, but not agreeing in the choice, Resolved, That the order for electing a secretary for foreign affairs, be postponed till Monday next.¹³¹

Ordered, That the election of a secretary for foreign affairs be postponed till Friday next.¹³²

While some of these “elections” were undoubtedly uncontested—reminiscent of Senate confirmations today—others clearly featured multiple candidates, as demonstrated by a letter from James Mitchell Varnum, a delegate to the Continental Congress from Rhode Island, to General George Washington: “We have attempted to elect a Secretary at War—Genls Greene, Lincoln & Knox are in Nomination; all the Votes for one or other of those Gentlemen. We effected nothing.”¹³³

We also found evidence that in their private correspondence, many delegates (and former delegates) to the Continental Congress used the words “appoint” and “elect” as synonyms, even when talking about the same position. For example, in 1781, George Washington wrote a letter to Phillip Schuyler—the father-in-law of his young protegee Alexander Hamilton—informing Schuyler of his “prospect of . . . election” as “Minister of War” and urged him not to refuse the post “if the choice should fall on [him].”¹³⁴ But four years later, Washington wrote to Henry Knox after the latter was appointed to the same position, saying “It gave me great pleasure to hear of your appointment as Secretary at War – without a complimt [sic], I think a better choice could not have been made.”¹³⁵ Likewise, in 1779, John Adams wrote to Samuel Huntington, the President of the Continental Congress, stating: “I had Yesterday the Honour of receiving your Letter of the twentyeth of October inclosed with two Commissions, appointing me Minister Plenipotentiary, from the United States, to negotiate Peace and Commerce with Great Britain.”¹³⁶ Yet he told the Comte de Vergennes, “The Congress of the United States of America did me the honour to elect me their Minister Plenipotentiary, to negotiate a Peace with Great Britain.”¹³⁷

As in state constitutions, sometimes the terms were used interchangeably even in the same document. Consider the following letter from Samuel Huntington, President of the Continental Congress, to Thomas Jefferson, informing the latter of his appointment as a peace commissioner:

Before this comes to Hand your Excellency will have received my Letter of the 2d Instant with it’s [sic] Enclosures, by which you will be informed that a Negotiation for Peace between the Belligerent Powers may probably take Place through the Mediation of the Empress of Russia and Emperor of Germany.

In Consequence of which Congress have thought proper to add four other Plenipotentiaries to the Honorable John Adams Esquire to assist in the expected Negotiation, of which you are elected one . . . Your Appointment is ordered to be kept secret that the Enemy may not get Intelligence of your Embarkation.

With very great Respect I have the Honor to be Your Excellency’s Most obedient & most humble Servant,

Sam. Huntington President

ENCLOSURE

By the United States in Congress assembled June 14. 1781

Resolved That four persons be joined to Mr. Adams in negotiating a peace between these United States and Great Britain.

The following were elected

The honble. Benjamin Franklin

The honble John Jay

The honble. Henry Laurens

The honble. Thomas Jefferson¹³⁸

Given this clear linguistic pattern in both early state Constitutions and the Articles of Confederation, it seems odd to suggest that a rigid distinction between the two words would suddenly appear in 1787 in the U.S. Constitution.

4. Addresses to and from George Washington

We have shown that the founding generation frequently used the word “elect” to describe the process of appointment, but is the inverse true as well? Were the “elected” positions in the Constitution ever referred to as “appointments”? The answer is a resounding yes.

When George Washington was elected president in 1789, he was universally “sated with adulation.”¹³⁹ His trip from Mount Vernon to New York for his inauguration took longer than he expected because he was constantly delayed by public feasts. As one biographer described his procession through Pennsylvania, “[a]s Washington entered Philadelphia, he found himself, willy-nilly, at the head of a full-scale parade. Twenty thousand people lined the streets, their eyes fixed on him in wonder.”¹⁴⁰ Newspapers around the country reprinted a host of “addresses” from prominent citizens and government figures given that day congratulating the President on his electoral victory. These messages often spoke of Washington as having been appointed President. The “President and Supreme Executive Council of Pennsylvania” stood “to congratulate you upon the establishment of the federal constitution, and felicitate ourselves, and our country, upon your unanimous appointment to the Presidency of the United States.”¹⁴¹ The “Mayor, Recorder, Alderman and Common Council of the city of Philadelphia . . . assembled[] to present you our sincerest congratulations on your appointment to the station of President of the United States.”¹⁴² The “standing committee of the Pennsylvania State Society of the Cincinnati” congratulated Washington on his “appointment, by the people, to the office of first Magistrate of this great empire.¹⁴³ “The Synod of the Reformed Church in North-America” met for the first time since Washington’s “appointment” in order to congratulate him and “join in that great and general joy testified by all descriptions of citizens on your acceptance of the highest office in the nation.”¹⁴⁴ Not to be outdone, the leadership of the German Lutheran Church “announce[d] the joy we entertain” in Washington’s “appointment to the station of President in Chief.”¹⁴⁵ To be sure, other speeches spoke of his election or ascension to the Presidency, but references to his “appointment” were commonplace.

The newspapers also reprinted the President-elect’s response to each elegy, where we see Washington himself employing appointment language to describe his election. He thanked the leadership of Pennsylvania for their “affectionate congratulations . . . on my appointment to the Presidency of the United States.”¹⁴⁶ To the leaders of Philadelphia, he said he felt “particularly obliged . . . for your congratulatory address on my appointment to the station of President of the United States.”¹⁴⁷ In response to praise from the Governor of New Hampshire, Washington promised, that “[i]n discharging the duties of my civil appointment . . . the love of my country will be the ruling influence of my conduct.”¹⁴⁸ He also accepted “with peculiar pleasure, the address of the University of the state of Pennsylvania, upon my appointment to the first office in the Union,” but conceded that he suspected that his “fellow-citizens anticipate[d] too many and too great advantages from the appointment.”¹⁴⁹

Nor was this appointment language limited to the celebrations in Philadelphia. Two months later, Vice President John Adams gave a speech on behalf of the Senate, thanking the President for his address to Congress:

We the Senate of the United States return you our sincere thanks for your excellent speech, delivered to both houses of Congress; congratulate you on the complete organization of the federal government, and facilitate ourselves, and our fellow-citizens, on your elevation to the office of President—an office, highly important by the powers constitutionally annexed to it, and extremely honorably from the manner in which the appointment is made. The unanimous suffrage of the elective body in your favor, is peculiarly expressive of the gratitude, confidence and affection of the citizens of America.”¹⁵⁰

In light of these examples, we feel comfortable concluding that the terms appointment and election were truly interchangeable, at least to the extent that an election was one form of appointment.

* * *

To summarize, we have shown extensive evidence—including the text of the Constitution itself, early state constitutions, the Articles of Confederation and Journals of the Continental Congress, and statements to and from George Washington, John Adams, and James Madison—to demonstrate a consistent linguistic pattern of using the words “appoint” and “elect” interchangeably. In light of this, we feel confident in rejecting “the distinction [Blackman, Tillman, and others] dr[a]w between elected and appointed federal positions.”¹⁵¹

C. The Text of the Constitution Identifies the Presidency as an Office

Third, the original Constitution of 1789 repeatedly refers to the Presidency as an “Office”—a fact that is undisputed. For example, in Article I, it states “The Senate shall chuse . . . a President pro tempore, in the Absence of the Vice President, or when he shall exercise the office of President of the United States.”¹⁵² Likewise, in Article II, it states that the President “shall hold his Office during a Term of four Years” and limits eligibility “to the Office of President” to “natural born citizens” who have “attained the age of thirty-five years.”¹⁵³

In an amicus brief submitted to the Colorado Supreme Court, Tillman acknowledges this. But he claims that “although the President holds an ‘office,’ he is not an ‘Officer of the United States.’ ”¹⁵⁴

We find this distinction difficult to square with early case law. In United States v. Maurice—an important Appointments Clause case Chief Justice John Marshall heard while riding the Circuit—Chief Justice Marshall concluded that “an office is defined to be ‘a public charge or employment,’ and he who performs the duties of the office, is an officer. If employed on the part of the United States, he is an officer of the United States.”¹⁵⁵ While not binding precedent, Maurice was frequently cited by lower courts both before and after the ratification of the Fourteenth Amendment and has been cited approvingly by the U.S. Supreme Court seventeen times, including in the majority opinion of Metcalf & Eddy v. Mitchell,¹⁵⁶ and more recently in Justice Thomas’ concurring opinion in Lucia v. SEC.¹⁵⁷ Blackman and Tillman have repeatedly quoted Justice Felix Frankfurter’s quip that when language is “obviously transplanted from another legal source”—as the phrase “Officers of the United States” in Section 3 clearly is—“it brings the old soil with it.”¹⁵⁸ We see no reason why that soil should not include Chief Justice Marshall’s early definition of an officer of the United States explicitly linking offices with officers.¹⁵⁹

D. Additional Context About the Original Meaning of “Officer of the United States” in the 1789 Constitution.

Fourth, we find Blackman and Tillman’s textual analysis of the original meaning of the phrase “officers of the United States” to be incomplete because it overlooks important contextual details. The phrase appears in the original Constitution of 1789 four times: in the Appointments Clause, the Impeachment Clause, the Oath or Affirmation Clause, and the Commission Clause. Context leads us to disagree with Blackman and Tillman’s readings of three out of four of these clauses. Along the way we critique the argument recently presented that suggests the President is not an Officer of the United States because he does not take an oath that has the words “support the Constitution” drawn from Article VI of the Constitution.

1. Appointments Clause

With respect to the Appointments Clause, it is not true that the Constitution empowers the president to appoint “Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States,” as has been asserted by President Trump. He only has the authority to appoint “Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States, whose Appointments are not herein otherwise provided for.”¹⁶⁰ We think the most natural reading of this proviso is that at least some of the other positions specifically enumerated elsewhere in the Constitution are (1) officers of the United States and (2) are potentially appointed through alternative channels other than those spelled out in the Appointments Clause.

We are not alone in that view. One of the original students of the Constitution, Alexander Hamilton, paraphrased the Appointments Clause for Federalist 67 as follows:

The capitalization—which was in the original—shows that Hamilton viewed the phrase “whose appointments are not herein otherwise provided for” as a modifier of “officers,” and that the phrase is making reference to officers mentioned elsewhere in the Constitution outside of the Appointments Clause.

Justice Antonin Scalia also reached this conclusion. Writing a concurrence in NLRB v. Noel Canning, Justice Scalia explained: “Except where the Constitution or a valid federal law provides otherwise, all ‘Officers of the United States’ must be appointed by the President ‘by and with the Advice and Consent of the Senate.’ ”¹⁶² Thus, Justice Scalia stated that there are Officers of the United States listed in the Constitution but not appointed by the President. Tillman actually wrote the Justice to ask what he meant by this statement. Justice Scalia surprised him by responding and left no doubt as to his position:

I meant exactly what I wrote. The manner by which the President and Vice President hold their offices is “provide[d] otherwise” by the Constitution. As is the manner by which the Speaker of the House and the President Pro Tempore of the Senate hold theirs.¹⁶³

The same view was taken by Thomas Merrill in a 2004 article, although he did not specifically mention the President and Vice President like Justice Scalia did.¹⁶⁴

Blackman and Tillman disagree, arguing for an alternative reading of “whose appointments are not herein otherwise provided for” in their third article in their ten-part series on Office and Officers:

The phrase . . . is, admittedly, a mouthful. We think this phrase tells the reader that the appointment of “Officers of the United States” is limited to the processes announced in Article II, Section 2. This sub-clause directs the reader not to scour the remainder of the Constitution for other provisions that provide authority to fill other federal “Officers of the United States” positions—by election or appointment. In other words, the Appointments Clause’s “not herein otherwise provided for”-language is not an invitation to search for other constitutional provisions providing authority to create or fill federal offices; rather this language puts the reader on notice that no such constitutional provisions exist beyond the textual bounds of Article II, Section 2.¹⁶⁵

In their view, anyone advocating for an “alternative reading that leads readers to look for other constitutional mechanisms to fill ‘Officers of the United States’ positions is mistaken.”¹⁶⁶ They critique the statements of both Hamilton and Scalia mentioned above, calling the former unclear¹⁶⁷ and the latter wrong.¹⁶⁸

But why? Because if Hamilton, Scalia, and Merrill are correct, the other officers “whose appointments are . . . provided for” elsewhere in the Constitution almost certainly includes at least some positions who are elected, either by the people or a multimember body such as the Electoral College or legislature, as shown in Table 1 below.

Table 1.
Position	Appointment Mechanism
Members of the House of Representatives	“[C]hosen every second Year by the People of the Several states”a
Electors [of Members of the House]	“Shall have the Qualifications requisite for Electors of the most numerous Branch of the State Legislature”b
Speaker and other Officers of the House	Chosen by Housec
Senators	“[C]hosen by the Legislature [of each state]”d
President Pro Tempore and “other Officers” of the Senate	Chosen by the Senatee
President of the United States	Electoral Collegef
Vice President	Electoral Collegeg
Electors for President and Vice President	“[I]n such Manner as the Legislature thereof may direct”h
Sources: a U.S. Const. art. I, § 2. b Id. c Id. d U.S. Const. art. 1, § 3. e Id. f U.S. Const. art. II, § 1. g Id. h Id.

As Blackman and Tillman explain, in their view “[o]nly appointed positions can be ‘Officers of the United States,’ i.e., positions ‘whose Appointments are not herein otherwise provided.’ Therefore, it would be a mistake to scour the Constitution for positions that are filled by election.”¹⁶⁹ Chad Squitieri reached a similar conclusion for similar reasons:

Article II, Section 2, Clause 2’s reference to “Appointments . . . not herein otherwise provided for” should not be understood as a reference to [other positions such as] Members of Congress. Instead, the use of ‘herein’ in Article II, Section 2, Clause 2 is best understood as a reference to Article II, Section 2, Clause 2 itself. Specifically, when Article II, Section 2, Clause 2 states “herein,” it references the types of appointed officers mentioned within the very same clause – i.e., “Ambassadors, other public Ministers and Consuls, Judges of the supreme Court.”¹⁷⁰

For this reason Squitieri did not feel that “Members of Congress” could be officers of the United States: “Article I does not speak to the ‘appointment’ of Members of Congress—it speaks to their election.”¹⁷¹

But we have already shown that the “the distinction [these scholars draw] between elected and appointed federal positions is a linguistic anachronism that did not exist at the time of the Founding. The Articles of Confederation, Journals of the Continental Congress, state constitutions, various founding fathers including George Washington, John Adams, and James Madison—not to mention the text of the Constitution itself—all used the terms “appoint” and “elect” interchangeably, at least to the extent that an election was a valid form of appointment. Once one understands this linguistic convention, we think the meaning of “whose Appointments are not herein otherwise provided for” is clear and see no reason that it would not include the President.

But what about Chief Justice Roberts’ statement in Free Enterprise that Blackman and Tillman frequently invoke to support their conclusion that officers of the United States cannot be elected: “The people do not vote for the ‘Officers of the United States’ ”?¹⁷² We do not think that our reading of the Appointments Clause is at odds with Chief Justice Roberts’s statement. The people do not vote for the President, the Electoral College does.¹⁷³ As our discussion in Section II.B and Appendix A and B demonstrates, there was a long history of multimember bodies “appointing” chief executives (and other officers) by ballot. That is exactly the process followed by the Electoral College. If the 152-member Virginia Legislature could “appoint” a judge,¹⁷⁴ surely the Electoral College—which in 1789 had only sixty-nine members¹⁷⁵—could “appoint” a President and Vice President as well.

There is a second reason we disagree with Blackman and Tillman’s reading of the Appointments Clause: it would limit the scope of the word herein to just that specific clause. And while that may not sound totally absurd when looking at the Appointments Clause alone, it makes no sense in the two other places the word appears in the Constitution. Take for example, the Direct Tax Clause in Article I, Section 9: “No Capitation, or other direct, Tax shall be laid, unless in Proportion to the Census or enumeration herein before directed to be taken.”¹⁷⁶ There are only two other references to an enumeration or census in the 1789 Constitution, neither of which is in Article I, Section 9. They are earlier, side-by-side in Article I, Section 2. The third and final use of the word herein is in the first sentence in the Constitution following the Preamble: “All legislative Powers herein granted shall be vested in a Congress of the United States.”¹⁷⁷ Such a narrow construction of “herein” there would completely destroy the Separation of Powers. Surely, Blackman and Tillman are not suggesting that there might be some legislative powers reserved for the President in Article II, so we are not sure why they insist on such a narrow construction of the inverse phrase in the Appointments Clause.

2. Impeachment Clause

We likewise are unpersuaded by Blackman and Tillman’s reading of the Impeachment Clause. They conclude that the “the president and vice president’s [express] enumeration in the Impeachment Clause in addition to ‘all civil Officers of the United States’ shows that the president and vice president are not deemed ‘officers of the United States’ themselves.”¹⁷⁸ We concede that this is a plausible reading of the clause, but we do not think it is the best reading. In English, this grammatical construction is often used to highlight the most important or most famous member of a broader group. Consider the following actual line from a 1963 speech given by Congressman Hale Boggs of Louisiana on the floor of the House of Representatives: “He leads the orchestra when his records are playing. He’s Dave Seville and Alvin and all the Chipmunks. He dances the twist like his life depended on it.”¹⁷⁹ Was Representative Boggs really suggesting that Alvin was not a chipmunk? Or consider this line from an article from USA Today: “[Vice President] Pence told Hannity Monday that all of his discussions with Zelensky, and all of the administration’s contacts, ‘were based upon proper considerations of how we support Ukraine.”¹⁸⁰ Surely, the article wasn’t suggesting that President Zelensky was not one of the Trump administration’s contacts.¹⁸¹

Nor is this construction a modern development. It appears frequently in documents from the Founding Era. Consider the closing line from a letter sent from General Charles Lee to his Commander-in-Chief: “My love to Mrs Washington and all the Ladies–adieu.”¹⁸² General Lee and General Washington may have had their differences, but Lee was clearly not suggesting that his commanding officer’s wife was not a “lady.” Likewise, one set of General Orders signed by Washington in 1777 stated, “The Commander in Chief thanks the Majors General Sullivan and Greene, and all the officers, and soldiers, engaged this day, to pursue the enemy, for their alacrity and zeal manifested in that service.”¹⁸³ Per Blackman and Tillman’s logic, we would have to conclude that Washington was suggesting that neither Sullivan—one of the heroes of the Battle of Trenton—nor Greene—the Continental Army’s Quartermaster-General—were “officers.” We think not.

The fact that Justice Joseph Story also felt that the language of the Impeachment Clause “would rather lead to the conclusion that they were enumerated, as contradistinguished from, rather than included in the description of civil officers of the United States”¹⁸⁴ does not dissuade us. Others have suggested that this express enumeration might be because the President as Commander-in-Chief, and the Vice President with his potential to serve as Acting President, are not strictly speaking civil officers of the United States, but rather both military and civil officers. We think this interpretation is compelling, but also not necessary. While we are loath to disagree with Justice Story, we do not think he was infallible.

Context matters here. And we think that context points to the best reading of the Impeachment Clause being that the President and Vice President being the two most important members of the broader category “officers of the United States”—the Alvin among the proverbial Chipmunks.

3. The Presidential Oath and the Article VI Oath

In their latest article, Blackman and Tillman reference an argument from the litigation.¹⁸⁵ That argument begins with the premise that the President takes an oath to “preserve, protect, and defend” the Constitution, found in Article II, and does not take the oath to “support” the Constitution, found in Article VI. Because Section 3 refers to officers who have “previously taken an oath . . . to support the Constitution of the United States,”¹⁸⁶ the President, the argument goes, has not taken such an oath and is not in the scope of Section 3.

This argument does not persuade us. In the first place, we are confident that one cannot take an oath to “preserve, protect, and defend” the Constitution without implicitly swearing to “support” the Constitution. By swearing to preserve, protect and defend the Constitution, one swears to support it more.

Evidence from the time of the Fourteenth Amendment confirms our view. Recall that Section 3 extended to any “person . . . who, having previously taken an oath, . . . as an executive or judicial officer of any State, to support the Constitution of the United States” and subsequently engaged in insurrection.¹⁸⁷ Thus, no one doubts that executive officers in the Southern states—for example, South Carolina—who had taken an oath prior to the rebellion, were covered by Section 3.

But when you look at the oath South Carolina officers were required by the South Carolina Constitution to take, the language mirrors the presidential oath, not the Article VI oath:

Every person who shall be chosen or appointed to any office of profit or trust; before entering on the execution thereof, shall take the following oath: “I do solemnly swear, (or affirm), that I will be faithful, and true allegiance bear to the State of South Carolina, so long as I may continue a citizen thereof; and that I am duly qualified, according to the constitution of this State, to exercise the office to which I have been appointed; and that I will, to the best of my abilities, discharge the duties thereof, and preserve, protect, and defend the constitution of this State, and of the United States: So help me God.¹⁸⁸

A newspaper transcript confirms this is the one and only oath that Governor William Henry Gist took in 1858.¹⁸⁹ Governor Gist went on to sign South Carolina’s Ordinance of Secession.¹⁹⁰

Given no one doubts Section 3 was to apply to Governor Gist and other South Carolina rebels, it is clear that the drafters of the Fourteenth Amendment viewed an oath to “preserve, protect, and defend” the United States Constitution as inherently an oath to “support” the United States Constitution. Any other reading of Section 3 appears absurd to us.

And South Carolina is not the only state. As documented in Appendix C, it appears Florida also had an oath that mirrored the presidential oath, not the Article VI oath. Georgia’s oath for its governor likewise mirrored the presidential oath, while other officers received an oath that mirrored the language of Article VI.¹⁹¹

We also note that, in Florida, the constitution was drafted as a prerequisite to admission into the union.¹⁹² Thus, Congress viewed Florida’s antebellum Constitution, complete with its “preserve, protect and defend” language, as acceptable language to satisfy Article VI’s “support” requirement. The presidential oath, just like the oaths in Florida, Georgia, and South Carolina, qualifies as an oath under Section 3 of the Fourteenth Amendment

For the sake of completeness, we cite the Texas Constitution’s oath, again approved by Congress prior to Texas’ admission. Ignoring a section about dueling, the oath reads in full: “I, (A. B.) do solemnly swear (or affirm) that I will faithfully and impartially discharge and perform, all the duties incumbent on me as ———, according to the best of my skill and ability, agreeably to the Constitution and laws of the United States and of this State.”¹⁹³

If we were writing on a blank slate, we would doubt that an oath to “discharge and perform all the duties . . . agreeably to the Constitution” has the same vigor as an oath to either “support” or “preserve, protect, and defend” the Constitution. But this is not a blank slate: several Texans took an oath under their state Constitution to “discharge and perform all the duties . . . agreeably to the Constitution” and then forced Governor Sam Houston (who was loyal to the union) out of office as a part of joining the Confederacy.¹⁹⁴ They were obviously covered by Section 3. There is no basis for arguing the President is not covered by Section 3 because his oath is, if anything, more rigorous than the requirement to “support” in Article VI: “preserve, protect, and defend.”

But the problems with the argument that the President does not take an Article VI oath did not die with the confederacy. Today, South Carolina,¹⁹⁵ Georgia,¹⁹⁶ and Texas,¹⁹⁷ all administer a “Preserve, Protect, and Defend” oath to at least some of their officers—oaths that do not include the word support. Near as we can tell from videos of the inaugurations, these officials, like Governor Gist in 1858, only take one oath. Were this enough to avoid taking an Article VI and Section 3 oath, numerous state officials today would be exempt from Section 3 even if they someday participate in an insurrection.¹⁹⁸

Blackman and Tillman rely on parallels between the structure of the Oath or Affirmation Clause of Article VI and Section 3 to suggest that the drafters used the Oath or Affirmation Clause as a model for Section 3. We have no quarrel with that. But because the presidential oath, like oaths of some state officers, is an oath to support the constitution through the language “preserve, protect, and defend,” we make two suggestions. First, it is entirely possible that the President is mentioned as an officer in Article VI, as the President is “bound by Oath or Affirmation, to support this Constitution.”¹⁹⁹ He simply takes the more specific presidential oath to do so. Second, because we’ve already shown that oaths to “support” include oaths to “preserve, protect, and defend” and oaths to act “agreeably” we should be slow to read terms in Section 3 narrowly because they were (supposedly) used narrowly in the Oath or Affirmation Clause.

4. Commission Clause

Blackman and Tillman’s best evidence comes from the Commission Clause. It is true that Section 3 of Article II of the Constitution states that the President “shall Commission all the Officers of the United States.”²⁰⁰ If viewed alone, this might be a silver bullet. But as mentioned above, the Appointments Clause indicates that there are other Officers of the United States whose appointment mechanisms are provided for elsewhere in the Constitution.²⁰¹ But none of the officials listed in Table 1, supra, receive presidential commissions. This produces a bit of a conundrum. If Blackman and Tillman are right that officials that do not receive commissions cannot be “officers of the United States,” then the Appointments Clause contains a meaningless surplusage. By contrast, if the phrase “whose Appointments are not herein otherwise provided for”²⁰² is not surplusage, then either “all” does not mean “all” or the Commission Clause has not been liquidated appropriately and other officials—including the President and Vice President—should receive a Commission.

