Category: Lit/Pub

The Big Deal in Mahmoud et al. v. Taylor is The First Amendment

first amendment

The U.S. Supreme Court last week heard oral arguments in Mahmoud et al. v. Taylor—a case brought by three families petitioning, on First Amendment Free Exercise grounds, to have their young children opt out of class time involving age-appropriate books that depict homosexual characters. The families—one Muslim, the other two Catholic—are not seeking to ban the books or to amend the Maryland elementary school curriculum. Instead, they are suing to overcome the school board’s refusal to allow an opt-out in this circumstance.

Opting out might appear to be a modest request, and Amy Howe’s Scotusblog summary of oral arguments describes the conservative majority of the Court as sympathetic to the parents’ petition. “What’s the big deal about allowing them to opt out of this?” asked Justice Alito. Although an unsurprising comment from Alito, the big deal raises conflicts the Court may be inclined to ignore—but it should not.

The Petitioners Misrepresent the Facts

The first big deal is that by allowing the petitioners to opt out on First Amendment Free Exercise grounds, this would make doctrine from an improper reading of the facts before the Court. In order to object to the books on the basis of the Exercise right, the parents should be required to show that the books promote an ethical or moral viewpoint that is not only anathema to their beliefs but also show how mandatory classroom exposure to that viewpoint is an abridgement of their right of Exercise.

But that bar cannot be met based on the content of the books at issue. “One book describes the story of a girl attending her uncle’s same-sex wedding, for example, while another book, Pride Puppy, tells the story of a puppy that gets lost during a Pride parade,” writes Howe. It may be the case that in the minds of the petitioners, the books imply or impose a moral position on homosexuality, but in fact, the books at issue simply depict the world as it is—a world that includes gay people.

If the books “promote” any viewpoint at all, it is tolerance in general, and it is extraordinary for these families to ask the Court to find that tolerance is tantamount to abridgement of religious exercise. It is clear that many religious people feel this way, but if the Court sanctions this view, it will erode the Framers’ purpose in writing the Exercise and Establishment clauses—namely that religious tolerance in a nation as complexly diversified as the United States is necessary for ensuring domestic tranquility.

So, if mere tolerance is unacceptable for certain families, it is unreasonable for those families to expect the public school to accommodate this view—let alone find that failure to make that accommodation is an abridgement of the Exercise right. The American public school has a responsibility to provide curricula and materials in literature, science, and history that reflect the world as it is—indeed failure to do so would violate the meaning of education—and the world includes gay people. It is, therefore, not the job of teachers or administrators at a public school to help certain parents shield their children from this fact any more than they would shield a child from a book depicting a prosperous Black character because some family sincerely believes that non-White people are subordinates in the “eyes of God.”

An Atheistic Hypothetical Makes the Point

Although my atheism is not conditioned on the fact that I find many religious people to be morally objectionable, I nevertheless sincerely believe that many religious people are morally objectionable. In fact, I could cite centuries of history to support that view in contrast to the dearth of evidence the petitioners in Mahmoud might bring to show a similar pattern among associations of gay people. Regardless, if the Court decides in favor of these families, would it likewise find that my secular child may opt out of a lesson in which a storybook merely depicts a family going to church?

If this hypothetical children’s book does not preach, for instance, that churchgoers are morally superior to non-churchgoers, then the book simply depicts the world as it is—a world in which some families go to church. In this case, would the Court find, as it should, that I am overreacting by seeking an exemption for my child on the false interpretation that the book’s mere depiction of churchgoing is implying that her secular family is morally wrong?

If the Court would rule against me in this hypothetical, then its finding should be no different for the plaintiffs in Mahmoud. Further, if my child gleans from the storybook that she is expected to tolerate, if not agree with, the religious kids in her class, that is both a sound lesson and one that is conducive to the atmosphere necessary for teaching a roomful of young children. This is especially true where young children come from a diverse range of families, including the likelihood that one may be the child of same-sex parents.

We Cannot Each Be a Law Unto Ourselves

Although opting out of a kindergarten class is a small matter—indeed, the parents could have done so without literally making a federal case out of it—asking the Court to grant the exemption on the basis that compliance with the public-school curriculum is an abridgement of Exercise conflicts with precedent, and possibly in a new and dangerous way. As the court held in Reynolds v. United States (1878), and Justice Scalia restated in Employment Division v. Smith (1990), the Free Exercise Clause is not “a system in which each conscience is a law unto itself.”

Reynolds and Smith address matters of state abridgement of specific exercise (polygamy in the first instance, and peyote ritual in the second) as a question of balance with “generally applicable law.” In response to Smith, Congress and President Clinton passed the Religious Freedom Restoration Act (RFRA) to reset judicial review and reverse Scalia’s holding in that case. RFRA was subsequently key in 2014 to then Judge Gorsuch’s 10th Circuit holding that Hobby Lobby, a privately held company owned by devout Christians, may be exempted from certain birth control mandates in the Affordable Care Act.

But between Reynolds and Hobby Lobby, in addition to the shift from natural persons to corporations engaging in “religious exercise,” we also see a contrast between direct prohibition of specific conduct (polygamy) to a more nuanced concept of abridgement (compliance with federal insurance law). But even Hobby Lobby is narrow in contrast to the implicitly broad exemption being sought in Mahmoud.

It is clearly the view of many religious Americans—especially American Christians—that mere toleration of fellow citizens outside their orthodoxy is perceived as an infringement of their Exercise rights. Or as The Authors Guild states in its amicus brief in this case, “Petitioners argue that the mere exposure to books that represent this community violates their free exercise rights under the First Amendment….we are now at an especially fraught moment in our history, as book removal initiatives—couched as attempts to shield children and young adults from harmful subjects and dangerous ideas—are wreaking havoc on school districts and public libraries across the country.”

The Court should refuse to endorse this agenda with a poor reading of the Exercise clause. Not only would it be bad law, but it invites pedagogical chaos in public education and animates viewpoint extremism in general. After all, what are the boundaries for public-school parents to opt out, grade-by-grade, topic-by-topic, belief-by-belief—in a society increasingly belabored by subjective truths? The family who clings to the geocentric model of the universe, for instance, is at liberty to home-school or find a parochial option that fits their beliefs, but the First Amendment does not permit them to demand that the public school accommodate their personal brand of scientific ignorance.

Thus, if we look beyond kindergarten and the Court finds for the plaintiffs in Mahmoud, the disarray it prefaces in public schools may only be resolved by an unconstitutional cabining of the opinion as an exception for certain religious exercise—perhaps just the three Abrahamic religions?—which would be in conflict with the Establishment clause. And quite possibly, that is the underlying intent of this lawsuit: to frame tolerance of diverse beliefs as intolerance of a few preferred beliefs until the latter overwhelms the former.

In that