One way out of this Mexican standoff is to not read the Commission Clause literally. As Lawrence Solum has noted, originalism is not literalism:

[A] grave misunderstanding of contemporary formalism is the idea that formalists are seeking the literal meaning of legal texts, and nothing could be further from the truth. And that’s because once we become acquainted with the philosophy of language, we realize that verbal communication, oral communication, written communication does not rely on words and punctuation marks alone to convey meaning, it also relies on context.

We almost never say, explicitly, everything we wish to convey. Instead, we rely on a mutual recognition of reader or listener and author or speaker of the context of communication to fill in the gaps.

So a famous example from the philosophy literature, Jack and Jill are married. And most of the time, we fill in that utterance with to each other. Because usually, when you say Jack and Jill are married, you mean to each other, although there are contexts where you might say those words in order to convey that Jack and Jill are actually married to other people. [i.e. “I saw Jack and Jill coming out of the hotel room together, but Jack and Jill are married!”]

In the law, it is the same. Context does much of the work of legal communication.²⁰³

We think that the context suggests that the Commission clause should be understood to read the President “shall commission all the officers of the United States” other than himself or perhaps “shall commission all of the officers of the United States” that he appoints.²⁰⁴ While critics will argue that this is circular reasoning, we think it is superior to the alternative reading promoted by Blackman and Tillman and President Trump, which views a whole clause of the Constitution as a mere “ink blot.”²⁰⁵

* * *

In sum, we believe context matters, and the added context we’ve supplied here calls into question the conclusion that the 1789 Constitution implies that the President was not an Officer of the United States.

III. A PRIMER ON LINGUISTIC DRIFT

While our own review of the evidence from the Founding Era leads us to firmly conclude that the original public meaning of the term “officer of the United States” included the President and Vice President, we agree with Blackman and Tillman that “it is the sort of question on which dispassionate, reasonable minds can disagree after having reviewed competing streams of authority, argument, and evidence.”²⁰⁶ As we noted above, at a minimum the Appointments Clause and Commission Clause are in tension with each other, and some sort of saving construction is necessary to harmonize the two.²⁰⁷

We likewise appreciate Blackman and Tillman’s open-mindedness about the possibility that the phrase “officer of the United States” could have undergone “some linguistic drift or slippage between the 1788 ratification of the Constitution and the 1868 ratification of the Fourteenth Amendment.” As they explain in their 2021 article:

Let’s assume that the President is not an “officer[] of the United States” for purposes of the 1788 Constitution. Under that assumption, it is still possible that the President might be an “officer of the United States” for purposes of Section 3. Thus, a reader might take the limited position that the President is an “officer of the United States” for the purposes of Section 3’s jurisdictional element.

This position is conceivable. Indeed, more than a decade ago, Tillman suggested that linguistic drift may have occurred with respect to this phrase between 1788 and 1868. He wrote that “[t]he stretch of time between the two events [1788 and 1868] was more than half a century . . . . It is hardly surprising that in the post-bellum epoch new meanings might have accrued to older language. Such linguistic slippage is common.” . . . Absent contrary evidence, [however] the default presumption should be one of linguistic continuity, rather than a presumption of linguistic drift. In other words, the proponents of the view that Section 3’s jurisdictional element applies to the presidency have the burden to prove two propositions. First, proponents must show that the particular linguistic drift involving the Constitution’s “officer of the United States”-language had actually occurred. And second, proponents must show that Section 3’s “officer of the United States”-language, in fact, drifted to include the presidency. In other words, even if the meaning shifted over time, it is not self-evident that the shift would embrace the presidency. Both propositions must be proven.²⁰⁸

Although we continue to respectfully disagree with Blackman and Tillman about the original meaning of the phrase “officer of the United States” in the original Constitution—and therefore believe that “linguistic continuity” favors holding that the President is an “officer of the United States”—we will devote the remainder of our Article to marshaling evidence to demonstrate that even if they are correct about the meaning of the phrase in 1788, by 1865 the phrase had morphed to include elected officials, including the President of the United States.

But first, it is important to understand what we mean by “linguistic drift.” The phrase was coined by Edward Sapir—an American anthropologist and linguist—back in 1921 when he observed:

The drift of a language is constituted by the unconscious selection on the part of its speakers of those individual variations that are cumulative in some special direction. This direction may be inferred, in the main, from the past history of the language. In the long run any new feature of the drift becomes part and parcel of the common, accepted speech, but for a long time it may exist as a mere tendency in the speech of a few, perhaps of a despised few.²⁰⁹

It is difficult to be able to discern exactly when a word reaches its “tipping point,” the moment in which the meaning that was favored initially by just “a despised few” becomes the prevailing norm. There is an analogy here to the difficulty courts face in determining when a registered trademark has gone generic. Even though many people use the word “coke” to refer to any soda, it is still largely a reference to the principal product of the Coca-Cola Company. But the same is true after the tipping point has been reached. Even after the word “trampoline” and “escalator” reached a point of genericide, there were likely still those who used those words in their branded sense for some time.²¹⁰

Thus, we are not surprised that Blackman and Tillman have identified a few sources from around the time that the Fourteenth Amendment was ratified that use the phrase “officer of the United States” in a way that excludes the President. Whether these are vestiges of an earlier understanding of the phrase—as Blackman and Tillman suggest—or early-adopters of a linguistic innovation does not matter. As the subsequent sections will show, we think that at the time the 39th Congress convened to draft the Fourteenth Amendment, the public meaning of the phrase included the President of the United States.

IV. EVIDENCE THAT OFFICERS (INCLUDING OFFICERS OF THE UNITED STATES) MAY BE ELECTED

Blackman and Tillman assert that one of the principal reasons that the President cannot be an officer of the United States is because officers are appointed, not elected.²¹¹ They are not alone in this view. Steven Calabresi—a long-time Trump critic—has advanced similar arguments.²¹² However, as shown in Sections II.B above, this position is based on a linguistic anachronism. At the time of the Founding, the strict dichotomy between “appointments” and “elections” that we employ today did not exist. Rather, an election—either by the people or a multimember body such as a legislature or the electoral college—was viewed as one potential mode of appointment.

That officers could be elected at the time the Fourteenth Amendment was ratified was even more clear. In the following Section, we will detail evidence gleaned from the text of the Amendment itself, legislative history of the Fourteenth and Fifteenth Amendments, the ratification debates in the states, and other sources that show that people regularly talked of officers being “elected.” We note that in each of the subsequent sections, we have not limited ourselves to explicit references to “officers of the United States.” While we acknowledge that this is the best evidence, because we do not find the term “officers of the United States” to be a term of art, we believe that clear references to federal officers, officers generally, or analogous state officers are still relevant for understanding the original meaning of “OFFICERS of the United States” as used in Fourteenth Amendment, as we discussed in greater detail in Section II.A.

A. Evidence from the Text of the Fourteenth Amendment

The best evidence that at the time of the ratification of the Fourteenth Amendment the word “officers” was understood to encompass elected officials is the text of Section 3 itself:

No person shall be a Senator or Representative in Congress, or elector of President and Vice-President, or hold any office, civil or military, under the United States, or under any State, who, having previously taken an oath, as a member of Congress, or as an officer of the United States, or as a member of any State legislature, or as an executive or judicial officer of any State, to support the Constitution of the United States, shall have engaged in insurrection or rebellion against the same, or given aid or comfort to the enemies thereof.²¹³

Most of the scholarship about the scope of Section 3 of the Fourteenth Amendment has focused exclusively on federal officers, without considering the analogous state positions.²¹⁴ But having shown that “officers of the United States” was not a legal term of art at the time of the Founding,²¹⁵ the selection mechanism for the parallel state officials mentioned in Section 3 is equally valid evidence for whether “officers” can be elected as a general matter. This is especially true if Blackman and Tillman are right that “Section 3 was modeled after the structure and language of the Oath or Affirmation Clause.”²¹⁶ The parallel structure and language of the Oath or Affirmation Clause presents “Officers of the United States” and “Officers of the several states” as closely analogous positions: “The Senators and Representatives before mentioned, and the Members of the several State Legislatures, and all executive and judicial Officers, both of the United States and of the several States, shall be bound by Oath or Affirmation, to support this Constitution.”²¹⁷

As noted above and shown in detail in Appendix A, at the time the original Constitution was ratified, few states had a Governor elected directly by the people. The rest had their governors selected by the state’s General Assembly, usually through a ballot process that resembled (and perhaps inspired) the Electoral College. However, by the time the Fourteenth Amendment was ratified, these facts had changed considerably. By 1865 the vast majority of states had governors elected directly by the people.

A similar evolution took place with respect to judicial officers. As shown in Appendix B, at the time of the Founding, judicial elections—at least as we conceptualize them today—were unheard of. Instead, judges were typically selected by the General Assembly, appointed by Governors, or were themselves legislators wearing a separate hat. But as Harvard Law Professor Jed Handelsman Shugerman has noted, beginning in the 1840s, America experienced something of a Constitution-writing renaissance, with many states adopting amendments or rewriting their constitutions entirely, introducing judicial elections in the process as part of a broader set of anti-legislative reforms:

The constitutional revolutionaries of the time believed elected judges were more likely to enforce limits against legislative excesses. From 1846 to 1851, twelve states adopted judicial elections for their entire court systems, and five states adopted partially elective systems. By 1860, out of thirty-one states in the Union, eighteen states elected all of their judges, and five more elected some of their judges. There were also proposals to subject federal judges to election, but the federal constitution was far more difficult to change.²¹⁸

In fact, the language of Section 2 of the Fourteenth Amendment acknowledges this evolution explicitly. Section 2 abolished the Three-Fifths Compromise of the original Constitution, replacing it with a fairer calculation for representation: “Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed.”²¹⁹ In an effort to prevent Southern states from disenfranchising African Americans, the Amendment then ties future representation to the number of eligible voters.

But when the right to vote at any election for the choice of electors for President and Vice-President of the United States, Representatives in Congress, the Executive and Judicial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States, or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced in the proportion which the number of such male citizens shall bear to the whole number of male citizens twenty-one years of age in such State.²²⁰

“The right to vote at any election for the choice of . . . the Executive and Judicial officers of a State”—it is difficult to be more explicit that officers can be elected than that.

Thus, even if Blackman and Tillman were right about officers not being elected at the time of the Founding, we think these seismic changes of state constitutional law between 1789 and the outbreak of the Civil War would have necessitated an evolution of the meaning of “officers” generally to include elected officials.

B. Evidence from the Legislative History of the Fourteenth Amendment

Another rich source of evidence that the officers mentioned in the Fourteenth Amendment included elected officials is the legislative history produced by Congress while debating the merits of the Amendment. In citing this evidence, we are well aware—and agree with—much of the criticism about over-reliance on legislative history.²²¹ But in this and subsequent Sections, we are not invoking legislative history in an attempt to divine Congressional intent. Instead, we are looking for clues about the way Congress used certain words and phrases. As Judge Frank Easterbook, one of the great critics of the use and abuse of legislative history once stated, “Clarity depends on context, which legislative history may illuminate. The process is objective; the search is not for the contents of the authors’ heads but for the rules of language they used.”²²²

In total, we found statements from at least ten Senators and six Congressmen that suggest that according to “the rules of language they used” the word “officer” included elected officials. One of these statements included explicit references to “Officers of the United States.” Senator Thomas A. Hendricks of Indiana proposed a change to the language of Section 3 that would have limited those barred from holding office in the future to those who entered the rebellion while they were still officers of the United States or one of the states. The reason was because he felt that an individual’s oath of office expired at the end of each term:

Everybody, by virtue of his allegiance, is bound to obey the Constitution of the United States, to stand by the Union. But this oath of itself is an oath of office binding upon him as an officer, else why is it that if a Senator taking this oath, serves six years and is reelected, he is sworn again? For the simple reason that he is entering upon another term of service, and for that term of service he must take this official oath to obey the Constitution of the United States. I presume this oath means as if it read, “Senators and Representatives and all other officers in the United States and in the States shall be bound by an oath or affirmation to support the Constitution of the United States in their offices.” I know of no other purpose that there can be to require a special oath from an officer.²²³

By sweeping Senators and Representatives into the category of “officers of the United States,” he made clear that he believed the category to be broad enough to include positions elected by multimember bodies (such as Senators) or directly by the people (as with Congressmen).²²⁴

Other statements made clear that the speakers thought that federal officers could be elected, even if they did not use the full phrase “officers of the United States.” Since we have debunked the notion that “officers of the United States” was a legal term of art at the time of the Founding, these synonyms are equally valuable clues as to the intended meaning of the full phrase. Senator Luke Poland of Vermont stated that he felt the Amendment as written was more merciful than the rebels deserved because it preserved their right to vote: “we leave the great mass [of Southerners] utterly untouched, and the leaders with their lives, their property, the full enjoyment of all their civil rights and privileges, with the right of voting for all officers, both State and national, with the single restriction they shall not hold office.”²²⁵

A number of these statements came during the debate in the House over an ultimately rejected section which would have stripped former Confederates of the right to vote until 1870.²²⁶ For example, future president James A. Garfield—then a Congressman from Ohio—stated:

If the proposition had been that those who had been in rebellion should be ineligible to any office under the Government of the United States, and should be ineligible to appointment as electors of the President and Vice President of the United States, or if all who had voluntarily borne arms against the United States had been declared forever incapable of voting for a United States officer, it would, in my judgment, be far more defensible.²²⁷

Congressman Robert C. Schenck, also from Ohio, used similar language while supporting the ultimately rejected proposal, claiming that it

does not disfranchise, but refuses to enfranchise. If you say that the people of these States, because of their having been engaged in the rebellion, shall not vote for Federal officers, there is nothing taken from them, because they have already divested themselves of that privilege, voluntarily abandoned, given it up, flung it away by breaking loose from the rest of the Union, as far as by their act, disposition, and power they could do so.²²⁸

Likewise, Congressman Henry J. Raymond of New York, stated that the rejected section “proposes to exclude the great body of the people of those States from the exercise of the right of suffrage in regard to Federal officers.”²²⁹ Representative Rufus P. Spalding of Ohio supported this proposal to “disqualif[y] active and known rebels from participating in the election of Federal officers.”²³⁰

There were also a number of other statements that discussed electing officers in general.²³¹ For instance, while arguing that Section 3 would not impose a punishment on former Confederates, but merely withhold a privilege, Senator Edgar Cowan of Pennsylvania stated that “[a]n elector is one who is chosen by the people to perform that function, just the same as an officer is one chosen by the people to exercise the franchises of an office.”²³² Later in the debates he returned to this distinction, asking “is not the elector just as much the choice of the community as an officer is the choice of it, except that the electors are chosen by a class and described by a general designation, whereas the officer is chosen by name to perform certain functions?”²³³

The widespread understanding that officers could be elected was repeatedly highlighted in the back and forth between Senator John B. Henderson of Missouri and Senator William Pitt Fessenden of Maine, as the pair debated an amendment to Section 2 proposed by Henderson. At the time, the language of Section 2 stated that “whenever the elective franchise shall be denied or abridged in any State on account of race or color, all persons of such race or color shall be excluded from the basis of representation.”²³⁴ Henderson wanted to make the section more explicit, changing the language to read “[b]ut whenever the right to vote at any election held under the Constitution and laws of the United States, or of any State, is denied to any of the male inhabitants of such State, &c.”²³⁵ He stated that “the inference [of this amendment] will be that it applies only to those general elections at which political officers are elected, members of the Legislature, Governor, judges, &c.”²³⁶ While Fressenden disputed whether the amendment was really necessary, he clearly agreed that officers could be elected, stating that he believed that the original language was “intended to cover the election of officers generally.”²³⁷ Some time later, Senator George Williams of Oregon proposed his own amendment to Section 2 along the same lines, adding words which were ultimately ratified—“But when the right to vote at any election for the choice of electors for President and Vice President of the United States, Representatives in Congress, the executive and judicial officers of a State, or members of the Legislature thereof”²³⁸—in order to (in his words) “specify[] particularly the officers for which these people must be allowed to vote in order to be counted.”²³⁹

Finally, we found a number of statements that support the proposition that state officers could be elected. While this proposition is hardly controversial—as shown in the last Section, the language of the Fourteenth Amendment itself acknowledges it as fact—these statements are still relevant evidence for showing that officers as a class—be they federal or state—can be elected. Senator Henderson, after acknowledging that any effort to strip ex-Confederates of the right to vote would be unworkable, stated that under the Amendment, “Lee, Johnston, Wade Hampton, Moseby, and even Jeff Davis, are left as qualified electors, competent to vote for State officers and members of Congress.”²⁴⁰ Congressman John A. Bingham—the father of the Fourteenth Amendment—made a similar statement in the House,

This amendment does not disqualify any rebel or aider of the rebellion from voting at all the State elections for all State officers, nor does it disqualify them from being appointed presidential electors. It amounts, therefore, to this: though it be adopted, and made part of the Constitution, yet all persons “who voluntarily adhered to the late insurrection, giving it aid and comfort,” may vote at all the State elections for State officers, and, being largely in the majority in every insurrectionary State, may elect the State Legislature, which may appoint electors for President and Vice President of the United States, and from aught in the amendment may appoint rebels as such electors.²⁴¹

Another example came during a debate over whether Confederate officials who had taken advantage of President Johnson’s general pardon should be barred from holding future office under Section 3. Senator James Doolittle of Wisconsin believed that they should not. To demonstrate that the Amendment would still punish those most culpable, he noted that a number of “prominent rebel officials” remained unpardoned—535 of them—including thirty-seven “cabinet officers and governors of States.”²⁴² Senator Hendricks likewise spoke of “[j]udicial officers” being “elected.”²⁴³ Senator Henderson spoke of the people “elect[ing] . . . members of the Legislature, Governor, judges, &c”²⁴⁴ as “political officers.”²⁴⁵ And Senator Timothy O. Howe of Wisconsin quoted and summarized a letter from W.L. Sharkey to Secretary of State William H. Seward “in which he tells him that the Governor and all the officers elected by the people had been duly installed, qualified, and taken possession of their offices.”²⁴⁶ Senator Henderson, Senator Fessenden, and Senator Daniel Clark of New Hampshire even briefly discussed the election of “municipal officers” and “township officers” such as mayors and recorders.²⁴⁷

Taken together, these statements reveal a consistent speech pattern among the Framers of the Fourteenth Amendment of referring to elected officials at all levels of government—federal, state, and local— as “officers.” It is also worth noting that while there may be examples suggesting the contrary—examples that could be revealed by a future corpus linguistics analysis of the Congressional Globe—we did not find any.

C. Evidence from the Ratification Debates of the Fourteenth Amendment in the States

Next, we turn to the ratification debates of the Fourteenth Amendment in the States. While not as well documented as the debates in Congress, they can still be a valuable source of evidence about how particular words or phrases were understood by the broader public at the time. Research into these debates has been greatly aided by a collection published by Kurt Lash in 2021.²⁴⁸ It includes transcripts of state legislative history as well as contemporary newspaper articles reporting on these debates. Here, too, we see a consistent pattern—mined from the debates in Alabama, Louisiana, and North Carolina—of the word “officer” being broad enough to include elected officers.

Alabama: On the day Alabama ratified the Fourteenth Amendment (reversing its earlier rejection), the Alabama Senate Journal recorded the following two statements. First, “The Senate met at 12 m. and elected officers. The 14th amendment was ratified and the Senate adjourned until to-morrow.”²⁴⁹ And second, “In the House, officers were elected and the 14th amendment ratified.”²⁵⁰ Although these are legislative officers—as opposed to general state officers—the statements still show that officers can be elected as a general principle, not to mention elected by a multimember body.

Louisiana: An article reporting on the ratification of the Fourteenth Amendment by Louisiana, which was published by the Boston Daily News, contained the following order from General Buchanan, the Commanding General of Union forces in the state: “All civil officers acting under military appointment will transfer their offices and everything pertaining thereto to their successors, who have been duly elected, and who are qualified under the laws of the State.”²⁵¹

North Carolina: A Joint Committee Report Rejecting the Fourteenth Amendment contained the following statement:

A leading feature of this second section is, that, virtually, it makes the basis of representation to consist of the voters only, which is manifestly inconsistent with the theory of our political system. The voters are merely the appointing power, whose function is to select the representative; but his true constituency is the whole population. It is a great fallacy to maintain that an officer represents only those who vote for him.”²⁵²

Not only does this show that officers are elected, but also demonstrates that Founding Era understanding of election being a type of appointment continued into the 1860s.

The Lash collection also contains a proposed “compromise” amendment—reported in the New York Times—which was proposed by Southern Governors to President Johnson after a number of Southern legislatures initially refused to ratify the Fourteenth Amendment. The language of the Compromise Amendment’s Section 4—which relates to apportionment of representatives—is particularly relevant for our purposes:

Sec. 4. Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed. But when any State shall, on account of race or color, or previous condition of servitude, deny the exercise of the elective franchise at any election for the choice of Electors for President and Vice President of the United States, Representatives in Congress, Members of the Legislature and other officers elected by the people, to any of the male inhabitants of such State, being 21 years of age and citizens of the United States, then the entire class of persons so excluded from the exercise of the elective franchise shall not be counted in the basis of representation.”²⁵³

This statement not only shows that officers can be elected, but the phrase “officers elected by the people” also suggests that officers can be elected in other ways, such as by multimember bodies like legislatures or the Electoral College.

D. Evidence from the Legislative History of the Fifteenth Amendment

We also looked at the legislative history of the Fifteenth Amendment, which was passed by the 40th Congress. Although one Congress removed from the cohort that passed the Fourteenth Amendment, it is still a valuable source of evidence of the linguistic conventions used in the Fourteenth Amendment, especially since so many of the members of the 40th Congress were also members of the 39th Congress. As with the legislative history and ratification debates of the Fourteenth Amendment, we found a consistent linguistic pattern of referring to elected officials—including federal ones—as both “officers” and “officers of the United States.”

Senator Frederick Theodore Frelinghuysen of New Jersey:

The consequences, therefore, of adopting any separate system of qualifications for the right of voting under the Constitution of the United States would have been that in some of the States there would be persons capable of voting for the highest State officers, and yet not permitted to vote for any officer of the United States; and that in the other States persons not admitted to the exercise of the right under the State constitution might have enjoyed it in national elections.²⁵⁴

Representative Samuel Shellaburger of Ohio:

I understood the first proposition of the gentleman’s argument to be substantially this: that if the Constitution had reposed in the States the unlimited power to regulate the matter of voting for Federal officers it would involve this mischief, to wit: that thereby the power would be placed in the States to withhold from the Government the election of Federal officers at all, and that that mischief might be fatal to the Government itself. Am I right in that statement?²⁵⁵

Representative Charles A. Eldridge of Wisconsin:

If the power exists in the Federal Government to pass this bill, whether under any one or all the provisions referred to, then I admit that Congress has the right to control the whole question of suffrage and the qualification of electors for all officers, State and national. There can be no reason for its entering the State and determining the qualification of those who are to elect the officers named in the bill that will not apply to every officer of the State so far as the question of power is concerned. The electors of President and Vice President are not named in section four of the first article. The power claimed, therefore, under the word “manner” in this section can no more apply to them than to the Governor of the State or any other State officer. So that if it covers electors it may as well cover, and does as necessarily cover, all that is contemplated by the amendment proposed by the joint resolution.²⁵⁶

Representative James Burnie Beck of Kentucky:

It is contended by the gentleman from Massachusetts that this is only a political punishment to be imposed on such States as refuse to obey the mandates of the first section till such time as Congress can enforce its provisions, which he asserts provides that the right to vote for certain officers cannot be denied or abridged.²⁵⁷

E. Evidence from Popular Sources that Officers Are Elected

Finally, we looked at popular sources such as newspapers and found numerous references to “officers of the United States,” “federal officers,” “national officers,” and “officers of the general government” being elected. Searching the Newspapers.com database for the years 1850–1870, we found examples from almost every state and several territories. While our search was by no means exhaustive—we hope to perform a more comprehensive corpus linguistic analysis of the subject at some point in the future—it demonstrates the ubiquitousness of referring to “officers of the United States” in a way that includes elected officials.

Alabama:

The South, for the humble privilege of being allowed to have a hand in the election of federal officers, has permitted her rights to be assailed and our leading politicians have compromised their principles for the sake of currying favor with their Northern allies.²⁵⁸

Arkansas:

To every marshal or duly elected officer of the United States.—You, and each of you are hereby commanded to bring up the body of J.W. Brown, said to be held in unlawful confinement on board the steamer Commodore.²⁵⁹

Connecticut:

Mr. Blaine asked Mr. Stevens if the third section would not be considered objectionable, as it excluded from the right to vote for national officers all who have voluntarily aided rebellion, and asked if the amnesty would exempt such.²⁶⁰

Delaware:

A universal suffrage bill has been prepared for presentation at the next session of Congress. It does not extend the suffrage beyond the election of Federal officers.²⁶¹

Georgia:

However desirable it may be, in the minds of many, to abrogate the unjust discrimination on account of color which prevails in the qualification for voters in most of the States, and to establish a uniform rule in that respect, particularly in the election of Federal officers, the loyal people of the land have recently made too great a struggle for the maintenance of the Constitution, to seek to accomplish the object by Congressional enactment, at a sacrifice of the obvious meaning and spirit of that instrument.²⁶²

Idaho:

It will be remembered a bill was some time ago prepared and introduced in Congress, in anticipation of this so-called ratification of the establishment of a despotism upon the ruins of the Republic, taking the management and control of elections of all Federal officers entirely away from the States, and subjecting the whole to the dictation and control of Congress.²⁶³

Indiana:

While conservative and law-abiding citizens, who are deprived of the privilege of voting, may obey the law, others, with no characters to sustain and no reputation to lose—lawless in person and purse—such as are found in all our large cities, will vote for national officers in defiance of the law.²⁶⁴

Iowa:

They know that the present Rebellion is the unprovoked work of bad, ambitious Demagogues, who have made a legal election of National officers an assumed justification for the worst of crimes.²⁶⁵

Kansas:

It was then resolved that all who participated in the rebellion should be disfranchised from voting for Federal officers, and that the rebel debt should be repudiated.²⁶⁶

Kentucky:

That a faithful execution of the fugitive slave law—a non-interference with slavery where it exists in the States, by citizens of the non-slaveholding States—a non-interference with the slave owner in the Territories while Territories, and the condemnation and rejection for office of politicians of all parties who shall hereafter attempt to agitate the subject of slavery, or make it a question in elections for officers of the United States, would restore peace and harmony to the States and people thereof.²⁶⁷

Louisiana:

A radical member of Congress, now here, has prepared a bill, which will be presented at the opening of Congress, providing for national suffrage. It differs very materially from similar bills presented by Mr. Sumner last session, and confines the suffrage to elections for national officers.²⁶⁸

Maine:

In the coming campaign for the election of the officers of the national government, let the watchwords be the rights of the people, the rights of humanity.²⁶⁹

Massachusetts:

If they are not citizens of the United States, then they have no right to vote for officers of the United States.²⁷⁰

Maryland:

[T]he right of the loyal States to decide for themselves the suffrage question does not in our opinion, give them power to prevent citizens of the United States from voting for officers of the United States.²⁷¹

Minnesota:

He said that one singular thing in the report was comparing Minnesota to Wisconsin in regard to the election of her federal officers. Wisconsin elects her officers and pays them out of the State Treasury; and it would be inconsistent for Minnesota to elect her federal officers and then have them paid out of the U.S. Treasury.²⁷²

Missouri:

Mr. Raymond, of New York, while not willing to accept it as a condition precedent to Southern representation was willing that all of the amendment, but the third section, depriving those who voluntarily aided in the rebellion, from voting for Federal officers.²⁷³

Nevada:

We do not believe that it is one of the rights of any State to deny any citizen of the United States a voice in the election of officers of the general government.²⁷⁴

New Jersey:

This act gives United States officers power to make arrests at the polls, and to inspect all records of elections for Federal officers.²⁷⁵

New York:

Charles C. Burleigh supported the resolutions against allegiance to the Constitution, and opposed voting for officers of the United States.²⁷⁶

North Carolina:

Mr. Lincoln distinctly contends for the right of any State to confer upon negroes citizenship, and the right to vote for Federal officers.²⁷⁷

Ohio:

[S]upposing that no one should vote for a United States’ officer, only for State officers, the General Government would cease to be, in four years.²⁷⁸

Oregon:

There was an informal meeting of a good many Republican Senators and Representatives to-day, to see if some action could not be had in the Senate to strike out the third section of the Constitutional Amendment, which disfranchises rebels from voting for Federal officers.²⁷⁹

Pennsylvania:

The unprecedented position of the legally elected officers of the United States should have at least gained for them the generosity of their former political foes.²⁸⁰

South Carolina:

To make out the inconsistency, he leaves out all the State elections ‘so often recurring,’ and Mr. Calhoun’s influence, and represents me as having attributed our unanimity solely to the election of Federal officers.²⁸¹

Tennessee:

Hence all the arrangements for the election of Federal officers by the people were necessarily based upon the rule that the persons entitled by law of the States to vote for members of the popular branch of the State Legislature should be the persons who would have the right to vote for representatives to Congress and for the presidential electors.²⁸²

Texas:

The evils that follow from the concentration of the attention of the people to national offices are extravagance in expenditures, an intense excitement pending the election of national officers, and a neglect of the people and their representatives to look to their own home as calculated to benefit them in all the relations of life, and to make them a happy and prosperous community.²⁸³

Vermont:

[T]he people of the States of California will sustain and uphold the constitutionally elected officers of the United States government, in all constitutional efforts to preserve the integrity of the Union.²⁸⁴

Virginia:

Mr. Boutwell reported a bill declaring who may vote for Federal officers, which he gave notice he would call up for action in ten days.²⁸⁵

Wisconsin:

He has not only sought no office, but has been so scrupulous that, feeling it might be inconsistent and dishonorable to take any part in a government which he considered in league with injustice and wrong, he has for years abstained from voting for federal officers.²⁸⁶

When viewed collectively, we think it is beyond dispute that at the time of the ratification of the Fourteenth Amendment, the ordinary meaning of the word “officer” in general and “officers of the United States” in particular included elected officials.

V. EVIDENCE THAT THE PRESIDENT IS AN OFFICER OF THE UNITED STATES FOR PURPOSES OF THE FOURTEENTH AMENDMENT

Having shown the text, drafters of the Fourteenth Amendment, ratifiers of the Fourteenth Amendment, and others understood the word officers—including “officers of the United States”—to encompass elected officials, we now turn to the precise question of whether the President of the United States is an officer of the United States. In some respects, this is overkill. Having shown that the full phrase “officer of the United States” was not a legal term of art, President Trump’s concession that the President is an “officer” is lethal to his case. However, in the following Sections we will amass additional evidence to show that at the time of the drafting of the Fourteenth Amendment, it was a common linguistic convention to refer to the President as an officer of the United States.

A. Evidence from the Legislative History of the Fourteenth Amendment

As noted above, we looked to the legislative history of the Fourteenth Amendment not to determine the intended meaning of the Fourteenth Amendment, but to look for evidence of how the legislators used the phrase “officer of the United States” and its synonyms in the course of their duties.²⁸⁷ Unfortunately, we did not find any explicit references to the President (or Vice President) as an “officer of the United States.” We suspect that Blackman and Tillman would argue that this proves their point. As Tillman explained in his amicus brief to the Colorado Supreme Court, “These references to the President may have been made in a more colloquial sense, but they did not state the President was an ‘Officer of the United States.’ ”²⁸⁸ But that is exactly our point. The phrase “Officer of the United States” is not a term of art, and therefore its original public meaning is the “colloquial sense.”

As noted in Section II.A, even during the first few years of the Republic, when Congress was busy creating positions within the new government, Congress almost never used the full phrase “officer of the United States.” The same is true of the debates over the Fourteenth Amendment. We found only twelve explicit uses of the phrase “officer of the United States” and one use of “officers of the United States.” Of these, ten were quotations of the exact language of the proposed amendment and two were close paraphrases. But they did refer to the President as an “officer of the government,” “executive officer,” and “officer.” This is exactly what one would expect if the full phrase was not a term of art, and as such is still probative of the proposition that the Framers of the Fourteenth Amendment viewed the President as an officer of the United States.

For example, in discussing who had the power to declare the insurrection over, Senator Garrett Davis of Kentucky referred to the President as an “officer of the Government”:

[T]here was a necessity for some power, some officer of the Government to declare when the insurrection was suppressed. There is such a power and such an officer to execute it; and who is he? The Constitution had been attacked by an armed resistance to the execution of the laws, and an attempt to set up an independent power and government within the United States. It is made the duty of the President, by the Constitution, to the best of his ability to preserve, protect, and defend that Constitution, and to take care that the laws be faithfully executed throughout the United States.²⁸⁹

Senator James Doolittle of Wisconsin used the same phrase to discuss the relationship between the President and other officers within the Executive Department. He had been accused by Senator Trumbull of Illinois of suggesting that inferior officers were “officers of the President.” Senator Doolittle disagreed:

I stated that executive officers were responsible to the President as the chief executive officer of the Government. My friend from Illinois seems to think that because I made this statement that they are responsible to the President, because he under the Constitution has placed upon him the responsibility of seeing that the laws are faithfully executed, I intended to say that these men were subject merely to the will of the Executive and not to the laws of the land. Not at all, sir.²⁹⁰

In addition, Senator Timothy Howe of Wisconsin once referred to the President as an “executive officer” and Senator Davis twice referred to him as the “chief executive officer.”

Senator Timothy Howe of Wisconsin:

It was argued, I recollect, by the Senator from Pennsylvania [Mr. Cowan] some time since that the President had a peculiar gift, or a peculiar right, for doing these things because he was an executive officer.²⁹¹

Senator Garrett Davis of Kentucky:

We now see, though, that this majority, lately the friends of the President, are engaged in a war upon him, and that war manifests itself in various aspects and modes. They denounce him; they denounce his measures, his policy. He is a coordinate branch of the Government; or at least the executive department is, and he is the chief executive officer.²⁹²

Senator Garrett Davis of Kentucky:

The powers of a Government are unavoidably augmented and energized during war, and then there is generally an accord between the legislative and executive branches, produced by the active presence of a common danger and a mutual effort to avert it, that makes the chief executive officer the instrument to give effect to their common policy and purposes.²⁹³

We found this language particularly probative given the connection identified by Blackman and Tillman between the Oaths and Affirmation Clause and Section 3.²⁹⁴

We also found a fourth reference by Senator Davis to the President as simply an “officer.” He referenced a debate back at the start of the Civil War about whether to seat the Senators elected from the loyal portions of Virginia—i.e., what would become West Virginia—after the rest of the state had voted to secede. The question was whether “notwithstanding the State of Virginia had passed an ordinance of secession and was in the condition of armed and active insurrection against the United States, still she was one of the United States and in the Union.” Senator Davis said that the Senate decided that the question was a “political question” and “[t]hat the President is the proper officer and power to decide” it.²⁹⁵

B. Evidence from the Impeachment Trial of Andrew Johnson

We were unsatisfied with the relatively few references we found in the legislative history of the Fourteenth Amendment, especially since four of the six references we found came from a single Senator. After all, individuals can be linguistic rebels, part of the “despised few” Sapir discussed.²⁹⁶ We therefore looked at the transcript of the impeachment trial of President Andrew Johnson for more examples of legislative speech.²⁹⁷

Following the assassination of Abraham Lincoln, his Vice President, Andrew Johnson, became President. Johnson, a loyal southern Democrat, had replaced a Republican, Hannibal Hamlin, as Lincoln’s running mate in 1864. Given the Republican majorities in the House and Senate, conflict with Johnson soon occurred. Relevant to our discussion, Congress passed a law over President Johnson’s veto that restricted his ability to fire officers appointed with the advice and consent of the Senate.²⁹⁸ When Johnson ignored that law and removed Edwin Stanton as Secretary of War, he was impeached.

We selected the transcript of the trial as a document to examine because it involves frequent use of the word “officer” by the Congress after the Congress that passed the Fourteenth Amendment. We view this transcript as a resource to answer multiple questions about the term “officer of the United States.”

A search for the term “officer of the United States” reveals a limited number of hits during the debates over the Fourteenth Amendment, but several actually use that term to refer to the President. For example, during a lengthy speech explaining his views on the impeachment, Senator George Edmunds of Vermont said that “[t]o this tribunal, sworn to impartiality and conscientious adherence to the Constitution and the laws, they [the founding fathers] committed the high powers indispensable to such a frame of government, of sitting in judgment upon the crimes and misdemeanors of the President, as well as all other officers of the United States.”²⁹⁹

A statement of Senator Joseph Fowler of Tennessee is likewise evidence that the term “officer of the United States” includes the President. In explaining the Impeachment Clause of the Constitution, he stated:

The framers of the Constitution . . . defined in their great charter the offences for which a President or other officer could be impeached and divested of his office. The Constitution says that “the President, Vice-President, and all civil officers of the United States shall be removed from office on impeachment for and conviction of treason, bribery, or other high crimes and misdemeanors.”³⁰⁰

Here, the parallel structure of these sentences plainly indicates (1) that Senator Fowler viewed the President as an officer under the Impeachment Clause, and (2) that he did not see a distinction between “officer” and “civil officer of the United States.”³⁰¹

In addition, the trial transcript twice quotes an article by John C. Hamilton, the son of Alexander Hamilton, which specifically identifies the Vice President as an officer of the United States, while discussing how the Constitutional Convention decided to have the Senate try impeachments.³⁰² In this discussion, Hamilton recounts that on

the 8th of September, Roger Sherman raised the objection that the Supreme Court was “improper to try the President because the judges would be appointed by him.” This objection prevailed, and the trial was [e]ntrusted to the Senate by the vote of all the States with one exception; and thus, on the same day, immediately after, the subjects of impeachment were extended from treason and bribery to “other high crimes and misdemeanors,” and thus [e]ntrusted and thus enlarged, it was on the same day made to embrace “the Vice-President and other civil officers of the United States.”³⁰³

Obviously, the inclusion of the word “other” in the phrase “the Vice-President and other civil officers of the United States” implies that the Vice President is a civil officer of the United States. Thus, the trial reveals that John Hamilton viewed the Vice President as a civil officer of the United States. Since all of Blackman and Tillman’s arguments apply with equal force to the Vice President as to the President, we think that evidence that the Vice President is an officer of the United States is equally probative for the President, and vice versa. (We also note that if “officer of the United States” was understood at the time of the founding or subsequently to be a term of art that excluded certain officials including the President and Vice-President, one would imagine John Hamilton, as a son of one of the writers of the Federalist Papers, would have understood that.)³⁰⁴

While these are the only direct references to the President as an officer of the United States, several Senators referred to the President as an officer. We reproduce them below:

Senator Garrett Davis of Kentucky:

The Constitution has no provision declaring a violation of any of its provisions to be a crime; that is a function of the legislative power, and it has passed no law to make violations of the Constitution, or of official oaths, by the President or any other officers, crimes.³⁰⁵

Senator Reverdy Johnson of Maryland:

[B]ut the Constitution for wise purposes says that in the contingency of an impeachment of a President of the United States or any other officer falling within the clause authorizing an impeachment they are to become, as I understand, a court. So have all our predecessors ruled in every case; and who were they?³⁰⁶

Senator Charles Beckalew of Pennsylvania:

The Constitution provides that when there is no President or Vice-President to discharge the duties of the presidential office, such duties shall be discharged by some other officer to be designated by law, until a new President shall be chosen.³⁰⁷

Senator John Sherman of Ohio:

The power of removal is expressly conferred by the Constitution only in cases of impeachment, and then upon the Senate, and not upon the President. The electors may elect a President and Vice-President, but the Senate only can remove them. The President and the Senate can appoint judges, but the Senate only can remove them. These are the constitutional officers, and their tenure and mode of removal are fixed by the Constitution.³⁰⁸

Senator Thomas Tipton of Nebraska:

It appears that while General Emory was acting under a commission requiring him to observe and follow such orders and directions as he should receive from the President and other officers set over him by law, an order reached him embodying a section of law, which law had been previously approved by the President himself, but, as it provided that orders from the President and Secretary of War should be issued through the General of the army, or next in rank, and the President being engaged to remove the Secretary of War and thwart the action of the Senate, in a discussion with General Emory, as to his duty as an officer, said, “This (meaning the order) is not in conformity with the Constitution of the United States, which makes me Commander-in-chief, or with the terms of your commission.”³⁰⁹

We also examined the House proceedings on the impeachment of President Johnson, as well as material from surrounding weeks.³¹⁰ Here there are also references to the President as an Officer of the United States. Representative Aaron F. Stevens, for example, stated “the executive officers of the United States, from the President down, are creatures of the people, and not creatures of the President.”³¹¹ Representative John Bingham³¹² from the House floor in the final days before President Johnson was impeached similarly referred to the President:

Did the gentleman from New York [Mr. Brooks] not know . . . that it is written in the Constitution of this country that the President, the Vice President, and every other civil officer of the United States shall be removed from office on impeachment for, and conviction of high crimes and misdemeanors.³¹³

Other representatives referred to the President as an executive officer. Referencing President Johnson, Representative Shelby M. Cullom said, “We are to-day [sic] considering the report of the committee appointed by the House to prepare and report articles of impeachment against that high officer of the Government.”³¹⁴ Representative Robert T. Van Horn referred to the President as “the executive officer of the nation.”³¹⁵ Representative William H. Kelsey stated, “In England, the chief executive officer of the Government cannot be impeached. Here he can be.”³¹⁶ Representative Ebon C. Ingersoll stated, “The President is merely an executive officer and cannot rightfully exercise any of the functions belonging to a judicial officer.”³¹⁷

We also found references to the President as an officer. Representative Rufus P. Spalding called the President “this high officer of our Government.”³¹⁸ In a discussion on appropriations, Representative William E. Nilback stated, “The President is the commander in chief of the Army and Navy, and as such is as much entitled as any other officer of the Army or Navy to have detailed for his assistance any subordinate officer.”³¹⁹

Taken together, we believe that the legislative history of the Fourteenth Amendment and the Impeachment debates and Trial of President Andrew Johnson demonstrates a consistent linguistic practice of identifying the President as an officer generally, and “Officer of the United States,” specifically.

C. Evidence from President Andrew Johnson’s Appointment Proclamations

We also found that Andrew Johnson—the President at the time the Fourteenth Amendment was ratified—referred to himself as an “officer of the United States” in numerous official proclamations appointing individuals to important posts in the former Confederate states. For example, consider this May 29, 1865, Proclamation appointing William W. Holden Provisional Governor of North Carolina:

Whereas, The President of the United States is by the Constitution made Commander-in-Chief of the army and navy as well as chief Executive officer of the United States and is bound by solemn oath, faithfully to execute the office of President of the United States, and to take care that the laws be faithfully executed . . . . I, Andrew Johnson, President of the United States and commander-in-chief of the army and navy of the United States, do hereby appoint Wm. W. Holden provisional governor of the State of North Carolina.³²⁰

We found similar proclamations by Johnson appointing governors over Alabama,³²¹ Georgia,³²² Mississippi,³²³ Texas,³²⁴ and South Carolina.³²⁵ In each of them, he referred to himself as an “officer of the United States.” While these proclamations were largely formulaic, using almost word-for-word language, there were some interesting variations. In the Alabama, Mississippi, and North Carolina proclamations, he refers to himself as the “chief executive officer of the United States,” but in the ones for Georgia, Texas, and South Carolina he adds a word, identifying himself as the “chief civil executive officer.” This tiny difference persuades us that the terms that “chief,” “civil,” and “executive” were all just adjectives modifying “officers of the United States”—lest anyone try to argue that that a “chief executive officer of the United States” or “executive officer of the United States” is somehow different from an “officer of the United States” for purposes of Section 3.

D. Evidence from the Amnesty Proclamations of Presidents Lincoln and Johnson

A fourth strain of evidence that, at the time the Fourteenth Amendment was ratified, the phrase “officers of the United States” included the President is the amnesty proclamations issued by Presidents Abraham Lincoln and Andrew Johnson pardoning confederates. On December 8, 1863, President Lincoln issued “a full pardon” which “restor[ed] [] all rights of property” to “all persons who have, directly or by implication, participated in the existing rebellion,” provided that they willingly took an oath to “support, protect and defend the Constitution of the United States, and the union of States thereunder” and respect all laws and proclamations issued by Congress and the President respecting slavery during the Civil War.³²⁶ Then in May 1865, President Andrew Johnson issued his own amnesty proclamation “grant[ing] to all persons who have directly or indirectly participated in the existing rebellion . . . amnesty and pardon, with restoration of all rights of property, except as to slaves.”³²⁷ Both of these proclamations contained a long list of exemptions—individuals participating in the rebellion that were not covered by the general pardon—chief among them “all who are, or shall have been, civil or diplomatic officers or agents of the so-called Confederate government” as Lincoln put it,³²⁸ or in the words of Johnson, “All who are, or shall have been, pretended civil or diplomatic officers, or otherwise, domestic or foreign agents, of the pretended Confederate Government.”³²⁹

Subsequent history demonstrates that Confederate President Jefferson Davis and Vice President Alexander H. Stephens were not covered by either of these amnesty proclamations. Davis was dogged with prosecutions for years.³³⁰ As for Stephens, he was elected to the U.S. Senate in 1866 but prohibited from taking his seat due to restrictions on former Confederates.³³¹ While he would go on to serve as a Congressman from the State and Georgia’s fiftieth governor, he held both of these positions only after Congress passed the Amnesty Act of 1872.³³²

But why were they excluded from Lincoln and Johnson’s amnesty proclamation? Obviously because they were “civil officers . . . of the pretended Confederate Government.” It is the only exemption that could possibly apply. And yet, the Confederate Constitution was modeled after the U.S. Constitution, and the four clauses that Blackman and Tillman cite to support their thesis—that the President and Vice President are not officers of the United States—are copied word-for-word as shown below, with the exception that “Confederate States” is substituted in place of “United States,” and some tweaks to capitalization.³³³

Table 2.
United States Constitution	Confederate States Constitution
Art. II, § 2: “[The President] shall nominate, and by and with the Advice and Consent of the Senate, shall appoint Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States, whose appointments are not herein otherwise provided for . . .”	Art. II, § 2: “[The President] shall nominate, and by and with the advice and consent of the Senate shall appoint, ambassadors, other public ministers and consuls, judges of the Supreme Court, and all other officers of the Confederate States whose appointments are not herein otherwise provided for . . .”
Art. II, § 3: “[The President] shall . . . Commission all the Officers of the United States.”	Art. II, § 3: “The President shall . . . commission all the officers of the Confederate States.”
Art. II, § 4: “The President, Vice President and all civil Officers of the United States, shall be removed from Office on Impeachment for, and Conviction of, Treason, Bribery, or other high Crimes and Misdemeanors.”	Art. II, § 4: “The President, Vice President, and all civil officers of the Confederate States, shall be removed from office on impeachment for and conviction of treason, bribery, or other high crimes and misdemeanors.”
Art. VI: “The Senators and Representatives before mentioned, and the Members of the several State Legislatures, and all executive and judicial Officers, both of the United States and of the several States, shall be bound by Oath or Affirmation, to support this Constitution.”	Art. VI: “The Senators and Representatives before mentioned, and in the members of the several State Legislatures, and all executive and judicial officers, both of the Confederate States and of the several States, shall be bound by oath or affirmation to support this Constitution.”

Yet, surely no one in the North would have allowed Alabama to elect Jefferson Davis to the Senate on grounds that he was just the President, not an officer, of the Confederate States.

Critics might quibble that neither Amnesty Proposal used the phrase “officers of the Confederate States” or at least “officers of the so-called Confederate States,” and that the broader term “officers of the so-called Confederate government” was more all-encompassing. But does anyone really believe that it would have made a difference if it did? We have already shown that the phrase “officers of the United States” was not a term of art, so why would its counterpart be? An “officer of the so-called Confederate government” was the same thing as an “officer of the so-called Confederate States.”

Others might argue that this line of reasoning is irrelevant because the Confederate Constitution was never recognized by the United States as valid law due to the Confederate States never being recognized as a legitimate country. But it is still evidence of the linguistic norms of the day for at least twelve states—twelve states that ultimately ratified the Fourteenth Amendment. Davis and Stephens were therefore officers only in “the colloquial sense.”³³⁴ But that is exactly the point. The colloquial understanding—or to put it in legal terms, original public meaning—of the officers of a country, pretended or otherwise, included the President and Vice President.

E. Evidence from Other Contemporary Sources

Having shown that the President was frequently referred to as an officer of the United States, federal officer, and officer in various legal sources, we turn now to other more popular sources as evidence of the original public meaning of the Fourteenth Amendment.

First, we found dozens of newspaper articles written between 1850 and 1870 that refer to the President explicitly as an officer of the United States. These articles—which we found by searching the Newspapers.com database—came from more than two-thirds of the states that were part of the Union when the Fourteenth Amendment was ratified. They included newspapers from the Deep South, the far West, the mid-Atlantic, the Midwest, and New England; papers in large cities like New York and Philadelphia and small towns like Rock Island, Illinois; papers that were pro-Union and pro-Confederacy. Some of the articles were actually reprintings of official government documents or Congressional speeches, while others were written as letters to the editor or mere gossip; some were written by local authors, while other articles we saw reprinted in papers in multiple states. Through it all, we noticed a consistent linguistic pattern of using the phrase “officer of the United States” in a way that included the Presidency.³³⁵ A fraction of the quotations are listed below:

Alabama:

On the 20th of June, the day of his letter, there were a President of the United States, a Cabinet, Judges of the Supreme Court, and thousands of other civil officers of the United States.³³⁶

Arkansas:

This creature [i.e., Brigham Young] and his deluded followers are in the constant habit of denouncing the President and all the other officers of the United States in the most indecent terms.³³⁷

California:

Great power is contided [sic] to the President, Vice President, and othhr [sic] civil officers of the United States.³³⁸

Connecticut:

[T]he President of the United States is by the Constitution made Commander-in-Chief of the Army and Navy, as well as chief executive officer of the United States, and is bound by solemn oath faithfully to execute the office of President of the United States.³³⁹

District of Columbia:

Mr. Fillmore . . . has been a faithful and honest President. . . . No chief executive officer of the United States ever displayed more wisdom, moderation, and conciliation.³⁴⁰

Georgia:

[W]hereas, the President of the United States is, by the Constitution, made commander-in-chief of the army and navy, as well as chief civil executive officer of the United States . . . I, Andrew Johnson, President of the United States . . . do hereby appoint James Johnson, of Georgia, Provisional Governor of the State of Georgia whose duty it shall be . . . to prescribe such rules and regulations as may be necessary.³⁴¹

Idaho:

The president and other officers of the United States receives a very small salary compared to the crowned heads of Europe.³⁴²

Illinois:

Their contest has been one of principle alone—a principle which, if Mr. Lincoln is the chief executive officer of these United States for the next four years, he will be compelled to carry out.³⁴³

Indiana:

Now, the President is an officer of the United States.³⁴⁴

Iowa:

This vain old man was made to believe that he was in communication with the Secretary of State, the President, and other important officers of the United States.³⁴⁵

Kansas:

Jefferson Davis . . . personally advised and assisted in maturing the plan for the cowardly murder of the President and other officers of the United States government.³⁴⁶

Kentucky:

The ‘august master’ of Russia, in his letter to President Lincoln, has given the Chief Executive officer of the United States some wholesome advice.³⁴⁷

Louisiana:

No provision of this nature has ever been made for the widows or families of any one of the Presidents or other civil officers of the United States.³⁴⁸

Maine:

Whereas the President of the United States is by the Constitution made commander-in-chief of the army and navy, as well as chief Executive officer of the United States . . . I, Andrew Johnson, President of the United States, and commander-in-chief of the army and navy of the United States, do hereby appoint William W. Holden, Provisional Governor of the State of North Carolina.³⁴⁹

Michigan:

No provision of this nature has ever been made for the widows or families of any one of the Presidents or other civil officers of the United States.³⁵⁰

Minnesota:

The Evening Post argues editorially that the plot for the murder of Lincoln, Johnson, Seward, Stanton and Grant, was known and approved by Jeff. Davis and other rebel leaders. It says at the very time when the assassins in Washington were preparing to do their work, Davis opened negotiations with Sherman, in which he dealt with that General as if Sherman were in fact chief officer of the United States Government, the others [including Lincoln] being supposed to be killed.³⁵¹

Mississippi:

At the instance of President Johnson, and to facilitate his patriotic work of reconstruction, the people of the South elected members [to] Congress. Without this invitation from the Chief Executive officer of the United States, they would not have afforded the radical majority in Congress an opportunity for perpetrating an outrage which virtually defeats the end for which so much blood and treasure were expended during the past four years.³⁵²

Missouri:

Then the clause after being so amended so as to include besides the President, the Vice President, and other civil officers of the United States . . . was agreed to as now found in article 2, section 4 of the Constitution.³⁵³

New York:

[T]he result was that neither the President, Vice President, nor other civil officer of the United States could lawfully do an act . . . contrary to the good morals . . . of the office he holds.³⁵⁴

North Carolina:

Here the President declares, not merely as Commander-in-chief, but as “Chief Executive officer of the United States,” that under the Constitution of the United States it is his duty to enforce the laws.³⁵⁵

Ohio:

The design of the provision undoubtedly was to prevent the juncture of executive and legislative authority in the same individual; and unless its force is destroyed by some other provision, it is evident that neither the President nor any other officer of the United States . . . can legally be a member of either House.³⁵⁶

Pennsylvania:

The official papers of Davis captured under the guns of our victorious army in the Rebel capitol identified beyond question or shadow of doubt, and placed upon your record, together with the declarations and acts of his conspirators and agents, proclaim to all the world that he was capable of attempting to accomplish his treasonable procuration of the murder of the late President, and other chief officers of the United States.³⁵⁷

South Carolina:

[T]he Provisional Governor is hereby authorized and empowered to appoint a competent Agent . . . and also as Agent of the Provisional Governor of this State in all matters which he may desire to bring through such Agent before the President or other officers of the United States Government.³⁵⁸

Tennessee:

[T]he President is an officer of the United States.³⁵⁹

Vermont:

no less so in respect to Senators or Representatives than in respect to the President or any other officer of the United States³⁶⁰

Virginia:

All persons who shall have knowledge of such plot, and shall not disclose the same to the President of the United States or some other officer of the United States, shall be guilty of misprision of treason . . . .³⁶¹

Wisconsin:

bill declaring the effect of impeachment by the House of Representatives, on the President and other officers o[f] the United States³⁶²

It is worth noting that at least a few of these articles were reprintings of the official proclamations mentioned above, where President Johnson explicitly identified himself as the “Chief Executive Officer of the United States” or “chief civil executive officer of the United States.”

Other newspaper accounts clearly referred to the President as a federal officer without explicitly using the phrase “officer of the United States.” A few are shown below:

Connecticut:

Mr. Covode on Monday also moved a resolution inquiring into the outlay of money for the purpose of electioneering, &c.,—and also inquiring whether the President or any other officer of government has with the use of money, patronage, or any other improper means, sought to influence Congress.³⁶³

Delaware:

This left me [James Buchanan] no alternative, as the chief executive officer under the Constitution of the United States, but to collect the public revenue and protect the public property, so far as might be practicable under the existing laws.³⁶⁴

Maine:

While at Washington, subsequent to his escape from Richmond, the loyalty of Mr. Starrett was abundantly substantiated to the satisfaction of the President and other officers of the Government.³⁶⁵

Maryland:

Hon. John Cochrane accompanied the officers of the Seventh Regiment to the President’s House this morning, and introduced them to the President and other officers of the Government.³⁶⁶

Massachusetts:

It declares the title of all abandoned lands to be in the United States, and forbids the President or any other officer of the Government from surrendering it or doing any act to impair or affect the title of the United States.³⁶⁷

New Jersey:

The Embassy first landed at Washington and will be received by the President and other officers of the Government with great ceremony.³⁶⁸

Pennsylvania:

He was the President, the chief officer of the government . . . .³⁶⁹

West Virginia:

[T]hey not only often call it requisition, but find it to answer the end desired, which is proven by their nomination for President, and other officers of the government.³⁷⁰

We also found evidence in legal treatises of the day. In Commentaries on American Law, the great American jurist James Kent stated, “The President is the great responsible officer for the faithful execution of the law, and the power of removal was incidental to that duty, and might often be requisite to fulfil it.”³⁷¹ And Calvin Townsend in his educational reader, An Analysis of Civil Government, was even more explicit: “The Vice-President is an Officer of the United States.”³⁷² Because all of Blackman and Tillman’s arguments apply equally to the Vice President as it does to the President, we find this to be relevant as well.

Finally, in the Republican Party Platform of 1868, we found the following statement that explicitly identifies the President as an officer:

We profoundly deplore the untimely and tragic death of Abraham Lincoln, and regret the accession of Andrew Johnson to the Presidency, who has acted treacherously to the people who elected him and the cause he was pledged to support; has usurped high legislative and judicial functions; has refused to execute the laws; has used his high office to induce other officers to ignore and violate the laws.³⁷³

* * *

In summation, we have compiled a significant body of evidence—including numerous proclamations by President Andrew Johnson, statements by Senators and Representatives during the debates over the Fourteenth Amendment and the impeachment trial of Andrew Johnson, Amnesty Proclamations by President Lincoln and Johnson, as well as dozens of contemporary newspaper articles which suggest that at the time the Fourteenth Amendment was ratified, the term “officer of the United States” was broad enough to include the President.

V. HARTWELL AND MOUAT REVISITED

Having marshaled significant evidence that the original public meaning of the phrase “officer of the United States”—both at the time of the Founding and the ratification of the Fourteenth Amendment—was broad enough to include elected officials generally and the President, in particular, we now look with fresh eyes at the two cases Blackman and Tillman cite in support of their conclusion to the contrary: United States v. Hartwell and United States v. Mouat.

A. United States v. Hartwell Supports Our Conclusion that the President Is an Officer of the United States

United States v. Hartwell was a criminal case brought under the Act of August 6th, 1846, which criminalized embezzlement of public funds.³⁷⁴ The Defendant was a clerk in the office of the assistant treasurer stationed at Boston.³⁷⁵ The case focused on whether as a clerk, Hartwell was an “officer” within the meaning of the statute.³⁷⁶ The Supreme Court said that he was.³⁷⁷ Blackman and Tillman summarized the Court’s holding as follows:

Justice Noah Swayne wrote the majority opinion. He offered a two-part definition of an office. First, “[a]n office is a public station, or employment, conferred by the appointment of government.” Second, “[t]he term [office] embraces the ideas of tenure, duration, emolument, and duties.”

In Hartwell, the clerk “was appointed by the head of a department within the meaning of the constitutional provision upon the subject of the appointing power.” The court did not expressly connect the term “officer” in the embezzlement statute with the phrase “officer of the United States” in the Appointments Clause. However, the court’s discussion of the appointment being made by the head of the department suggests the two concepts were closely related—rightly so, in our view.³⁷⁸

They then conclude that because “Presidents are not appointed by the government” but are instead “elected by the people,” they cannot be Officers of the United States.³⁷⁹

As an initial matter, it bears repeating that Presidents are not “elected by the people.” They are elected by the Electoral College, which is as much an organ of the government as Congress or the Supreme Court is. Furthermore, as we showed in Section II.B, at the time of the Founding, the words “elect” and “appoint” were used interchangeably. Remember James Madison’s comment at the Constitutional Convention about the Electoral College we quoted above? “The option before us then lay between an appointment by Electors chosen by the people — and an immediate appointment by the people.”³⁸⁰ The Constitutional Convention chose the prior. The Joint Committee Report from North Carolina, which we quoted in Section IV.C, shows that this understanding of the word “appoint” continued at the time Justice Swayne was writing: “The voters are merely the appointing power, whose function is to select the representative.”³⁸¹ In light of this linguistic insight, we think Presidents easily satisfy the Hartwell test.

We think the opinion supports our conclusion in at least two additional ways. First, as Blackman and Tillman note, the opinion does not use the full phrase “officer of the United States,” instead using the words “officer” and “public officer.” Yet it is clear that the Court is analyzing Hartwell’s position under the Appointments Clause. If there was an understood legal or colloquial distinction between “officers” and “officers of the United States,” we think Justice Swayne would have felt it necessary to use the latter phrase. Instead, we think the Hartwell opinion strengthens our view that all references to the President as an “officer” is evidence that he is an “officer of the United States.”

Second, we think the opinion supports our reading of the Impeachment Clause. This actually comes from Justice Miller’s dissenting opinion which argued that the Defendant fell outside the contours of the embezzlement statute because he had not been explicitly entrusted with the money by an act of Congress. But in reaching that conclusion, we couldn’t help but notice one of the sections that he quoted:

That the Treasurer of the United States, the treasurer of the mint of the United States, the treasurers and those acting as such of the various branch mints, all collectors of customs, all surveyors of customs acting also as collectors, all assistant treasurers, all receivers of public moneys at the several land offices, all postmasters, and all public officers of whatever character, be, and they are hereby, required to keep safely . . . all the public moneys collected by them.³⁸²

Surely, Congress was not suggesting that the Treasurer of the United States, the treasurer of the mint of the United States, and other enumerated positions were not public officers. This is yet another example of the Alvin and the Chipmunks rule.

B. United States v. Mouat’s Test Misconstrues the Appointments Clause by Ignoring the Modifying Clause

United States v. Mouat considered whether a paymaster’s clerk—appointed by a paymaster in the navy with the approval of the Secretary of the Navy—was entitled to mileage reimbursement under the Act of June 30, 1876.³⁸³ The Act limited reimbursement to “actual travelling expenses” and prohibited “disbursing officers of the United States” from collecting “for mileages and transportation in excess of the amount actually paid.”³⁸⁴ Writing for the majority, Justice Samuel Miller stated:

What is necessary to constitute a person an officer of the United States in any of the various branches of its service has been very fully considered by this Court in United States v. Germaine, 99 U. S. 508. In that case, it was distinctly pointed out that under the Constitution of the United States, all its officers were appointed by the President, by and with the consent of the Senate, or by a court of law or the head of a Department, and the heads of the departments were defined in that opinion to be what are now called the members of the Cabinet. Unless a person in the service of the Government, therefore, holds his place by virtue of an appointment by the President, or of one of the courts of justice or heads of departments authorized by law to make such an appointment, he is not, strictly speaking, an officer of the United States.³⁸⁵

But, as pointed out in Section II.D.1 above, that is not actually what the Constitution says. The President, courts of law, and department heads do not appoint all of the officers of the United States. There is another category: those officers “whose Appointments are . . . otherwise provided for” elsewhere in the Constitution. Words, we note, that the Germaine Court failed to quote.

As such, we find Justice Miller’s statement that

Unless a person in the service of the government, therefore, holds his place by virtue of an appointment by the President or of one of the courts of justice or heads of departments authorized by law to make such an appointment, he is not, strictly speaking, an officer of the United States,³⁸⁶

to be simply wrong. It flies in the face of the express language of Article II. And we feel that both its rigid test—based as it was on an incomplete version of the Appointments Clause—and its suggestion that “Congress may have used the word ‘officer’ in some other connections in a more popular sense” should be disregarded as dicta.

As such, we actually think Blackman and Tillman are interpreting the historical record exactly backwards. Mouat is not a linguistic continuity of the original meaning of the phrase “officer of the United States” but rather a departure from it. As we have shown, the phrase was not a term of art at the time of the Founding. Instead, it referred broadly to almost all federal officials whose positions were established by law—be that the Constitution or a federal statute. And it was broad enough to encompass both elected officials generally and the President of the United States specifically.

That understanding—shared by Chief Justice Marshall in his opinion in Maurice and by Congress in the Postal Act—continued at the time of the drafting and ratification of the Fourteenth Amendment. As we have shown, the explicit text, legislative history, and ratification debates of the Fourteenth Amendment and legislative history of the Fifteenth Amendment—not to mention newspapermen across the country—consistently spoke of electing officers, including officers of the United States. And Congress, Presidential proclamations, newspapers, and academic works published around the time the Fourteenth Amendment was ratified routinely referred to the President and Vice-President as an “officer of the United States.” Blackman and Tillman stated that the burden was on “proponents of the view that Section 3’s ‘officer of the United States’-language includes the presidency” to “put forward evidence as probative as Mouat and Hartwell.”³⁸⁷ We think we have more than met that challenge.

CONCLUSION

There is plenty, frankly, that we do not know. We do not know the meaning of the word “insurrection” in the Fourteenth Amendment, or how that meaning would apply to the events of January 6, 2021. We have not done historical research on if the Amendment is “self-executing.” We do not know many things about Section 3 of the Fourteenth Amendment. And we emphatically take no position on pending litigation other than the issue this paper addresses. We understand this piece is entering a complex national debate accompanying a presidential election. We would bury this paper with disclaimers if needed to get this point across that we cannot and will not answer many important questions surrounding these big topics. As is, we’ve settled for the first paragraph of our conclusion.

But this we know: The term “officer of the United States” in the 1788 Constitution is not a term of art. It thus applies to all “OFFICERS of the United States,” as a standard textualist interpretation of the phrase implies. There is no doubt that the person who holds the office of President of the United States becomes an officer of the United States when the person takes the presidential oath. Donald Trump was an officer of the United States.

Even assuming that was not the end of the matter, we also know this from a wide range of sources: At the time of the Fourteenth Amendment, the term “officer of the United States” included elected officials. Many references in that era refer to the President himself, as well as the Vice President, as an “officer of the United States.” The historical record in 1868 confirms what has been true since 1789: the President of the United States is an officer of the United States.

APPENDIX A. SELECTION MECHANISM FOR GOVERNORS IN THE EARLY STATES
State Constitution	Selection Mechanism	Referred to as Appointment, Election, or Both?	Relevant Passages
Connecticut (1818)	General Election	Election	Art. IV, § 1: The supreme executive power of the State shall be vested in a Governor, who shall be chosen by the electors of the State, and shall hold his office for one year from the first Wednesday of May next succeeding his election, and until his successor be duly qualified. No person who is not an elector of this State, and who has not arrived at the age of thirty years, shall be eligible.
Delaware (1776)	Joint ballot of both houses	Appointment	Art. 7: A president or chief magistrate shall be chosen by joint ballot of both houses’ to be taken in the house of assembly . . . and the appointment of the person who has the majority of votes shall be entered at large on the minutes and journals of each house.
Georgia (1777)	Chosen by ballot by the General Assembly	Both	Art. II: On the first day of the meeting of the representatives so chosen, they shall proceed to the choice of a governor, . . . and of an executive council, by ballot out of their own body. Art. XXIV: “I, A B, elected governor of the State of Georgia, by the representatives thereof, do solemnly promise and swear that I will, during the term of my appointment, to the best of my skill and judgment, execute the said office faithfully and conscientiously’ according to law, without favor, affection, or partiality; that I will, to the utmost of my power, support, maintain, and defend the State of Georgia, and the constitution of the same.”
Maryland (1776)	Joint ballot of both houses	Both	XXV: That a person of wisdom, experience, and virtue, shall be chosen Governor . . . on the second Monday in every year forever thereafter, by the joint ballot of both Houses (to be taken in each House respectively) deposited in a conference room; the boxes to be examined by a joint committee of both Houses, and the numbers severally reported, that the appointment may be entered . . . . [I]f the ballots should again be equal between two or more persons, then the election of the Governor shall be determined by lot.
Massachusetts (1780)	General Election	Chosen/Election	Ch. II, Art. II: The Governor shall be chosen annually: And no person shall be eligible to this office, unless at the time of his election . . . . Ch. II, Art. III: Those persons who shall be qualified to vote for Senators and Representatives within the several towns of this Commonwealth, shall, at a meeting, to be called for that purpose, on the first Monday of April annually, give in their votes for a Governor.
New Hampshire (1776)	Appointed by Council	Appointment	Accordingly pursuant to the trust reposed in us, WE DO Resolve, that this Congress assume the name, power and authority of a house of Representatives or Assembly for the Colony of New-Hampshire And that said House then proceed to choose twelve persons, being reputable freeholders and inhabitants within this colony, in the following manner, viz. five in the county of Rockingham, two in the county of Stratford, two in the county of Hillsborough, two in the county of Cheshire, and one in the county of Grafton, to be a distinct and separate branch of the Legislature by the name of a COUNCIL for this colony, to continue as such until the third Wednesday in December next; any seven of whom to be a quorum to do business. That such Council appoint their President.
New Hampshire (1784)	General Election	Chosen/Election	Part II: The President shall be chosen annually; and no person shall be eligible to this office, unless at the time of his election, he shall have been an inhabitant of this state for seven years next preceding, and unless he shall be of the age of thirty years; and unless he shall, at the same time, have an estate of the value of five hundred pounds, one half of which shall consist of a freehold, in his own right, within the state; and unless he shall be of the Protestant religion.
New Jersey (1776)	Election by the Council & Assembly	Election	Art. VII: [T]he Council & Assembly jointly at their first Meeting, . . . shall, by a Majority of Votes, elect some fit Person within the Colony to be a Governor for one Year, the Governor.
New York (1777)	Election by freeholders of the State	Election	XVII: [T]he supreme executive power and authority of this State shall be vested in a governor; and that statedly, once in every three years . . . shall be, by ballot, elected governor . . . which elections shall be always held at the times and places.
North Carolina (1776)	Joint ballot of both houses	Election	Art. XV: [T]he Senate and House of Commons, jointly at their first meeting after each annual election, shall by ballot elect a Governor for one year.
Pennsylvania (1776)	Joint ballot of the General Assembly & Council	Chosen	§ 19: All vacancies in the council that may happen by death, resignation, or otherwise, shall be filled at the next general election for representatives in general assembly, unless a particular election for that purpose shall be sooner appointed by the president and council. The president and vice-president shall be chosen annually by the joint ballot of the general assembly and council.
Rhode Island (1842)	Election at the town, city, or ward meetings	Election	Art. VII, § 1: The chief executive power of this State shall be vested in a governor, who, together with a lieutenant governor, shall be annually elected by the people. Art. VII, § 11: The compensation of the governor . . . shall not be diminished during the term for which they are elected. Art. VIII, § 1: The governor . . . shall be elected at the town, city, or ward meetings, to be holden on the first Wednesday of April, annually.
South Carolina (1776)	Joint ballot of both houses	Chosen/Appointment	§ III: That the general assembly and the said legislative council shall jointly choose by ballot from among themselves, or from the people at large, a president and commander-in-chief and a vice-president of the colony. § XIV: That in case of the death of the president and commander-in-chief, or his absence from the colony, the vice-president of the colony shall succeed to his office, and the privy council shall choose out of their own body a vice-president of the colony, and in case of the death of the vice-president of the colony, or his absence from the colony, one of the privy council (to be chosen by themselves) shall succeed to his office, until a nomination to those offices, respectively, by the general assembly and legislative council for the remainder of the time for which the officer so dying or being absent was appointed.
Virginia (1776)	Joint ballot of both houses	Chosen/Appointment	A Governor, or chief magistrate, shall be chosen annually by joint ballot of both Houses (to be taken in each House respectively) . . . who shall not continue in that office longer than three years successively.

APPENDIX B. SELECTION MECHANISM FOR JUDGES IN THE EARLY STATES
State Constitutions	Selection Mechanism	Referred to as Appointment, Election, or Both?	Relevant Passages
Connecticut (1818)	Nomination by the Governor and appointment by the General Assembly for Supreme Court and lower court	Appointment	Art. V, § 1: The judges of the supreme court and of the superior court shall, upon nomination by the governor, be appointed by the general assembly in such manner as shall by law be prescribed. Art. V, § 3: Judges of the lower courts shall, upon nomination by the governor, be appointed by the general assembly in such manner as shall by law be prescribed, for terms of four years.
Delaware (1776)	Joint ballot of the President and General Assembly	Appointment	Art. 12. The president and general assembly shall by joint ballot appoint three justices of the supreme court for the State, one of whom shall be chief justice, and a judge of admiralty, and also four justices of the courts of common pleas and orphans’ courts for each county.
Georgia (1798)	General Election	Election	Art. III, § 1: The judicial powers of this state shall be vested in a superior courts . . . . The judges of the superior court shall be elected for the term of three years, removable by the governor.
Maryland (1776)	Appointment by Governor with advice and consent of the Council	Appointment	XLVIII: That the Governor, for the time being, with the advice and consent of the Council, may appoint the Chancellor, and all Judges and Justices.
Massachusetts (1780)	Appointment by Governor with advice and consent of the Council	Appointment	Ch. 2, Art. 9: All judicial officers . . . shall be nominated and appointed by the Governor, by and with the advice and consent of the Council.
New Hampshire (1784)	Appointment by President & Council	Appointment	Part II: All judicial officers . . . shall be nominated and appointed by the president and council . . . no appointment shall take place, unless three the council agree thereto.
New Jersey	N/A	N/A	N/A
New York (1777)	Appointment by Commission of Senators & the Governor	Appointment	XXIII: That all officers [including Chancellor, and Justices of the Supreme Court], shall be appointed in the manner following, to wit: The assembly shall, once in every year, openly nominate and appoint one of the senators from each great district, which senators shall form a council for the appointment of the said officers, of which the governor . . . shall be president and have a casting voice, but no other vote; and with the advice and consent of the said council, shall appoint all the said officers.
North Carolina (1776)	Appointment of the General Assembly by joint ballot of both houses	Appointment	XIII: That the General Assembly shall, by joint ballot of both houses, appoint Judges of the Supreme Courts of Law and Equity, Judges of Admiralty . . . who shall be commissioned by the Governor.
Pennsylvania (1776)	Appointment by the President with the Council	Appointment	§ 20: The president, and in his absence the vice-president, with the council, five of whom shall be a quorum, shall have power to appoint and commissionate [sic] judges.
Rhode Island (1842)	Election by the two Houses in grand committee	Election	Art. X, § 4: The Judges of the Supreme Court shall be elected by the two Houses in grand committee.
South Carolina (1776)	Chosen by ballot jointly by the General Assembly and Legislative Council and commissioned by the President and Commander-in-Chief	N/A	XX: That all other judicial officers shall be chosen by ballot, jointly by the general assembly and legislative council, and except the judges of the court of chancery, commissioned by the president and commander-in-chief.
Virginia (1776)	Appointed by joint ballot of the two Houses	Appointment	The two Houses of Assembly shall, by joint ballot, appoint Judges of the Supreme Court of Appeals, and General Court, Judges in Chancery, Judges of Admiralty, Secretary, and the Attorney-General, In case of death . . . the Governor . . . shall appoint persons to succeed in office. House of Assembly or the Privy Council.

APPENDIX C. OATHS THAT STATE EXECUTIVE OFFICERS TOOK IN SOUTHERN/FUTURE REBEL STATES IN THE ANTEBELLUM ERA
State Constitutions	Oath
Alabama (1819)	Art. VI, § 1: The members of the General Assembly, and all officers, executive and judicial, before they enter on the execution of their respective offices, shall take the following oath or affirmation, to wit: “I solemnly swear (or affirm, as the case may be) that I will support the Constitution of the United States, and Constitution of the State of Alabama, so long as I continue a citizen thereof, and that I will faithfully discharge, to the best of my abilities, the duties of ——— according to law: so help me God.”
Arkansas (1836)	§ 28: The appointment of all officers not otherwise directed by this constitution shall be made in such manner as may be prescribed by law; and all officers both civil and military acting under the authority of this State shall before entry on the duties of their respective offices take an oath or affirmation to support the Constitution of the United States and of this state and to demean themselves faithfully in office.
Florida (1838)	§ 11: Members of the General Assembly, and all officers, Civil or Military, before they enter upon the execution of their respective offices, shall take the following oath or affirmation: I do swear (or affirm,) that I am duly qualified, according to the Constitution of this State, to exercise the office to which I have been elected, (or appointed) and will, to the best of my abilities, discharge the duties thereof, and preserve, protect, and defend the Constitution of this State, and of the United States.
Georgia (1798)	Art. I, § 19: Every member of the senate or house of representatives shall, before he takes his Seat, take the following oath or affirmation. to wit: “I, A B, do solemnly swear (or affirm, as the case may be) that I have not obtained my election by bribery, treats, canvassing, or other undue or unlawful means, used by myself, or others by my desire or approbation, for that purpose; that I consider myself constitutionally qualified as a senator, (or representative,) and that, on all questions and measures which may come before me, I will give my Vote and so conduct myself as may, in my judgment, appear most conductive to the interest and prosperity of this State; and that I will bear true faith and allegiance to the same; and to the utmost of my power and ability observe, conform to, support, and defend the constitution thereof.” Art. II, § 5: The governor shall, before he enters on the duties of his office, take the following Oath or affirmation: “I do solemnly swear (or affirm, as the case may be) that I will faithfully execute the office of governor of the State of Georgia; and will, to the best of my abilities, preserve, protect, and defend the said State, and cause justice to be executed in mercy therein, according to the constitution and laws thereof.”
Louisiana (1845)	Title VI, Art. 90: Members of the General Assembly, and all officers, before they enter upon the duties of their office, shall take the following oath or affirmation: “I (A B), do solemnly swear (or affirm) that I will support the Constitution of the United States and of this State, and that I will faithfully and impartially discharge and perform all the duties incumbent on me as, according to the best of my abilities and understanding, agreeably to the Constitution and laws of the United States and of this State; and I do further solemnly swear (or affirm) that since the adoption of the present Constitution, I, being a citizen of this State, have not fought a duel with deadly weapons within this State, nor out of it, with a citizen of this State, nor have I sent or accepted a challenge to fight a duel with deadly weapons with a citizen of this State, nor have I acted as second in carrying a challenge or aided, advised or assisted any person thus offending, so help me God.”
Mississippi (1832)	Art. VII, § 1: Members of the legislature, and all officers, executive and judicial, before they enter upon the duties of their respective offices, shall take the following oath or affirmation, to wit: “I solemnly swear (or affirm, as the case may be) that I will support the constitution of the United States, and the constitution of the state of Mississippi, so long as I continue a citizen thereof, and that I will faithfully discharge, to the best of my abilities, the duties of the office of __________________ according to law. So help me God.” Art. VII, § 6: The legislature shall pass such laws to prevent the evil practice of duelling as they may deem necessary, and may require all officers before they enter on the duties of their respective offices, to take the following oath or affirmation: “I do solemnly swear (or affirm, as they case may be) that I have not been engaged in a duel, by sending or accepting a challenge to fight a duel, or by fighting a duel since the first day of January, in the year of our Lord one thousand eight hundred and thirty-three, nor will I be so engaged during my continuance in office. So help me God.”
North Carolina	Provision not found in Constitution.
South Carolina (1790, as amended in 1834)	Art. IV, § 5: Every person who shall be chosen or appointed to any office of profit or trust; before entering on the execution thereof, shall take the following oath: “I do solemnly swear, (or affirm), that I will be faithful, and true allegiance bear to the State of South Carolina, so long as I may continue a citizen thereof; and that I am duly qualified, according to the constitution of this State, to exercise the office to which I have been appointed; and that I will, to the best of my abilities, discharge the duties thereof, and preserve, protect, and defend the constitution of this State, and of the United States: So help me God.”
Tennessee (1835)	Art. X, § I: Every person who shall be chosen or appointed to any office of trust or profit, under this Constitution, or any law made in pursuance thereof, shall, before entering on the duties thereof, take an oath to support the Constitution of this State, and of the United States, and an oath of office.
Texas (1845)	Art. VII, § 1: “I, (A. B.) do solemnly swear (or affirm) that I will faithfully and impartially discharge and perform, all the duties incumbent on me as ———, according to the best of my skill and ability, agreeably to the Constitution and laws of the United States and of this State: And I do further solemnly swear (or affirm) that since the adoption of this Constitution by the Congress of the United States, I being a citizen of this State, have not fought a duel with deadly weapons, within this State, nor out of it, nor have I sent or accepted a challenge to fight a duel with deadly weapons, nor have I acted as second in carrying a challenge, or aided, advised or assisted, any person thus offending – so help me God.”
Virginia	Provision not found in Constitution.

98 S. Cal. L. Rev. 65

Download

James A. Heilpern*

* Research Fellow at the Georgetown Center for the Constitution. The authors would like to thank Daniel Ortner, Sarah Jenkins Dewey, Christian Sanchez Leon, Alex Worley, and Samuel Clemence for their editorial help, as well as Kindra Heilpern and Robert Forsberg for facilitating access to important research materials. The authors would also like to thank Josh Blackman, Seth Tillman, Will Baude, and H. Jefferson Powell for comments on previous drafts of this article.

Michael T. Worley†

† Attorney, Provo, UT. The work of this paper should not be attributed for good or ill to my employer or any other entity. These entities had nothing to do with this paper, which was written off the clock.

The Lost History of “History and Tradition”

March 2025March 2025Dov Fox* & Mary Ziegler†

The Supreme Court has decided one blockbuster after another by appealing to “history and tradition,” deploying this trope to remake key features of the constitutional landscape: from overturning Roe to abolishing affirmative action; from narrowing the scope of public accommodations to widening the margin for church/state entanglements. The Court says that its history-and-tradition test emerged fully formed in 1997 from an assisted-suicide case that was designed to rein in the drift toward living constitutionalism under the Warren and Burger Courts. This origin story is compelling. The problem is that it is not true—not where the test came from or even what it is. The Court’s narrative erases decades of social-movement conflict that this Article is the first to excavate. This Article marshals original archives to reveal that the history-and-tradition test was fashioned from the crucible of earlier struggles: over the value of deep and more recent history, over which communities and what kind of evidence define tradition, and, ultimately, over the role of America’s past in our constitutional present.

The contours of this debate were sharpened across doctrines and eras, inside the courts and beyond them. Recovering this history uncovers a rival vision of the history-and-tradition test—not entrenched but evolving. Taking seriously this dynamic alternative makes three contributions. First, it casts doubt on major decisions about race, abortion, guns, and God, while challenging the fixed-in-time conception of traditionalism these rulings stand on. Second, the more adaptive version of history and tradition sheds light on puzzles, including the levels-of-generality problem, the constitutional-progressive response to charges of judicial activism, and how a modern conservative legal coalition was forged out of fierce divisions over originalist methods and outcomes. Finally, this lost method bears surprising implications for claims that span the ideological spectrum: from fetal rights and gay marriage to gender-affirming care and conversion therapy. For instance, entrenched history and tradition probably would not operate to protect widely accepted rights such as interracial marriage and medical refusal. And evolving traditionalism could protect yet-unrecognized rights that it would not protect right now, such as aid-in-dying or assisted reproduction.

[G]uided by the history and tradition . . . , we must ask what the Fourteenth Amendment means by the term “liberty.” . . . [The answer is that it] does not protect the right to an abortion.¹

—Justice Alito

[T]his Court has instructed that the Establishment Clause [for government involvement with religion] must be interpreted by “reference to historical practices and understandings.”²

—Justice Gorsuch

Only if a firearm regulation is consistent with this Nation’s historical tradition . . . [does] the individual’s conduct fall[] outside the Second Amendment’s “unqualified command.”³

—Justice Thomas

INTRODUCTION

How should courts determine when individual freedoms enjoy constitutional standing? This is the central question that the U.S. Supreme Court answered in a recent trilogy of rulings about abortion, guns, and God.⁴

All three ascertained the status and scope of constitutional rights by invoking “history and tradition.”⁵

In Dobbs v. Jackson Women’s Health Organization, the Court enlisted a history-and-tradition test to overturn Roe v. Wade⁶ on the ground that a right to abortion is not “deeply rooted in this Nation’s history and tradition.”⁷ New York State Rifle & Pistol Association v. Bruen revolutionized the Court’s Second Amendment doctrine by striking down gun control measures that do not “comport[] with history and tradition.”⁸ And in Kennedy v. Bremerton School District, the Court set aside its fifty-year-old Lemon test—which assesses church/state violations according to a law’s purposes, effects, and entanglement with religion⁹—and replaced it with an approach that’s anchored in the “traditions undergirding” the Establishment Clause, “consistent with a historically sensitive understanding” of its provisions.¹⁰

The role assigned to history and tradition varies across these contexts, as does the relationship between the two.¹¹ The constant in each is how rights claims are interpreted based on history and tradition: not what the text originally meant to people who were alive when it was ratified or the normative understandings that are most desirable to us today, but the social practices that resonate with our shared history and tradition. Reliance on history and tradition has made waves in a range of other constitutional domains too: from free speech, affirmative action, and voting rights, to tribal authority, immigration, and sentencing.¹² All told, the Court’s appeal to history and tradition has upended decades of settled law across the constitutional landscape, with a raft of new occasions on the horizon.¹³

A rich literature has emerged to unpack this turn to history and tradition. Sherif Girgis identifies the distinguishing mark of that test as gleaning constitutional meaning from “practices” that are widely accepted or deeply embedded within American life.¹⁴ Miranda McGowan expounds that these meaning-making practices may be undertaken by elected officials, social movements, or other ordinary citizens working together.¹⁵ Larry Solum and Randy Barnett explain that the relevant social practices enjoy density, breadth, and staying power in relation to a constitutional amendment or provision at issue.¹⁶

Marc DeGirolami distinguishes those practices from other sources of interpretation.¹⁷ They’re not moral principles like the “evolving standards of decency” that shape which punishments the Eighth Amendment bars as cruel or unusual, consistent with “the progress of a maturing society.”¹⁸ Nor do judicial precedents count as practices¹⁹—that is, unless it is state court judges applying state constitutional texts. Curtis Bradley and Neil Siegel clarify that the social practices which make up history and tradition constitute more than “historical gloss” on constitutional powers or limits.²⁰ They’re a standalone criterion for adjudicating disputes about when and why the Constitution recognizes individual rights.

Scholars have also criticized the use of history and tradition in Dobbs. Reva Siegel reveals that the majority’s method of discerning social practices by counting state laws was deployed by segregationists before it was rejected in Brown v. Board of Education.²¹ Aaron Tang argues that almost half the states that Dobbs credits wholesale abortion bans to, in fact, allowed the ending of pregnancies until quickening.²² Melissa Murray and Katherine Shaw show how democratic deficits like gerrymandering cast doubt on Dobbs’s rhetorically potent claim that abolishing the national right to abortion would facilitate deliberation in the states.²³

Yet the most salient feature of this history-and-tradition test has received just passing reference. Namely, it asks whether the putative right in question fits within a pattern of practice relative to the Bill of Rights or Civil War Amendments.²⁴ That inquiry is rooted in the era of ratification. The Supreme Court has presented this temporal limit as natural and neutral, detached from preferences or politics.²⁵ On the Court’s telling, the test emerged in response to the freewheeling, unenumerated-rights decisions of the Warren and Burger Courts. The Rehnquist and Roberts Courts turned to the history and tradition of a long-ago past in order to rein in that earlier impulse for Justices to make policy.²⁶

This fixed-in-time approach promises to anchor constitutional rights and even to resolve interpretive disputes about them by appeal to shared customs. Yet it distorts the origins and implications of the history-and-tradition test, obscuring fierce debates over its meaning and significance in 2024 cases about guns,²⁷ immigration,²⁸ presidential immunity,²⁹ and free speech.³⁰ The history-and-tradition approach has fractured Justices on matters including the space it makes for post-ratification evidence,³¹ what other kinds of evidence help to establish a tradition,³² what level of generality a tradition should be articulated at,³³ whether history and tradition can constrain Justices,³⁴ and its relationship to originalism.³⁵ These conflicts cannot be fully understood without a sense of the historical struggles over history and tradition that shape today’s debate. This Article excavates the origins of these debates to uncover a rival version of the history-and-tradition test missing from contemporary cases and commentary. Our analysis chronicles the deep roots that this hidden conception has in the Court’s own jurisprudence and the social-movement struggle that has unfolded alongside it. And it spells out distinctive attractions of this alternative for constitutional theory and practice.

Part I and II mine original archives to unearth a more dynamic understanding of this test that sees the potential for consequential practices to change over time when newer customs dislodge older ones. On this account, traditions aren’t entrenched but evolving. Those of more recent vintage must still have a longstanding pedigree and be deeply embedded into the fabric of the American life, even if they’re contested in a diverse and polarized country. This evolving form of traditionalism runs through half a century of conservative and liberal opinions, amicus briefs, and social movements. “[T]radition is a living thing” is how Justice John Marshall Harlan II described this ideal in 1961.³⁶

Part II reveals that entrenched traditionalism did not emerge until the 1980s. That is when Christian conservatives and the right-wing legal movement sought to bridge growing divides among them by harmonizing originalist interpretive methods with natural-law principles that were seen as predating the Constitution and informing it. The Federalist Society embraced interpretive methods based on original public meaning.³⁷ But antiabortion lawyers and other social conservatives sometimes found originalism too limiting and worried that it would not deliver their preferred outcomes that states be made to promote Christian values, and that laws permitting abortion be struck down as unconstitutional.³⁸ An entrenched approach to history and tradition forged a powerful coalition on the political right, allowing for evidence of what social conservatives deemed Christian teachings and Western values.

Part III spells out three payoffs of recovering the evolving vision of that test. First, it makes progress on doctrinal puzzles like the levels-of-generality problem;³⁹ the enduring charges that reliance on tradition is both too manipulable (covering for judicial activism) and too intransigent (rooting injustices);⁴⁰ and the blurry lines between traditionalism and methods that resemble it, like historical gloss (to clarify ambiguous terms) and liquidation (to lock in their meaning).⁴¹ Second, resurrecting that evolving history-and-tradition test offers a shared vocabulary with the mediating potential to broker ideological compromise (between the prevailing interpretive extremes of conservative and progressive constitutional theory) by splitting the difference between originalism and living constitutionalism.⁴² Finally, we apply both traditionalisms—entrenched and evolving—to show how each would resolve controversies from fetal rights and gay marriage to gender-affirming care and conversion therapy. The implications are surprising. For example, the entrenched history-and-tradition test likely would not protect widely accepted rights such as interracial marriage and medical refusal. Meanwhile, evolving traditionalism might protect yet-unrecognized rights like aid-in-dying and assisted reproduction in a future that made these practices routine, but it probably would not protect any of them right now.

I. SOCIAL MOVEMENT POLITICS

Today’s Supreme Court majority advances a distinctive origin story for the history-and-tradition test. Dobbs directs our attention to struggles over substantive due process in the 1960s and 1970s.⁴³ But there are other plausible places to begin telling the history of the history-and-tradition test. The uses of history in jurisprudence could be documented from the Court’s early cases at the outset of the nineteenth century. A more recent point of departure emerges in the second half of the twentieth century, when social movements first contested the meaning of history and tradition. Or one could start with the uses of history and tradition that have become so salient over the Court’s last couple Terms. A sensible middle ground for launching the history of history and tradition is when the Court got into the business of recognizing unenumerated rights in the early twentieth century.

It is true that Supreme Court decisions of the nineteenth century mentioned rights that animate the “fundamental principle[s] of a republican government”⁴⁴ or that applied to “all civilized nations.”⁴⁵ For example, in the 1823 case of Corfield v. Coryell, George Washington’s nephew, Justice Bushrod Washington, described “privileges and immunities which are, in their nature, fundamental; which belong, of right, to the citizens of all free governments; and which have, at all times, been enjoyed by the citizens of the several states which compose this Union.”⁴⁶ History and tradition also appears to have figured centrally in early jurisprudence on the Seventh Amendment.⁴⁷ These earlier discussions matter for various purposes adjacent to ours.⁴⁸

We focus our historical analysis on the twentieth and twenty-first centuries because that reflects the most faithful understanding of the history-and-tradition test applied by the Court today.

The question at the center of our inquiry—whether tradition and history are entrenched or evolving—became far more central to social-movement and judicial debate in the past several centuries than it was before. Consider the account of “general law” developed by William Baude, Jud Campbell, and Stephen Sachs—they argue that the Fourteenth Amendment “secured but did not confer” rights already recognized in general law, which was “derived from general principles and customs and operating across jurisdictions.”⁴⁹ These scholars acknowledge that the scope of the general law was characterized by “imprecision and woolliness”—not least when it came to the question of whether general-law principles were fluid, “capable of developing over time, through a course of long-standing legal practice”⁵⁰—and whether the “rights of Englishmen,”⁵¹ as Sachs writes, were a “closed set.”⁵² Baude, Campbell, and Sachs recognize that nineteenth-century Justices were comfortable with the imprecision of general law and did not always openly grapple with questions about the scope of common-law custom and tradition in ways we might expect today.⁵³ In the twentieth and twenty-first centuries, by contrast, the fixed-versus-fluid question became a flashpoint for constitutional conflicts for a range of actors.⁵⁴ By studying their debates, we can gain crucial perspective on the workings and origins of both an evolving and fixed approach to history and tradition today.

Earlier discussions of common-law tradition bear an ambiguous relationship to present-day case law. With the decisions of Slaughter-House Cases⁵⁵ and later, Erie Railroad Co. v. Tompkins,⁵⁶ the Court rejected the readings of the Privileges or Immunities Clause favored by some originalist scholars. Accordingly, Baude, Campbell, and Sachs acknowledge that their approach to general law might be—and remain—“legally dead.”⁵⁷ Earlier understandings of tradition—perhaps rooted in general law or in the Privileges or Immunities Clause—may not tell us much about how the Court will apply a history-and-tradition test in the future. While Sachs reads Dobbs as consistent with originalist arguments about the Privileges or Immunities Clause or general law,⁵⁸ profound tensions separate the approaches rejected in Slaughter-House and Erie from present-day traditionalism, which is better understood as a new phenomenon that’s distinct from originalism and worth taking seriously in its own right.⁵⁹

To be sure, by the twentieth century, the Court had long referred to tradition when the Justices decided the 1905 working-hours case of Lochner v. New York. There, the Supreme Court recognized a “liberty of person or of free contract” provided for in the Federal Constitution with little mention of where that liberty came from or how to identify it.⁶⁰ But in dissent, Justice Oliver Wendell Holmes advanced a role for history and tradition, suggesting that the meaning of the Fourteenth Amendment might be gauged by evaluating whether “a rational and fair man necessarily would admit that the statute proposed would infringe fundamental principles as they have been understood by the traditions of our people and our law.”⁶¹

Holmes did not explain how one should go about identifying such a tradition beyond suggesting that many a “reasonable man” would find nothing wrong with the disputed New York law.⁶² It would be another decade until a majority hinted at a role for tradition and history.⁶³ In the parental liberty case of Meyer v. Nebraska, the Court held that the Due Process Clause protected “those privileges long recognized at common law as essential to the orderly pursuit of happiness by free men”—without explaining how they counted, or how much.⁶⁴ We focus on the period after Meyer because judges, litigators, and eventually, grassroots movements began a more robust debate about how tradition mattered in law. Which traditions should count—and was tradition necessarily backward looking? What kinds of evidence could establish a tradition? These questions began to more centrally influence debates on the pages of the United States Reports and beyond in the first decades of the twentieth century.

A. The Pre-History (and Tradition)

Arguments about history and tradition played a more dominant role in incorporation debates surrounding the Fourteenth Amendment.⁶⁵ In the years following the ratification of the Fourteenth Amendment, the Supreme Court has eviscerated its Privileges or Immunities Clause, leaving litigants to rely on the Due Process Clause when challenging the constitutionality of state policies.⁶⁶ Competing ideas about the effect of the Fourteenth Amendment on the Bill of Rights emerged after the Court began incorporating parts of the Bill of Rights against the states in Gitlow v. New York in 1925.⁶⁷ While Justice Hugo Black maintained that the Fourteenth Amendment had only incorporated the Bill of Rights,⁶⁸ other Justices believed that the Due Process Clause was itself a source of fundamental rights.⁶⁹

1. Holmes and Moody

Justice Holmes’ reflections on history and tradition came into clearer view in one early case, Twining v. State of New Jersey, where he joined a majority opinion suggesting that the meaning of tradition shifted over time.⁷⁰ Twining, a bank manager facing misdemeanor charges, had declined to testify on his own behalf, and a prosecutor suggested that his reasons for doing so were less than pure.⁷¹ Twining protested, arguing either that the Fourteenth Amendment incorporated a right against self-incrimination or that such liberty wasn’t enumerated under the Fourteenth Amendment.⁷² Rejecting Twining’s claim, Justice William Henry Moody laid out three principles governing unenumerated rights. He wrote that such a right could be “ascertained by an examination of . . . the common and statute law of England before the emigration of our ancestors.”⁷³

But this historical inquiry was not the end of the matter.⁷⁴ Moody suggested that the meaning of history and tradition was subject to change, but only in rare circumstances when government action offended “fundamental principles of liberty and justice which lie at the base of all our civil and political institutions.”⁷⁵ Twining ultimately lost—because the Court did not think that the right against self-incrimination had adequate historical pedigree, to be sure, but also because the Court found no evidence that the relevant tradition had changed.⁷⁶

This vision of tradition echoed the conservatism of Edmund Burke, the Anglo-Irish theorist who loomed large in early conservative constitutional theory.⁷⁷ Burke celebrated the wisdom reflected in time-tested social practices more than whatever meaning could be derived from abstract theories or principles.⁷⁸ Yet, he squarely rejected the notion that tradition could not change. Tradition, as Burke saw it, did not change in response to “floating fancies or fashions.”⁷⁹ But its slow pace of reform is not the only reason that tradition worked as a welcome check on haste or revolution. Precisely because tradition was flexible is what invited a manageable and incremental kind of change that discouraged more radical ruptures with the past.⁸⁰ In Twining, Justice Moody implicitly embraces this Burkean conception of tradition.

2. Ozie Powell and Frank Palka (not Palko)

An evolving-tradition test circulated in the Court’s jurisprudence throughout the 1930s, though its contours remained fuzzy. In Powell v. Alabama, for example, Ozie Powell and four other Black teenagers had hitched a ride on a freight train crossing Alabama when they got into a fight with several white boys.⁸¹ Two white girls then accused Powell and his co-defendants of sexually assaulting them.⁸² The State tried Powell and his co-defendants without appointing defense counsel and sentenced all five to death.⁸³ Powell’s counsel responded that Alabama had denied the defendants due process of law by failing to appoint them counsel.⁸⁴ The Court recognized that, in the past, it had applied a primarily backward-looking history-and-tradition test, asking about “the settled usages and modes of proceeding under the common and statute law of England before the Declaration of Independence,” assuming they had not proven “unsuited to the civil and political conditions of our ancestors by having been followed in this country after it became a nation.”⁸⁵ Based on the evidence before it, the Powell Court concluded that the right to counsel failed this backward-looking test.⁸⁶ Nevertheless, the Court found that such a right was central to due process by focusing on “later cases,” including Twining itself.⁸⁷

By contrast, in Snyder v. Massachusetts, a man faced charges for murder and armed robbery, and prosecutors filed a motion requesting that jurors be brought to the gas station where the murder took place.⁸⁸ Snyder argued that defendants had an unenumerated right to tour crime scenes alongside jurors—a right deeply rooted in the nation’s tradition and history.⁸⁹ In rejecting Snyder’s claim, the Supreme Court likewise scoured decisions dating back to 1747 and said little about current practice.⁹⁰

But elsewhere, the Court reiterated that tradition was a living thing. In an incorporation case, Frank Palka (whose name would be forever misspelled as “Palko”) broke into a store, stole a phonograph, and murdered a police officer while making a getaway.⁹¹ A Connecticut jury convicted Palka, but the State appealed, invoking a Connecticut law permitting new trials when there had been an error “to the prejudice of the state.”⁹² Palka’s appointed lawyer, David Goldstein, argued that a second trial would violate Palka’s right to avoid double jeopardy.⁹³ In Palko v. Connecticut, the Court held that double jeopardy protections did not qualify as a “principle of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental.”⁹⁴ In reaching the conclusion, the Court said almost nothing about the constitutional past.⁹⁵ “Reflection and analysis” were all that was required—whether the Court could imagine an equitable justice system without such a right.⁹⁶

B. Post-War Family Planning

A more full-throated vision of an evolving history-and-tradition test emerged from conflicts about birth control—many of them unfolding outside the Supreme Court. Following World War II, Planned Parenthood combined arguments about the need to curb population growth with existing claims about the rights of married couples to plan their families.⁹⁷ But in the 1950s, it was segregationists rather than birth controllers who insisted that the past should guide constitutional interpretation—and that constitutional meanings were entrenched rather than evolving.⁹⁸ As Reva Siegel has shown, the famed Supreme Court litigator John Davis, who represented South Carolina in the Brown litigation, argued that the meaning of the Fourteenth Amendment was fixed at the time of ratification.⁹⁹ To determine the amendment’s original expected application, Davis pointed out that “23 [states] either then had, or immediately installed, separate schools for white and coloured children.”¹⁰⁰ Davis’s argument ultimately failed to convince the Brown Court. But other segregationists persisted in using state counting as a key method to establish constitutional meaning.¹⁰¹

1. “Tradition is a living thing.”

A different vision of the role for history and tradition appeared in conflicts about contraception at the Supreme Court. In the 1950s, Planned Parenthood lawyers hoped to challenge Connecticut’s anti-contraception law, the last in the nation to bar birth control by married couples.¹⁰² By 1961, when the Supreme Court was poised to hand down a decision on the law, the environment seemed ripe for a favorable decision. The FDA approved the birth control pill in May 1959, and within two years, one million people had used it.¹⁰³

The case that would become Poe v. Ullman had been winding its way through the Connecticut courts, and the plaintiffs in Poe told various horror stories about their lack of contraceptive access: one woman suffered three consecutive stillbirths, while a second nearly lost her life as a result of a foreseeably dangerous pregnancy.¹⁰⁴ Their attorney, Fowler Harper, made little mention of an unenumerated right to privacy. His argument instead mirrored an amicus brief from the Planned Parenthood Federation of America: the Connecticut law lacked a rational basis, forcing couples to use contraceptive methods that were “the most unreliable and least desirable.”¹⁰⁵ Nor did Connecticut mention history and tradition in defending the law, instead chastising Planned Parenthood for making an argument that was “sociological and physiological rather than legal.”¹⁰⁶

It was Justice John Marshall Harlan who raised the importance of history and tradition in his Poe dissent. The majority ruled that the plaintiffs had not suffered a redressable injury because they faced neither prosecution under the Connecticut law nor a realistic threat of one (after all, as the majority remarked, birth control was “commonly and notoriously sold in Connecticut drug stores”).¹⁰⁷ Justice Harlan disagreed. He said that the Court should identify unenumerated rights by consulting “what history teaches are the traditions from which it developed as well as the traditions from which it broke.”¹⁰⁸ The meaning of neither history nor tradition was fixed, in his view. “That tradition is a living thing.”¹⁰⁹ Harlan’s test pointed to a traditionalism that evolves. This change doesn’t come easy or rely on “merely personal and private notions.”¹¹⁰ But when a tradition did evolve, and when a new way of being fell into place, it was this more recent constitutional understanding that carried the day—in the tradition that married couples were using contraception to plan their families.¹¹¹

2. Constitutional Updating

Harlan’s opinion failed to carry the day in Poe, and his dissent did not command the change that he sought in social practices around privacy and birth control. But his proposed inquiry into evolving traditions raised a hard question about majority decisions that might influence such reform in the future. What if a judicial intervention itself nudges a tradition to change, or undermines a shift otherwise and already underway? Curtis Bradley and Neil Siegel observe that updating the Constitution in response to society’s changing needs “implicates an inherent tension between the benefits of customary evolution and centralized judicial review. Given the authority that federal courts possess in our constitutional system today, practice is likely to coordinate around judicial decisions.”¹¹² Won’t the recognition of salient rights like contraception and interracial marriage leave a mark on those practices? How can the Supreme Court identify an organically evolving tradition within American society if its own decision reworks the very content of that tradition or what it ultimately comes to be?¹¹³

Legal historians harbor well-founded skepticism about the degree to which the Supreme Court changes public attitudes or popular understandings of the Constitution, never mind the social traditions that might follow from those ideals. Scholars like Jeffrey Rosen, Reva Siegel, and Robert Post have documented the extent to which the Court tends to mirror or codify constitutional understandings, rather than transform them.¹¹⁴ They present robust evidence to reject a strong version of the transformation claim that the Court alone can dictate a new tradition. What of a weaker version? The first important thing to say is there is good reason to think the Court’s decisions have little meaningful effect on behavior at all—whether because most of its rulings don’t break through into the public conscience, or because, even if they do, most Americans may find the reasoning or result unconvincing.¹¹⁵

But two other possibilities emerge. First, the Court might sometimes reinforce or advance a shift that is already underway. Michael Klarman shows that seems to have happened with Brown, further legitimizing a new consensus in favor of integration.¹¹⁶ Also Obergefell, which might have done the same for same-sex marriage.¹¹⁷ On other occasions, backlash to a decision may push a tradition in the opposite direction. For example, the Massachusetts Supreme Judicial Court’s same-sex marriage decision in Goodridge v. Department of Health appeared to have triggered resistance to same-sex marriage, at least in the short term.¹¹⁸ Dobbs looks to have sparked a similar backlash.¹¹⁹ By the most reliable accounts, the incidence of abortions hasn’t gone down overall since trigger bans went into effect, with many Americans in ban states accessing pills by mail or crossing state lines for care.¹²⁰

In evaluating an evolving tradition, it makes little sense to disentangle the influence of an intervening decision that makes, at most, a modest contribution within a larger constitutional conversation. A decision might nudge changes in social practice in one direction or another. But that influence depends on the persuasiveness of its reasoning and the extent to which a court’s conclusion tracks what people already think. The evolving test needn’t stop its search for relevant practices when the Court hands down a decision that recognizes a new right or takes an old one away. Instead, it looks to the most current among longstanding traditions, which in turn incorporates how much, if at all, the Court ended up impacting that tradition and how.

3. Privacy and Penumbras

Harlan’s evolving-tradition test reflected this dialogic dynamism. It embodied the idea that social practices around contraception had changed, especially among married people, including Catholics.¹²¹ In the early 1960s, many believed that the Vatican would soften its opposition to birth control,¹²² and some Catholic leaders suggested that the church’s living traditions compelled them to do so.¹²³ By the mid-1960s, there were signs that lay Catholics made claims on tradition as well—over half of them in a 1966 poll favored federal funding for contraceptive programs.¹²⁴ While this dialogue took place mostly within Catholicism, its effects reverberated more broadly, for it had been Catholics who had most vigorously opposed the legalization of contraception since at least the 1930s.¹²⁵

This idea of an evolving tradition around contraception struck proponents of birth control as promising, and Harlan’s idea of a living tradition appeared to be a perfect vehicle for it. When Griswold came before the Court, Thomas Emerson and Katie Roraback, the attorneys for the Planned Parenthood League of Connecticut, argued that there was an unenumerated right to privacy that encompassed married couples’ use of birth control.¹²⁶ But how would a court define its limits? Citing Harlan’s dissent in Poe, Roraback and Emerson offered three guideposts for an evolving tradition test: “reason, tradition and current practice.”¹²⁷ Only proof of a new way of being, one that mirrored “current practice,” would carry the day because it would show that the relevant constitutional tradition had changed.

The Court’s decision in Griswold hardly settled questions about the utility of an evolving history-and-tradition test. Justice William O. Douglas’s majority opinion famously concluded that this right could be found in the Constitution’s penumbras.¹²⁸ The constitutional past seemed to matter in this analysis: Douglas stressed that marital use of contraception touched on a right to privacy older than the Bill of Rights, but the precise role for history or tradition was hardly clear.¹²⁹ Justice Goldberg’s concurring opinion centered on the Ninth Amendment and looked to history and tradition for guidance in fleshing out which liberties that amendment guaranteed.¹³⁰ By far the longest discussion of an evolving tradition test, however, came in Justice Hugo Black’s dissent.¹³¹ He argued that any evolving-tradition test required judicial activism.¹³² How could the Justices determine that a tradition had changed? Black quipped: “Our Court certainly has no machinery to take a Gallup poll.”¹³³

C. Sex, Abortion, and Cohabitation

After Griswold, the fate of an evolving-tradition test would intersect with a growing movement to reform criminal abortion laws.¹³⁴ In 1962, the American Law Institute (“ALI”) released changes to the Model Penal Code that included proposed reforms to criminal abortion laws, including exceptions for rape, incest, and certain fetal abnormalities and health threats.¹³⁵ By the mid-1960s, some state legislators had begun considering ALI bills, and antiabortion scholars, lawyers, and activists started to mobilize.¹³⁶

Leading antiabortion commentators insisted that respect for fetal life (and the criminalization of abortion at any stage in pregnancy) were deeply rooted in the nation’s history and tradition—and that this conviction had nothing to do with Catholicism, the faith practiced by most in the early antiabortion movement. At first, these arguments, too, echoed the idea of an evolving tradition. Antiabortion scholars did not limit their claims to the idea that Christian teachings, Western civilization, and the common law had always protected fetal life, even if some, like antiabortion scholar Eugene Quay, argued that “[p]rotecting the life of the unborn child has been a major concern of the earliest laws known to us.”¹³⁷

1. Catholics, Protestants, and Jews

Abortion-rights supporters questioned whether the nation’s history and tradition were as clear as abortion opponents claimed—in either the past or the present. In 1967, the Protestant Council of the City of New York and three Jewish organizations released a statement insisting the contemporary opposition to abortion did not reflect Judeo-Christian values or even national customs.¹³⁸ Harriet Pilpel of Planned Parenthood similarly explained in 1970 that “[t]hose of us who did not grow up in rigid religious tradition which decries all abortion . . . as murder, look at abortion . . . in terms of freedom of choice as to when and whether to have a child.”¹³⁹ The history and tradition of abortion remained contested as the conflict moved into the courts.¹⁴⁰

Significantly, both proponents and opponents of legal abortion at times framed tradition as evolving. In 1971, for example, Robert Byrn, an antiabortion law professor, sought to be named the guardian ad litem for all fetuses scheduled for abortion in New York City’s municipal hospitals.¹⁴¹ While claiming that rights for the fetus had been recognized since “time immemorial,” Byrn stressed that this claim constituted a living tradition, with emerging awareness of the rights and humanity of the unborn child in modern tort law, human rights law, and welfare law.¹⁴² Byrn thus conceded that constitutionally relevant traditions change over time.¹⁴³ In rejecting his fetal-rights argument on the merits, the New York Appellate Division defined history and tradition in a similar way, looking both to understandings at the time of the founding and the shifting meanings established in the present day.¹⁴⁴

2. From Poe to Roe

When the Supreme Court agreed to hear Roe v. Wade, those on both sides contested how a history-and-tradition test would apply to the constitutional status of abortion bans. Some antiabortion attorneys like Joseph Witherspoon, a member of the board of directors of the National Right to Life Committee, focused not on Poe’s evolving traditionalism, but on what he described as the original meaning of the Fourteenth Amendment.¹⁴⁵ In an amicus brief for the National Right to Life Committee, Juan Ryan, the long-standing president of the organization, likewise suggested that abortion could not be an unenumerated right unless there was an unbroken tradition supporting it.¹⁴⁶ Ryan argued that abortion “was always condemned at common law.”¹⁴⁷

In Roe, abortion-rights groups made their own claims about history and tradition. Planned Parenthood’s amicus brief in Roe sought to discredit the historical ancestry of fetal personhood, claiming that there was “only one instance other than the very recent abortion cases in which the contention has been made that a fetus is a person under the Fourteenth Amendment.”¹⁴⁸ An amicus brief submitted by a coalition of feminist organizations insisted that an evolving tradition test favored the recognition of abortion rights.¹⁴⁹ The Fourteenth Amendment, the feminist brief reasoned, required an analysis of “its full development and its present place in American life throughout the Nation.”¹⁵⁰

When the Court handed down a decision in Roe, the meaning of history and tradition lurked in the background of the debate between the majority and dissent. The majority acknowledged that medical customs around abortion had once been very different, culminating in abortion’s criminalization in most states during the nineteenth century.¹⁵¹ Critical to the majority’s telling was how the relevant tradition had changed: physicians had called for the reform of criminal laws, and abortion itself had become safer.¹⁵² Roe credited these changing practices and values in defining the scope of constitutional privacy.¹⁵³ By contrast, Justice William Rehnquist, writing in dissent, emphasized that the framers of the Fourteenth Amendment would not have recognized a right to choose abortion, but that’s not all.¹⁵⁴ He also suggested that the abortion right would fail an evolving history-and-tradition test. “Even today,” he said, “when society’s views on abortion are changing, the very existence of the debate [about abortion] is evidence that the ‘right’ to an abortion is not so universally accepted as the appellant would have us believe.”¹⁵⁵

3. City of East Cleveland

Harlan’s evolving history-and-tradition test reemerged in the Supreme Court in Moore v. City of East Cleveland.¹⁵⁶ Ordinances like the one in Moore, which zoned certain neighborhoods for nuclear families, formed part of an intensifying conflict about what the American Civil Liberties Union (“ACLU”) called sexual privacy.¹⁵⁷ A backlash to the sexual revolution in the 1960s and 1970s took aim at nonmarital cohabitors.¹⁵⁸ In 1972, twenty-eight states and several local communities criminalized cohabitation outside of marriage.¹⁵⁹ A more subtle strategy used family zoning ordinances like the one in Moore to exclude couples who did not wish to marry or were prevented from doing so by law.¹⁶⁰

This backlash belied fundamental changes in how Americans lived. Attitudes toward premarital sex were becoming more permissive, and premarital sex was becoming more common.¹⁶¹ By the 1970s, the advent of no-fault divorce meant that marriages were often temporary—the rate of no-fault divorces doubled¹⁶²—and the women’s liberation movement challenged a sexual double standard governing marriage while insisting that marriage should be unnecessary for women’s security or thriving.¹⁶³ The practice of living with non-nuclear family was increasingly common, even if states and cities sought to rein it in.

In Moore, Inez Moore was raising her grandsons, and her attorneys argued that the nation’s evolving history and tradition meant that East Cleveland could not use its zoning laws to exclude a family like hers.¹⁶⁴ Moore’s brief stressed that the extended family was an “institution that predates, and very likely will antedate, other legal and social institutions, such as the municipality.”¹⁶⁵ But contemporary recognition of the importance of the extended family was important too. “This practice of expanding the home to assist relatives,” the brief explained, “has continued to be an accepted tenet of our society.”¹⁶⁶ The ACLU’s brief in Moore asked the Court to go further, recognizing a substantive due process right to “determine basic family relationships and living arrangements,” as well as “broad rights to free association and privacy within the home.”¹⁶⁷

The Court voted to invalidate East Cleveland’s ordinance, and Justice Lewis Powell wrote a plurality opinion joined by three of the Court’s more liberal Justices.¹⁶⁸ Powell agreed with the ACLU that Moore’s decision to live with and raise her grandsons fell into the category of family rights protected under the Fourteenth Amendment.¹⁶⁹ Powell acknowledged that “[t]here are risks when the judicial branch gives enhanced protection to certain substantive liberties without the guidance of the more specific provisions of the Bill of Rights.”¹⁷⁰ At the same time, he made clear that traditions must change, and often did, but that the process took time and required the forging of new norms and consensus.¹⁷¹ The risks of judicial overreach were no reason for leaving unenumerated rights unenforced or artificially limited. Instead, the Court would look to an evolving tradition test, one that would show “respect for the teachings of history [and] solid recognition of the basic values that underlie our society,” while recognizing and honoring meaningful changes in tradition over time.¹⁷²

II. FORGING A CONSERVATIVE LEGAL MOVEMENT

A very different conception of history and tradition helped to bridge divides in an emerging conservative coalition. By the early 1980s the Federalist Society had formed and was rapidly expanding, with Ronald Reagan tapping the organization’s members to join his administration or nominating them to the federal judiciary. Originalism emerged as a foundational principle that guided the conservative legal movement’s approach to the Constitution and led the Reagan Administration to promise voters revolutionary changes in the law under the cover of neutrality. But originalism itself was not a perfect vehicle for the demands of the emerging conservative Christian bar or its counterparts in the antiabortion movement.

Conservative Christian lawyers, for their part, believed that the nation’s founding itself was explicitly Christian and wove religious teachings into the founding document—a claim that the Federalist Society was not prepared to make. A kind of Burkean history-and-tradition test did not suit conservative Christian lawyers either. Judges who adopted it had recognized the very rights that conservative Christians believed to be at odds with the nation’s founding character. For abortion opponents, arguments from original intent or original public meaning for fetal personhood were complicated because the framers of the Fourteenth Amendment said nothing about abortion, and because the antiabortion movement of the nineteenth century said nothing about the Constitution.¹⁷³

A looser focus on history and tradition—as opposed to original intent, original public meaning, or original expected application—had a unique appeal for social conservatives. For abortion opponents, looking more broadly at history and tradition could account for a range of other evidence beyond 1868 that included stigma surrounding abortion, criminal abortion laws that remained in place for nearly a century, and the like. For conservative Christians, a history-and-tradition approach could allow attorneys to weave in beliefs about a faith-based founding without explicitly tying interpretation to natural law or religious doctrine—and without imputing a desire to enforce Christian beliefs to framers who themselves had varied views of religion. This new history-and-tradition approach wasn’t evolving but entrenched. It assumed that the meanings of constitutional provisions were more or less fixed around the time of their ratification—and that later developments and constitutional understandings were entirely irrelevant.

A. Coalition and Constitution-Making

1. The Federalist Society

In 1968 and 1972, Richard Nixon had campaigned on criticism of the Supreme Court, which he painted as soft on crime.¹⁷⁴ Nixon’s comments galvanized conservatives who were critical of what they saw as unprincipled decisions of the Warren and Burger Courts. But in the 1970s, Nixon nominees hardly proved to be rock-ribbed conservatives. After all, it was a Nixon nominee, Harry Blackmun, who wrote the majority in Roe.¹⁷⁵

Ronald Reagan’s election in 1980 represented an unprecedented opportunity for the Federalist Society, which organized in 1983.¹⁷⁶ The founders and early members of the Federalist Society had any number of objections to the legal status quo. Some, aligned with business interests and libertarianism, sought to create a legal bulwark against the threat posed to business by progressive public interest groups like the Environmental Defense Fund.¹⁷⁷ Others had concerns about what they saw as race-based preferences or religious liberty.¹⁷⁸ The Federalist Society itself served as an umbrella for conservative lawyers that would “break what [members] see[] as the liberal control of many of the institutions of modern America.”¹⁷⁹

For Reagan, the Federalist Society offered a ready supply of smart and reliably conservative lawyers, both for positions in the administration and judicial nominations. Attacks on the Court delivered a potent political payoff.¹⁸⁰ On the campaign trail, Reagan channeled backlash to the decisions of the Warren and Burger Courts and vowed to stop the Court from “threatening to change traditional ways of life.”¹⁸¹ Reagan ran on a platform that included a commitment to nominate “judges at all levels of the judiciary who respect traditional family values and the sanctity of innocent human life.”¹⁸²

But before and after his election, the American Bar Association hammered Reagan for elevating ideology over qualifications, and the administration looked for an interpretive method—or set of traits—that would appear neutral and legitimate while ensuring the kinds of results that Reagan demanded.¹⁸³ As Reva Siegel has shown, aligning with the Federalist Society offered an alternative source of legitimacy for the judges the administration put forward. Originalism, in turn, commended itself to the Reagan Administration and the conservative legal movement as a way to frame the administration’s vision of the law and identify sympathetic judges, without calling for substantive and specific outcomes.¹⁸⁴

After Reagan’s re-election, the Federalist Society established itself as a pipeline to power: Edwin Meese, the attorney general, launched an attack on judicial activism at the organization’s national gathering in 1985; Supreme Court Justice Antonin Scalia was another prominent speaker.¹⁸⁵ Where conservatives had been disorganized and demoralized, the Federalist Society united them by appealing to originalism and judicial restraint—and by rallying against the influence of left-leaning legal scholars like the movement for Critical Legal Studies.¹⁸⁶ The Federalist Society claimed that the legal left had distorted the judiciary’s proper role insofar as “the independence of the Supreme Court [was] at risk of being subverted by political gamesmanship.”¹⁸⁷

The Federalist Society was an insurgent force challenging a legal orthodoxy that had been accepted by a broad spectrum of legal professionals for decades—and was epitomized by the putative legalism of the Warren Court.¹⁸⁸ Reframing that legal consensus as political allowed Federalist Society leaders to claim that they were not revolutionaries but stewards of the rule of law.¹⁸⁹

2. The Rutherford Institute

But antiabortion lawyers and conservative Christian litigators did not see originalism as ideal, at least not how Meese and his colleagues defined it. Antiabortion lawyers hoped eventually for a court that would recognize fetal personhood.¹⁹⁰ In recent decades, the movement promoted a wide range of arguments that reflected existing Supreme Court jurisprudence—the recognition of a new suspect class or a substantive due process right—more than it did textualism or originalism.¹⁹¹ Attorneys for conservative Christian groups, like the Rutherford Institute, believed that courts should adhere to the Christian principles they claimed animated the nation’s founding.¹⁹² An entrenched history-and-tradition test could reconcile these ambitions with the Federalist Society’s calls for judicial modesty.

A trial of this entrenched approach to history and tradition came in the form of a struggle over the fate of criminal sodomy laws. As Bowers v. Hardwick was making its way to the Supreme Court, an emerging conservative Christian public-interest bar changed the claims made on a history-and-tradition test.¹⁹³ The Rutherford Institute, which opened its doors in 1982, played a particularly important role in the creation of a fully entrenched version of the test.¹⁹⁴ There were other conservative Christian litigation firms when John Whitehead launched Rutherford, but Whitehead himself was renowned because of his success in Christian publishing.¹⁹⁵ Like other prominent conservative Protestants, Whitehead maintained that the Constitution was a Christian document.¹⁹⁶ The Fundamental Law, “clearly expressed in God’s revelation as ultimately found in the Bible,” had shaped the nation’s founding as well as the Constitution, which “presuppose[d] the Declaration [of Independence] and the higher, fundamental law to which the Declaration [bears] witness[].”¹⁹⁷

When Rutherford filed as amicus in Bowers, an evolving-tradition test might have served Rutherford’s aims: less than a third of Americans polled in 1986 opposed sodomy bans.¹⁹⁸ But Rutherford attorneys framed the nation’s traditions as both Biblical and unchangeable, and they wove this conviction into a vision for the history-and-tradition test.¹⁹⁹ Antiabortion lawyers had made a similar argument in struggles over end-of-life decision-making, insisting that legitimate substantive due process rights had to reflect the broader history of “Western Civilization.”²⁰⁰ Rutherford’s amicus brief argued that criminal prohibitions of sodomy had roots in “the Judeo-Christian Scriptures.”²⁰¹ Remarkably, Rutherford also acknowledged that attitudes toward same-sex intimacy had changed—in recent years, tolerance for same-sex intimacy had grown—if not impressively.²⁰² But this contemporary history did not matter, Rutherford insisted, for a deeply rooted tradition could not evolve.²⁰³

Originalism might have delivered the same result in Bowers as this kind of entrenched history-and-tradition test. The framers of the Fourteenth Amendment had drafted its text at a time when criminal sodomy laws were venerable and unquestioned. It would not be long before state lawmakers began expanding sodomy laws to cover oral sex, while applying them more selectively to same-sex intimacy.²⁰⁴ Looking at an entrenched history and tradition—which could include scriptural condemnations of sodomy—would likely have yielded the same result. But for Rutherford, the point was to advance an alternative that could take into account other Christian and “Western” values across the broader run of cases—and to advance a test that both the conservative legal movement and Christian legal movement could sometimes embrace.

The Bowers Court did not fully adopt an entrenched history-and-tradition test, even as the Court upheld Georgia’s sodomy ban.²⁰⁵ The majority noted that “[p]roscriptions against [sodomy] have ancient roots,” and that more than a dozen states criminalized sodomy, both at the time that the Bill of Rights was written and at the ratification of the Fourteenth Amendment.²⁰⁶ But the majority also emphasized that there was no new, tolerant tradition regarding sodomy: “24 States and the District of Columbia continue to provide criminal penalties for sodomy.”²⁰⁷ Justice Burger, who contended that condemnation of sodomy was “firmly rooted in Judeao-Christian moral and ethical standards,” picked up on even more of Rutherford’s claim.²⁰⁸

B. From Michael H. to Casey

Movement organizations were not very involved in the 1989 case of Michael H. v. Gerald D., but the history-and-tradition test was front and center.²⁰⁹ Conflict over that test continued to be central in constitutional struggles over abortion too. The Supreme Court whittled away at Roe in a series of decisions, including another from 1989, Webster v. Reproductive Health Services, which articulated the view that Roe’s trimester framework was hopelessly unworkable.²¹⁰ When the Justices agreed to hear a challenge to a Pennsylvania multi-restriction law, in Planned Parenthood of Southeastern Pennsylvania v. Casey,²¹¹ reproductive rights attorneys Kitty Kolbert and Linda Wharton argued that the history-and-tradition test required exploration of “evolving standards” for both liberty and equality.²¹²

1. Integrating the New Christian Right

Antiabortion lawyers, for their part, invoked Bowers as the prime example of an entrenched history-and-tradition test²¹³: they argued that abortion failed that test because it had been “a crime at common law.”²¹⁴ The plurality decision in Casey invoked Justice Harlan’s dissent in Poe and reinforced that tradition was a “living thing”—one defined not by a single moment in the past (or any one interpretation of that moment).²¹⁵ Casey described the Court’s own substantive due process decisions as helping to define past tradition, but said little else.²¹⁶

Casey, together with growing conflicts over same-sex marriage and end-of-life decision-making, increased interest in a history-and-tradition test as a mobilizing tool on the right. Before the second half of the 1990s, mostly single-issue antiabortion groups had fought for influence in the conservative legal movement, with key figures bridging the divide between the two. These included such figures as James Bopp Jr., the influential general counsel for the National Right to Life Committee tapped to head a new election law and campaign finance committee within the Federalist Society, and Leonard Leo, the Federalist Society attorney with deep ties to antiabortion activism and conservative Catholic voters who had been tasked with steering Clarence Thomas’s Supreme Court confirmation through Congress.²¹⁷

2. Alliance Defending Freedom

In 1994, Larry Burkett, a tycoon in the world of Christian financial consulting, teamed up with Bill Bright, the founder of the massively successful Campus Crusade, to develop a conservative Christian alternative to the ACLU that they would call the Alliance Defense Fund (“ADF”) (later the Alliance Defending Freedom).²¹⁸ Bright and Burkett convinced a group of leading televangelists in 1993 that conservative Christian litigators were massively underfunded—a conclusion that led to the founding of ADF, which would fund cases involving “religious freedom, the sanctity of human life, and family values.”²¹⁹ Central to this vision was a particular approach to history and tradition, one that fixed on a point in the nation’s supposed past while assigning Christian Scripture a foundational role in determining the Constitution’s meaning. “America’s founders were Christians who established our nation on Christian principles, deeply rooted in the Bible,” ADF leader Alan Sears explained in a 1996 fundraising letter.²²⁰

ADF quickly built up a formidable budget, desperately needed by even the best-funded antiabortion groups, and ADF’s claims about the persecution of Christians resonated with antiabortion activists angry about the prosecutions and fines of those seeking to block entrances to reproductive health facilities.²²¹ But most antiabortion groups remained loathe to stray from a single-issue framework, and members of the Federalist Society did not openly embrace ADF’s vision of a Christian Constitution. Rallying around a history-and-tradition test could overcome these differences.

C. Evolving and Entrenched Traditionalism

In the mid-1990s, a handful of states began passing aid-in-dying laws. Antiabortion attorneys argued that those laws violated the Constitution.²²² When the Supreme Court agreed to hear Glucksberg, antiabortion groups privileged arguments about history and tradition.²²³ “It is clear that a right to assisted suicide is neither implicit in the concept of ordered liberty nor deeply rooted in American history and tradition,” the National Right to Life Committee argued.²²⁴

1. The Battle in Glucksberg

The right-to-die advocates litigating Glucksberg stressed that Casey had not “removed history and tradition from the due process equation.”²²⁵ Instead,

Casey recognized that constitutional significance inheres . . . not in the historic legality or illegality of a specific act, but in the treatment, in our history and tradition, of interests like those in bodily integrity and autonomy, in avoiding pain and suffering, and in making profoundly intimate and personal life-shaping decisions.²²⁶

The ACLU’s amicus echoed this interpretation, stressing that the right of “a mentally competent, terminally ill person to choose an end to [physical] suffering” was “deeply rooted in [the] Nation’s history and tradition.”²²⁷

Glucksberg rejected constitutional claims against assisted suicide but did not clearly embrace an entrenched history-and-tradition test. Justice Rehnquist’s majority opinion juxtaposed living constitutionalism with an alternative focused on “fundamental rights found to be deeply rooted in our legal tradition.”²²⁸ And Rehnquist stressed themes of judicial restraint that conservatives had tried to associate with an entrenched traditionalism, suggesting that a focus on history and tradition could “rein in the subjective elements that are necessarily present in due process judicial review.”²²⁹

But Glucksberg hardly focused on just a few points in the distant constitutional past.²³⁰ The Court also stressed more recent support for bans on aid-in-dying, which had “been reexamined and, generally, reaffirmed.”²³¹ And Glucksberg reaffirmed that Casey had applied a history-and-tradition test.²³² “[T]he Court’s opinion in Casey,” Glucksberg reasoned, “described, in a general way and in light of our prior cases, those personal activities and decisions that this Court has identified as so deeply rooted in our history and traditions, . . . that they are protected by the Fourteenth Amendment.”²³³ This was a far cry from the evolving history-and-tradition test that Justice Harlan had set forth in Poe.²³⁴

2. From Lawrence to Obergefell

ADF used history and tradition to describe what its leaders understood to be divisive positions—for example, opposing heightened scrutiny for sex classifications—and to bridge possible divides with members of the Federalist Society and the conservative legal movement. Rather than denouncing sex equality or even feminism, ADF suggested that strict scrutiny for sex classifications would overturn deeply rooted laws recognizing “reasonable, time-honored differences between the sexes.”²³⁵ When it came to efforts to decriminalize same-sex intimacy, expand sex-education curricula, or bolster laws prohibiting same-sex marriage, ADF long framed its work as a fight to stop “homosexuals” from changing the law and using a “battering ram . . . to smash the biblical values on which our country was built.”²³⁶

By the end of the 1990s, these claims became politically toxic—and quite different from the kind of textualist and originalist arguments advanced by the Federalist Society. Polls showed growing tolerance for legalizing same-sex intimacy, and in the 1996 case Romer v. Evans, the Court struck down a Colorado state constitutional amendment for lacking a rational basis and codifying anti-gay animus.²³⁷ An entrenched traditionalism had operated to allow conservative movements to update what had become politically out-of-bounds claims. It also enabled them to frame the demands of the Christian legal movement and antiabortion forces in terms that the Federalist Society could accept.

Conservative Christians again sought to use arguments from history and tradition to create common ground in Lawrence v. Texas, a case involving a challenge to Texas’s sodomy ban.²³⁸ The State of Texas, for example, stressed that the nation had a tradition of “historically prohibiting a wide variety of extramarital sexual conduct.”²³⁹ Lawrence proved a devastating setback for proponents of an entrenched history-and-tradition test. Justice Kennedy’s majority began by casting doubt on the narrative of consistent hostility to homosexuality offered up by Bowers—as leading historians had documented, a focus on same-sex intimacy was relatively recent, as was an understanding of homosexuality as a stable identity.²⁴⁰ But the Lawrence Court suggested that it was the “laws and traditions in the past half century” that were of the “most relevance.”²⁴¹

Lawrence did nothing to decrease social-movement interest in arguments about history and tradition, especially as the fight for same-sex marriage continued. Groups like the National Organization for Marriage (“NOM”), founded in 2007 to preserve a California ban on same-sex marriage, presented their cause as a defense of “traditional marriage.”²⁴² In 2008, when the California Supreme Court held that the state constitution required access to marriage for same-sex couples, Maggie Gallagher, one of NOM’s co-founders, complained that the decision “brushed aside the entire history and meaning of marriage in our tradition.”²⁴³ In 2009, NOM moved its headquarters to Washington, D.C. and repackaged the group’s cause, as the Washington Post explained that year, to argue that “opposing gay marriage does not make them bigots, that the argument should have nothing to do with hate or fear, and everything to do with history and tradition.”²⁴⁴

Arguments about history and tradition also figured centrally in efforts to fortify the alliance between conservative Catholics, evangelical Protestants, and Orthodox Christians, spearheaded by Robert George, the Princeton professor who served as the chairman of NOM.²⁴⁵ George helped to draft and publicize the 2009 Manhattan Declaration, which identified traditional marriage, opposition to abortion, and support for a particular idea of religious liberty as the unifying demands of conservative Christians.²⁴⁶ The declaration, in turn, framed opposite-sex marriage as honored or even required by “Christian tradition, and . . . Western law.”²⁴⁷

History-and-tradition arguments also defined conservative movements’ strategies when a challenge to bans on same-sex marriage reached the Supreme Court in Obergefell v. Hodges.²⁴⁸ Some amicus briefs stressed that there was a history and tradition of allowing states to define marriage as they saw fit.²⁴⁹ Others suggested that only marriage between persons of the opposite sex was deeply rooted in the nation’s history and tradition.²⁵⁰ Obergefell embraced a kind of evolving traditionalism. Justice Kennedy’s majority underscored that “[h]istory and tradition guide and discipline” analysis of the Fourteenth Amendment, “but do not set its outer boundaries.”²⁵¹ What made this longer-term perspective so important, Kennedy said, is that “[t]he nature of injustice is that we may not always see it in our own times.”²⁵² But Obergefell would not usher back in the evolving traditionalism of yore.

An entrenched history-and-tradition test remained a priority for antiabortion lawyers, especially following the confirmations of Justices Gorsuch, Kavanaugh, and Barrett.²⁵³ And then, Dobbs elevated this entrenched species of traditionalism, alongside other key cases from the 2021–2022 and 2022–2023 Terms.²⁵⁴ But the meaning of a history-and-tradition test had, in fact, been a site of contestation for decades before the newly constituted conservative majority declared that “history and tradition” teaches that “what the Fourteenth Amendment means by the term ‘liberty’ . . . does not protect the right to an abortion.”²⁵⁵

III. REMAKING TRADITIONALISM

Recovering this lost approach to history and tradition is significant for three reasons. First, it sheds new light on old puzzles that have beset constitutional interpretation. These include charges of judicial activism and deep-rooted injustice, generality problems, and difficulties in differentiating between the traditionalist method and either “gloss” or “liquidation.” Second, a shared focus on longstanding and widespread practices promises to take the edge off interpretive divisions between the prevailing extremes of originalism and living constitutionalism, especially if looking to the way in which they have evolved over time. Third, distinguishing entrenched from evolving traditionalism provides a constructive way to clarify which rights merit constitutional recognition. The final Section details how these distinct approaches to traditionalism would go about analyzing a number of fundamental rights claims that the Supreme Court has already recognized (e.g., birth control, interracial marriage, medical refusal) as well as others that it hasn’t (e.g., assisted reproduction, aid-in-dying, gender-affirming care).

A. Doctrinal Puzzles

Traditionalism is criticized for being either too manipulable or too stuck in the past. These objections come from opposite directions. One casts this method of interpretation as so unmoored that it invites judicial activism. The other says it is so rigid that it entrenches injustice. The evolving approach to history and tradition proves less vulnerable to both criticisms than the entrenched version of that test. And there are a couple other doctrinal puzzles that evolving traditionalism makes meaningful progress on too. One relates to the levels-of-generality problem in determining the scope of constitutional rights and holdings. The other is about the complexities of distinguishing traditionalism from “gloss” or “liquidation.” The evolving kind of history and tradition isn’t immune to these challenges. But it fares better.

1. Activism and Calcification

The evolving test attracts charges of judicial activism because it doesn’t confine the inquiry to a preset timeframe in the distant past. And if a judge has more time periods to pick from, she has more discretion to pick and choose her favorite practices. But that analysis is still anchored in extra-judicial practices whose roots run deep. There’s no picking and choosing among isolated customs or fleeting fads—let alone freestanding normative principles or cost-benefit analyses. And this methodological point is critical: evolving history and tradition advises looking to the most recent among sufficiently longstanding practices. Accordingly, it offers the judge no wider variation than entrenched traditionalism.

Besides, fixing a traditionalist analysis on some earlier era isn’t determinate. It leaves ample space to exercise judicial discretion. For one, as Justice Amy Coney Barrett asked in the Bruen gun rights case, “How long after ratification may subsequent practice illuminate original public meaning?”²⁵⁶ The absence of any clear answer invites arbitrary line-drawing on either side. Adam Winkler catalogs the ways in which the Bruen “Court says that it is doing history and tradition analysis, but conveniently ignores any history it doesn’t like,” dismissing “all of the historical precedents for heavy restrictions on concealed-carry laws as outliers,” “early English common law as too old,” and “laws that were adopted after the mid-eighteen-hundreds as too young.”²⁵⁷ Joseph Blocher and Reva Siegel give the name “asymmetric updating” to the danger that Second Amendment rights expand to include new kinds of weapons, while the government’s regulatory authority stays frozen in 1791 or 1868.²⁵⁸ Entrenched traditionalism is vulnerable to cherry-picked traditions and badly done history too, while the evolving test can find expression in tangible practices grounded in objective, empirical evidence.

When it comes to the problem of intransigence, entrenched traditionalism is susceptible to injustices that reach further back in our past. All that matters is that a practice was accepted at the relevant ratification era—its animating rationale or subsequent rejection escapes scrutiny.²⁵⁹ So does the way in which tradition is deployed: to uphold a law that’s consistent with longstanding practices, to strike one down because it is at odds with them, or to interpret a provision that was enacted precisely to abolish a persisting evil like slavery. Consider the 2021 case Brnovich v. Democratic National Committee,²⁶⁰ which read into the Voting Rights Act a presumption that any practice is valid that was “standard” when the Act was enacted in 1965, even if it operated to deny black voters the opportunities afforded to others.²⁶¹ Justice Elena Kagan dissented on the ground that the purpose of the statute was “to eradicate then-current discriminatory practices, not to set them in amber.”²⁶²

Obergefell, the 2015 case affirming a right to same-sex marriage, underscores the injustices in an entrenched appeal to traditionalism: “If rights were defined by who exercised them in the past, then received practices could serve as their own continued justification and new groups could not invoke rights once denied.”²⁶³ Evolving traditionalism combats the reflexive appeal to “what we have always done,” when our past embodies grave wrongs. Longitudinal analysis about historical practices leaves space to recognize the ways in which traditions tend to grow progressively fairer over time. There is no guarantee that “what we have done more recently” ever and always bends toward decency and fairness. It is possible that more recent traditions are not especially more likely to be just. Later generations can certainly make moral mistakes too. What an evolving test enables us to do is avoid reifying the transgressions that we can see clearly when looking back in the rearview mirror, so that at least those we won’t perpetuate indefinitely into the future.

That’s what persuaded the Massachusetts Supreme Judicial Court (the state’s highest) in a 2022 case to grant due process protections to medical aid-in-dying under the state constitution. The Court identified two methods of traditionalism that it referred to as narrow and comprehensive: “a fundamental right may be determined either through a narrow view of this nation’s history and traditions or through a more comprehensive approach, which uses ‘reasoned judgment’ to determine whether a right is fundamental.”²⁶⁴

The Massachusetts high court adopted the “comprehensive” approach whose consideration of “modern precedent alongside history” enabled the majority “to cleanse our substantive due process analysis of the bigotry that too often haunts our history.”²⁶⁵ Social practices occupy a relatively weak place in this “comprehensive” conception of history and tradition. It holds that practices are neither a necessary nor sufficient condition of interpretative meaning. Evolving traditionalism reserves a stronger role for social practices. They give a judge a source of legal authority to cite when she agrees with it but also constrain her when she does not.

2. Levels of Generality

Generality problems pervade constitutional, statutory, and common law.²⁶⁶ They boil down to the description of a law, provision, or case at a lower level of generality or a higher one—in other words, in terms that are either more particular, and so would accordingly apply to a narrower set of future facts or questions, or stated as an abstract matter, which has relevance to a wider range of legal issues.²⁶⁷ This framing can have a profound impact on a legal conclusion. In 1937, legal theorist Jerome Hall put the problem like this: “Upon the level of generality selected for the criteria of likeness or dissimilarity depends the outcome.”²⁶⁸

The generality problem we focus on is the level at which a rights claim is articulated before the court, or in the understanding of its holding. It is a debate that Laurence Tribe and Michael Dorf launched in 1990,²⁶⁹ the year after Justice Antonin Scalia suggested that courts select “the most specific level at which a relevant tradition protecting, or denying protection to, the asserted right can be identified.”²⁷⁰ The question is how specifically or generally to identify a “relevant tradition”—for example, as allowing adulterous fathers to interfere with families, or letting genetic parents visit the children they learned of their biological connection to only at some point after their birth.²⁷¹ Marc DeGirolami argues that empirical appeal to the fact of social practices itself might help to rein in the choice of level by limiting the range of plausible answers to those that find expression in practices that are especially pervasive and longstanding.²⁷²

Evolving traditionalism can help to gain traction on the levels-of-generality problem. Take Obergefell. America has long respected the general framing of a liberty interest in two people getting married in cases like Loving v. Virginia, which recognized rights of marriage rather than “interracial” marriage.²⁷³ But the same interest might also be characterized in terms of a more specific framing about same-sex couples specifically. The question is whether two decades of same-sex marriage recognition in a growing minority of states is enough to root a tradition deeply. The answer depends on which practices endure most when the lens of tradition is refracted from the ratification era into more recently longstanding practices. The Obergefell majority reasoned about history not as a limit but as a guide that’s “entrusted to future generations” to protect individual freedom in a manner that evolves “as we learn its meaning.”²⁷⁴

By contrast, the reconstituted Supreme Court majority in Dobbs adopted a low level of generality to define a specific right to abortion, as that practice had been entrenched under nineteenth-century state restrictions.²⁷⁵ An evolving traditionalism that accounted for battles over birth control and medical refusal would have advised a more general right of reproductive freedom or healthcare. Distinguishing entrenched from evolving traditionalism also guards against selectively favoring whichever level of generality suits a preferred outcome: for example, describing the types of “arms” covered by the Second Amendment right at a high level of generality to include weapons unknown to the founding era, while casting gun control regulations at a low level of generality that requires laws which closely resemble past practice.²⁷⁶

3. Gloss and Liquidation

Distinguishing the evolving from entrenched tests for history and tradition also sharpens the differences between traditionalism and similar-but-distinct interpretive methods: gloss and liquidation. Historical gloss traces back to the separation-of-powers landmark Steel Seizure Case,²⁷⁷ specifically to Justice Felix Frankfurter’s concurrence, detailing that permissible sources of presidential power include “the gloss which life has written upon them” through “a systematic, unbroken . . . practice.”²⁷⁸ Gloss is just one data point among others, so relevant practices feature less prominently than they do in traditionalist interpretation, where social practices take center stage.²⁷⁹

Historical gloss more closely resembles evolving history and tradition in its analysis of relevant actions and inactions over time.²⁸⁰ The biggest difference is that gloss considers only government practices.²⁸¹ The history-and-tradition test widens the focus to salient and “systematic” practices by non-state communities too. Beyond legal advocacy, this could include widely shared primary conduct by unorganized groups of individuals, such as the growing use of contraceptives even among Catholics, or cohabitation of the sort cited in Moore.²⁸² By contrast to gloss, evolving traditionalism qualifies a social practice as relevant to recognizing a constitutional right, or declining to recognize it, when lots of people have undertaken or abstained from that practice for a long time.

Liquidation also goes beyond constitutional text to set the scope of indeterminate provisions in light of how political branches, states, or the people apply them after they’re enacted.²⁸³ Federalist No. 37 says liquidation is necessary when and because a newly ratified rule is “more or less obscure and equivocal.”²⁸⁴ This method features elements of both the evolving and entrenched versions of traditionalism. Liquidation is evolving in its openness to amplification by post-ratification practices that boast longevity and adherence. But then it fixes the meaning of that gap in the text through a period of acquiescence to a particular view in a process known as settlement.²⁸⁵

William Baude argues that liquidation tends to be achieved by people of great stature, working at the nerve centers of political power.²⁸⁶ By contrast, traditionalism finds the development of tradition in sometimes small places. And an evolving brand of that test looks for uniformity in the practices that cash out constitutional meaning across space and time. Liquidation also closes off interpretation to the possibility of future revision, without especially worthy justification. Whereas evolving traditionalism counts any change in longstanding social practices as meaningful to rights recognition.

B. Common Ground

The prevailing constitutional theories tend to talk past each other. Originalism and living constitutionalism appeal not only to different eras, but also to disparate values and core commitments.²⁸⁷ It is not that there haven’t been rigorous scholarly efforts to engage the other side.²⁸⁸ And yet these camps now find themselves more polarized than ever. They’re at odds about the Constitution’s meaning in most of the hard cases, that is, whenever the document is vague, silent, or cannot be substantiated with reliable historical context.²⁸⁹ That’s when they part ways for good, for example, about the role of original meaning (or original intent or law or methods). An evolving form of traditionalism holds out hope for bridging this gap, offering interpretive opponents a more conciliatory set of first principles and shared vocabulary.²⁹⁰

Most originalists think the meaning of each constitutional provision was pretty much fixed at the time that provision was adopted. So, it might seem reasonable to think that most would be reluctant to embrace social practices that were entrenched long ago in lieu of ones whose durability and pervasiveness evolved more recently. But there are other originalist commitments too. One is that constitutional practice constrains judicial decision-making. And this would seem to counsel openness to the external restraint that longstanding history and tradition places on this method of interpretation. The promise is more modest than any easy compromise. It is just the chance to bring constitutional antagonists to the same table to engage in meaningful conversation.

The evolving approach to history and tradition can be understood as a particular variant of living constitutionalism, defined as the idea that the Constitution’s authoritative meaning can and should be responsive to changing values absent any formal change to Article V.²⁹¹ Evolving traditionalism anchors itself in the most recent among those lasting social practices that go back at least decades, whereas most forms of living constitutionalism impose no such check on their interpretations that are grounded in changing circumstances and values.

The difference between entrenched traditionalism and originalism is subtler. Solum and Barnett diagnose their resemblance when they ask why Dobbs and Bruen fix the focus of traditionalism on historical evidence about the prevalence of state legislation around the time of constitutional ratification in 1791 or 1868. Their answer invokes originalism: firearms and pregnancy restrictions lend insight into what the public would have understood the Second and Fourteenth Amendments to mean when it comes to gun and abortion rights.²⁹² Solum and Barnett are right to observe that entrenched traditionalism and original public meaning can rely on similar evidence, even often. What distinguishes them, however, is the self-conscious attention each pays either to concrete social practices (traditionalism) or to abstract textual construction (originalism).

As for the entrenched and evolving variations of the history-and-tradition test, each focuses on different eras, either around the time of constitutional ratification or else extending for wider windows before and after that era. But both share minimal agreement on the interpretive importance of especially meaningful social practices.²⁹³ Those pervasive and entrenched practices are their North Star and point of departure. This first principle is what Justice Harlan described in his Poe dissent as the “regard to what history teaches are the traditions from which [this country] developed as well as the traditions from which it broke.”²⁹⁴ Compared with the entrenched approach to history and tradition, evolving traditionalism boasts the advantage of not forever preserving status inequalities and other past injustices, even if it might still risk ratifying contemporary evils.²⁹⁵

1. Shared Vocabulary

Traditionalism supplies a shared vocabulary that replaces abstract moral principles or contested balancing tests with tangible practices as the basis of constitutional meaning. These conversations about social practices can be mediated most effectively when tradition is treated as “a living thing,” in Harlan’s words.²⁹⁶ Practices that stand the test of time can form a shared way of talking about these questions. Analyzing the entire historical spectrum of tradition from ratification to present day moves litigants and courts into a common arena by eliminating temporal bounds and inviting inquiry into why practices came into being and stuck around, thereby shoring up their staying power. Even the most enduring practices may be subject to disagreement about their social meaning or moral value.²⁹⁷ And yet, as Harlan put it, “a decision which builds on what has survived is likely to be sound.”²⁹⁸

Entrenched history-and-tradition tests lack clear answers to the question about just how far back to go or even reach ahead up to the present. This temporal uncertainty emerges in the trilogy itself. In Dobbs, the majority’s historical analysis extends from thirteenth-century British common law until Roe was decided on January 22, 1973.²⁹⁹ Whereas in Bruen, the Court focuses on eighteenth- and nineteenth-century practices,³⁰⁰ in Kennedy, the Court looks instead from the Constitution’s ratification through that of the Fourteenth Amendment.³⁰¹ It is not just the trilogy, either. In McDonald v. City of Chicago and Timbs v. Indiana—two cases that Dobbs cited as examples of the history-and-tradition analysis³⁰²—the time periods for analysis also differed: McDonald covered ratification-era common law until the early nineteenth century,³⁰³ while the examination of relevant social practices in Timbs started with British common law and went not only up to the ratification of the Fourteenth Amendment but all way through to the present.³⁰⁴

2. Constitutional Trimming

An evolving history-and-tradition test starts from the first principle of enduring social practices, from before ratification until the present. Its shared vocabulary of tangible actions and inactions seeks to split the difference between the polarized margins of original meaning and living constitutionalism.

Evolving traditionalism satisfies some (albeit not all) of the considerations that move originalists to be originalists, and that attracts living constitutionalists to living constitutionalism. For originalists, traditionalism curbs judicial discretion by appealing to the longstanding social practices of politically accountable actors or the people they represent. It satisfies the popular sovereignty motivations of originalism through finding expression of the people’s will in diffuse practices over time. Meanwhile, living constitutionalists will find much to like in how evolving history-and-tradition isn’t rigidly fixed to meanings from long-ago eras very different from our own—and how it accedes to social practices that have embedded more recently based on the ethical understandings those practices reflect.

Cass Sunstein casts traditionalist interpretation as forging a constitutional compromise that demands concessions from both originalists and living constitutionalists, an example of what he calls “trimming.”³⁰⁵

Confronted with a disagreement between themselves and their critics, trimmers might conclude not that the original understanding is determinative, but that it is entitled to consideration, and that when precedents do not cut the other way, originalism should be followed. For this reason, some trimmers might be inclined to accept a form of “soft originalism,” giving weight to the original understanding without being bound by it.³⁰⁶

Shedding the least compatible of this ideological baggage can facilitate more discursive engagement on a meaningful common ground. Steering this middle course can be seen as a Goldilocks kind of exercise. Nobody wants to be entirely rigid or entirely fluid. Being too rigid makes you stiff, intransigent, and uncompromising, whereas being too fluid leaves you unmoored, floating, and aimless. DeGirolami argues that traditionalist methodology habituates judges to strike the right balance by thinking “in preservationist and custodial terms, asking them to extend and renew long-lived practices in an ongoing argument about the political virtues they serve.”³⁰⁷ An evolving kind of history and tradition holds particular promise to mediate between the extremes of originalism and living constitutionalism.

C. Remaking Rights

Beyond theoretical debates over constitutional jurisprudence, distinguishing evolving from entrenched traditionalism impacts the practical recognition of fundamental rights in profound and sometimes surprising ways. Fidelity to entrenched traditionalism would call into question rights that are clearly established and rarely contested—medical refusal, interracial marriage, and birth control. On the other hand, evolving traditionalism would not necessarily recognize new rights like medical aid-in-dying, assisted reproduction, and gender-affirming care.

That more dynamic approach also invites the possibility of an evolving right to treatments that the medical profession recognizes as standard-of-care, such as abortion, birth control, in vitro fertilization (“IVF”), and honoring advance directives, including for people who are pregnant when they fall into an irreversible coma.³⁰⁸

An important caveat is in order: the ultimate outcomes of an evolving history-and-tradition test will of course remain complicated and contested. And limited space here doesn’t permit working out definitive answers, especially in contexts as hard as these. The conclusions in this final Section are accordingly preliminary and tentative. The goal is simply to spell out the plausible operation of evolving traditionalism to give a sense of how this interpretive method would apply in practice. We start with the trilogy of Dobbs, Bruen, and Kennedy.

1. Abortion, Guns, and God

Dobbs concludes that “a right to abortion is not deeply rooted in the Nation’s history and traditions,” noting that, by the late-nineteenth century, thirty states “prohibited abortion at all stages except to save the life of the mother.”³⁰⁹ But most historians agree that abortion was in fact legal and common for most of American history, at least in the early stages of pregnancy, while half of the states that Dobbs appealed to did not make abortion a crime at all, let alone one that was prosecuted.³¹⁰ Selective history aside, Dobbs doesn’t reckon with the salient circumstance that, during the civil war era, women were deprived of now-uncontroversial freedoms that range from voting to property ownership.³¹¹ A plurality had put it plainly in the Frontiero pregnancy discrimination case that was decided the same year as Roe: “There can be no doubt that our Nation has had a long and unfortunate history of sex discrimination.”³¹²

An evolving traditionalism that extends beyond the age of coverture and disenfranchisement could find ample evidence of acquiescence in abortion. The Dobbs majority summarily dismisses the notion that even expanding the lens by another century to when Roe was decided in 1973 would provide social-practice support for a right to abortion. But the liberalization of abortion laws over that period makes it confusing to see how that pre-1973 state of affairs would not support a right to abortion. Nor does the majority even mention the fact of abortion practices that took place after 1973. One reason that Dobbs may have discounted post-1973 practices is that the Court’s intervention in Roe scrambled the signal, making it hard to read them off the public’s will. But the majority in Dobbs doesn’t say, or try to justify this move, beyond appealing to the incomplete history of traditionalism it misleadingly ascribes to Glucksberg.

On Bruen’s Second Amendment traditionalism too, the “definition of ‘arms’ is fixed according to its historical understanding,” yet “covers modern instruments that facilitate armed self-defense.”³¹³ Bruen looks into the nineteenth century, too, but dates the relevant firearm history back to the right of self-defense in the 1689 English Bill of Rights.³¹⁴ Beyond the right’s historic ties to state militias is the longstanding prevalence of trap-gun bans that reserved self-defense to one’s person—state legislation did not protect the use of firearms to defend one’s home or property.³¹⁵ The fixed-in-time account of traditionalism makes much of the gunpowder laws from the late-eighteenth century.³¹⁶

Only the evolving version of history and tradition can account for more recent kinds of gun control regulations. Consider the safe-storage laws enacted in the 1980s, designed to head off pervasive accidental gun deaths, and mass shootings or suicides by minors.³¹⁷ Yet, these are precisely the kinds of social practices that the entrenched traditionalism of Bruen closes its eyes to. There is a long history and tradition supporting Second Amendment rights. But lots of gun control measures that weren’t around in the nineteenth century have become extremely common for much of the twentieth, and not seriously contested until very recently.

In Kennedy, the Establishment Clause case, the Supreme Court again teaches that rights “must be interpreted by reference to historical practices . . . [that] faithfully reflec[t] the understanding of the Founding Fathers.”³¹⁸ Legislative prayers, school prayers, and Sunday observance laws are instructive.³¹⁹ Entrenched traditionalism finds considerable state intervention with religion in all three.³²⁰ There’s the revolutionary-era tradition of opening legislative sessions with prayers, dating back to the First Congress that hired a pastor to deliver them.³²¹ Then there are nineteenth-century norms of daily prayers and religious readings in public schools, accepted as a symbol of patriotic values and moral guidance to preserve a unified and peaceful society.³²² Also, Sunday closing laws were incorporated from the English common law during America’s founding for the purpose of observing Christian faith and facilitating church attendance.³²³

An evolving traditionalism might look very different. Deferring legislative prayers to outside clergy invites less risk of proselytizing or subtle coercion than the newer practice that has divided courts involving prayers led by lawmakers or other officials themselves.³²⁴ Public school prayer itself was held to violate the First Amendment in the 1962 landmark case of Engel v. Vitale.³²⁵ And Sunday closing laws have evolved to replace a religious purpose with a secular one to provide workers with a collective day of rest.³²⁶ Under an entrenched approach to history and tradition, ratification-era social practices deeply root all kinds of government involvement with religion. Only evolving traditionalism provides a determinate basis for rejecting such church/state entanglements.

2. Medical Refusal and Aid-in-Dying

The entrenched history-and-tradition test likely risks ruling out the uncontroversial right to medical refusal, while an evolving approach might not warrant recognition of aid-in-dying. In the 1989 Cruzan case, the Supreme Court recognized the right of competent persons to refuse unwanted medical interventions that extends to the denial of life-sustaining treatment.³²⁷ Cruzan itself appeals to the longstanding embrace of medical refusal in extrajudicial practices that include the common law tort principles governing informed consent and battery, state court landmarks, and state statutory law about vaccines, prisoners, children, and people with mental illness.³²⁸

Despite its rich lineage, the right to refuse life-sustaining treatment would fare poorly under the entrenched history-and-tradition test. When the Fourteenth Amendment was ratified in 1868, there were no ventilators or feeding tubes. The legal doctrine of informed consent itself wasn’t even invented until the twentieth century.³²⁹ Only a general freedom from nonconsensual touch was alive in the common law tort of battery. Neither would a right to assisted suicide be possible under an entrenched history-and-tradition test. It might not be under an evolving test either, at least not yet. The Model Penal Code barred assisted suicide well into the twentieth century.³³⁰ But the landscape has shifted in recent decades. Ten states have some form of legalized assisted suicide, whether legislatively or by voter initiative.³³¹

The scope of these practices remains partial and contested for now, even under the evolving test for history and tradition. But the pendulum is swinging. Evolving traditionalism would not have protected a right to assisted suicide in 1997, consistent with Glucksberg’s decision not to recognize that right. But if more and more states continue to codify a right to aid-in-dying under limited circumstances, that more adaptable history-and-tradition test could well protect that right some decades hence. That makes clear a significant implication of evolving traditionalism: depending on how prevailing social practices take shape, a future court applying this kind of history-and-tradition test could protect certain rights that a court today (or in the past) would and should say lack protection right now.

3. Birth Control, Sexual Intimacy, and Marriage

Justice Clarence Thomas, concurring in Dobbs, advocated “reconsider[ing] all of this Court’s substantive due process precedents, including Griswold, Lawrence, and Obergefell.”³³² This Section considers the entrenched-versus-evolving implications of each in turn, starting with the Court’s 1965 decision in Griswold to affirm a privacy right to birth control.³³³ That right would readily be vindicated under the evolving history-and-tradition test. By the 1930s, the contraceptive movement was large and visible, with birth control clinics opened across the country in a rebranding of what had once been deemed an immoral device now labeled as standard medicine.³³⁴ Contraceptives were widely recognized as essential healthcare and a social norm, even before Griswold and Eisenstadt v. Baird legalized birth control nationwide.³³⁵ But the birth control right may be unlikely to find solicitude under the entrenched approach to history and tradition. As the Dobbs dissent noted, “the American legal landscape in the decades after the Civil War was littered with bans on contraceptive devices.”³³⁶

A right to same-sex intimacy finds ample support under an evolving account of history and tradition, before such rights received recognition in Lawrence.³³⁷ In the era after World War II, dozens of activist groups formed to support sexual minorities.³³⁸ By 1955, the American Law Institute’s Model Penal Code advocated repeal of the sodomy laws.³³⁹ And in 1969, the Stonewall uprising inspired the formation of more than a thousand gay rights organizations nationwide, including the Human Rights Campaign, the Gay Liberation Front, and the Gay and Lesbian Alliance Against Defamation.³⁴⁰ The gay rights movement grew stronger still after Bowers.³⁴¹ In the 1990s and 2000s, state courts struck down several anti-sodomy laws under the equal protection clauses of state constitutions,³⁴² and the handful of remaining bans were rarely enforced.³⁴³

Entrenched traditionalism may look less kindly on the same-sex marriage right that Obergefell recognized under the Equal Protection Clause of the Fourteenth Amendment.³⁴⁴ In the era around the Civil War, sexual orientation was not a suspect class, and marriage was understood as being less about love and companionship than about raising the kids you had, something reserved to a man and woman.³⁴⁵ An evolving history and tradition, by contrast, would account for recognition of same-sex marriage in states like Vermont and Massachusetts by the early 2000s,³⁴⁶ and at least the benefits of civil unions and domestic partnerships in thirty other states in the following decade.³⁴⁷

Slavery and racial subordination are among America’s longest-standing “histories and traditions.” The constitutional reset that was the Reconstruction Amendments left pervasive redlining and institutional segregation, which extended to laws forbidding people of different races from getting married.³⁴⁸ The Supreme Court held that anti-miscegenation laws violated equal protection under the law in the 1967 case of Loving v. Virginia.³⁴⁹

An entrenched account of history and tradition might turn up conflicting evidence.³⁵⁰ On the one hand, anti-miscegenation statutes were on the books at some point in thirty-eight states by the end of the nineteenth century, and in twenty-nine still by 1951.³⁵¹ On the other hand, some states had begun to repeal their interracial marriage bans even before the Fourteenth Amendment, while others rejected proposals to write miscegenation bans into their state constitutions.³⁵² Only an evolving approach could account for the fourteen states that had repealed their prohibitions on interracial marriage in the decade or so before Loving—and the fact that sixteen bans remained up until that judgment, including Alabama’s until 2000.³⁵³

4. Fetal Rights and Assisted Reproduction

For the pro-life movement, the next major campaign after the demise of Roe will be a fight for fetal personhood—recognition that a fetal person has rights to due process and equal protection—and quite likely, a conclusion that liberal abortion laws are unconstitutional.³⁵⁴ Scholars from John Finnis to Josh Craddock argue that, as a matter of original public meaning, the word “person” in the Fourteenth Amendment applies from the moment of fertilization.³⁵⁵ Personhood claims have also captured headlines in contexts beyond abortion. An Alabama case in early 2024 involved several couples who sued over the wrongful destruction of embryos they had created during fertility treatment in the hopes of using them to have children.³⁵⁶ In LePage v. Center for Reproductive Medicine, the couples pursued a number of legal theories, including a negligence claim against the fertility clinic. The Alabama Supreme Court vindicated their action for wrongful death on the ground that the state’s Wrongful Death of a Minor Act applied to “all unborn children, regardless of their location.”³⁵⁷

LePage had the effect of pausing IVF in the state, as providers and intermediaries involved in the practice feared that even inadvertently destroying an embryo could lead to serious legal liability.³⁵⁸ The decision did not resolve the constitutional status of fetal personhood, even for the purpose of state law. But Chief Justice Tom Clark gestured toward a possible broader ruling in a concurring opinion, writing that “all human beings bear the image of God, and their lives cannot be destroyed without effacing his glory.”³⁵⁹ LePage made clear that recognizing personhood rights for the unborn would have implicated not just abortion but IVF as well. Conservative and progressive scholars have doubted originalist arguments for unborn personhood under the Fourteenth Amendment.³⁶⁰ At any rate, history and tradition may offer a more straightforward path to fetal personhood. Still, personhood proponents would struggle under an entrenched history-and-tradition test. For centuries, colonies, territories, and states allowed abortion before quickening, and British legal authorities dating back to Blackstone conferred the status of “natural persons” only after “an infant is able to stir in the mother’s womb.”³⁶¹ Even in the nineteenth century, when antiabortion doctors mobilized widely to criminalize abortion, no similar movement developed on behalf of fetal rights, whether at common law or under the Constitution.³⁶²

Surprisingly, an evolving history-and-tradition test would look kindlier on claims to fetal personhood. The 1960s is when a robust fetal rights movement began to take shape and forge a powerful political partnership with the Republican Party.³⁶³ A commitment to fetal rights, once dismissed as Catholic dogma, now holds sway among conservative evangelicals, Mormons, and some orthodox Jews.³⁶⁴ Support for fetal personhood has room to grow on the political right. Fetal rights still find too little support in social practices today, when polls show that the smallest fraction of the population approves of abortion bans early in pregnancy.³⁶⁵ So an evolving history-and-tradition test would not yet give champions of unborn personhood what they want. But history and tradition is their best bet in future constitutional struggles.

What of a right to assisted reproduction? IVF combines egg and sperm in a laboratory before testing the resulting embryo by implanting it into someone who can carry it to term. It is the most effective way to have biologically related kids for many single people, cancer patients, and infertile or same-sex couples.³⁶⁶ Many fear that this social practice will be forbidden if the law recognizes fetal personhood. In the nineteenth century, the notion of conceiving a child apart from sexual intercourse was far-fetched, let alone in some way that would make it possible to pick and choose certain aspects of offspring makeup before birth.³⁶⁷ IVF wasn’t invented until the late 1970s, and a right to access it fails under entrenched traditionalism.³⁶⁸ As early as 1897, the Catholic Church took a stand against any form of “artificial insemination.”³⁶⁹ Most physicians remained hostile to the idea of assisted reproduction during that time, while courts and commentators compared any use of donor sperm to adultery.³⁷⁰ A right to these and other forms of assisted reproduction would accordingly find next to no support under an entrenched form of traditionalism.

A right to assisted reproduction stands a better but still outside chance under the evolving test. Donor insemination has been relatively common since the 1930s and 1940s.³⁷¹ By 2018, a third of Americans either know someone who has used assisted reproductive technologies or have used it themselves.³⁷² Twenty-one states now require insurance coverage for infertility treatment, including fifteen that address IVF specifically.³⁷³ Over fifty percent of large employers (those with more than 20,000 employees) covered IVF in 2022, compared with only thirty-six percent in 2015.³⁷⁴ Still, a tradition favoring IVF access remains contested: federal legislation to enhance IVF access has stalled,³⁷⁵ and many patients live in states without IVF coverage or depend on state or federal Medicaid, which does not cover assisted reproductive technologies.³⁷⁶ Some antiabortion groups and their lawmakers in state legislatures frame IVF as lawless and immoral and have ambitions to restrict or ban it—an ambition that has become even clearer after LePage.³⁷⁷ A half-century of evidence doesn’t yet support a right to access IVF under the evolving-tradition test. But that may be changing.

5. Gender-Affirming Care and Conversion Therapy

The two history-and-tradition tests also have surprising implications for other due process claims. Two claims making headlines in the wake of state bans are conversion therapy and gender-affirming care. Parents have asserted due process rights to each form of medical treatment on behalf of their minor children—designed either to convert those youths’ sexual orientation from gay to straight, or to temporarily block puberty-related changes to breasts, muscle, and voice to buy time for adolescents who consistently identify as transgender to consider more permanent changes later in adulthood.³⁷⁸ The Supreme Court has held that the “interest of parents in the care, custody, and control of their children is perhaps the oldest of the fundamental liberty interests.”³⁷⁹ But the Court has also made clear that clinical decisions on their kids’ behalf must be weighed against the State’s own concern for the health and wellbeing of its vulnerable citizens.³⁸⁰ That leaves uncertain any rights claims by parents to access treatments for their children that the government has forbidden.³⁸¹

Entrenched traditionalism might look more favorably on parental rights to conversion therapy, notwithstanding overwhelming evidence that it harms the very children it purports to help. Conversion therapy traces to the Civil War era, when the medical profession widely regarded homosexuality as “either a criminal act or a medical problem, or both,”³⁸² and authorized parental attempts to “cure” it through invasive measures like lobotomy and castration that have since given way to hypnosis and talk therapy.³⁸³ Parental claims to gender-affirming treatments stand little chance under the entrenched history-and-tradition test. For one, hormones like estrogen and testosterone weren’t discovered until the 1920s and 1930s.³⁸⁴ That puberty blockers and cross-hormone therapy weren’t around when the Fourteenth Amendment was ratified is all the Sixth and Eleventh Circuit needed to conclude that “the use of these medications in general—let alone for children—almost certainly is not ‘deeply rooted’ in our nation’s history and tradition.”³⁸⁵

Under an evolving test, parental claims to gender-affirming care could stand an outside chance, while conversion therapy does not. Conversion therapy peaked in popularity in the 1960s when electroconvulsive therapy became increasingly common and the American Psychiatric Association briefly defined homosexuality as a mental disorder from 1968 to 1973.³⁸⁶ But healthcare organizations have roundly rejected conversion therapy over the ensuing decades as clinical evidence made clear that it is more likely to harm than benefit those distressed by a conflict between their sexual desires and their faith or family—especially if it reflects a disparaging view of same-sex attraction that inspires self-hatred.³⁸⁷ About half of the fifty states now ban the practice.³⁸⁸

The evolution of hormone therapy to affirm gender in minors has taken a different path.³⁸⁹ Puberty-blocking injections and implants emerged with programs in the Netherlands in the early 2000s.³⁹⁰ In 2009, the Endocrine Society published guidelines recommending the use of puberty blockers for adolescents whose distress, experienced from persistently identifying with a gender that doesn’t match their bodies’ sex-based traits, is not alleviated through psychosocial counseling and behavioral support.³⁹¹ By 2014, thirty-two clinics in the United States offered this treatment alongside parental support, with that number rising to about sixty by 2022.³⁹² Until 2023, gender affirming care was routinely provided to qualifying minors in forty-six states.³⁹³ But the months since have now presented a serious challenge for any rights claims under the evolving history-and-tradition test: the flurry of recent bans have left gender-affirming care legal in only thirty-one states.³⁹⁴

CONCLUSION

In declaring that the Supreme Court has “long asked whether the right is ‘deeply rooted in [our] history and tradition,” Justice Alito wrote that a history-and-tradition test emerged to constrain a more freewheeling living constitutionalism.³⁹⁵ His story takes off the stage entirely the heady days of substantive due process in the 1960s and 1970s that helped to forge the Roe decision. A history-and-tradition test, in turn, emerged in the 1980s and 1990s as the Justices retreated from the excesses of earlier decades. As Alito tells it, Glucksberg required the Justices to “exercise the utmost care whenever . . . asked to break new ground in this field, lest the liberty protected by the Due Process Clause be subtly transformed into the policy preferences of the Members of this Court.”³⁹⁶ In other areas of doctrine, the Court’s approach to history and tradition is different, but the common denominator is the same: to count, a tradition must be rooted in the ratification era. From there, it is timeless and unchanging.

This origin story is compelling. The problem is that it is not true: neither where the test came from nor what it is. For decades before Glucksberg, social movement contestation over the role of history and tradition had wrestled with the very meaning of that test. Progressive movements, at times, contended that the nation’s traditions evolved to include those once left out of old status hierarchies, and that any legitimate understanding of constitutional tradition included the present and recent past as much as it did Blackstone. Some of the Justices, in turn, came to define an evolving history-and-tradition test as a middle ground between a true living constitutionalism and a hidebound constitutionalism anchored only to one point in an imagined past.

Conservative movements, in turn, forged their own ideas of a history-and-tradition test, insisting that such an approach had always properly honored only traditions that reached back to the time of the nation’s founding—or before. Members of the conservative legal movement found this history-and-tradition test an effective way to bridge divides within an emerging coalition on the right; a way to reiterate shared commitments to judicial restraint while codifying deeply held beliefs and values on subjects from religious freedom to abortion.

Understanding the Court’s contemporary uses of history and tradition as the byproduct of social movement conflict reveals the other normative commitments that reliance on history and tradition disguises. Seeing it in historical context also calls into question the precedential pedigree the Dobbs Court claims for both its decision and its vision of traditionalism that overlooks a very different version of that test. And that one we have forgotten has profound implications for remaking the controversies of our time.

98 S. Cal. L. Rev. 1

Download

Dov Fox*

* Herzog Research Professor of Law, University of San Diego School of Law.

Mary Ziegler†

† Martin Luther King Jr. Professor of Law, University of California, Davis School of Law. Thanks for insights to Vic Amar, Niko Bowie, Joseph Blocher, Jud Campbell, Marc DeGirolami, Murray Dry, Sherif Girgis, Felipe Jiménez, Sally Gordon, Laura Kalman, Ken Kersch, Mike Ramsey, Eric Rueben, Stephen Sachs, Neil Siegel, Reva Siegel, and Aaron Tang. We are grateful to Sasha Nuñez and Liz Parker for editing support. Caroline Bulger, Jenna Gatto, Alyssa Guthrie, Emmanuela Kubari, Bailey Neal, Emily Serleth, and Joseph Wilhardt provided wonderful research assistance. We are grateful for expert edits and cite-checking to Charles Murphy, Jacob Karlin, and the staff of the Southern California Law Review.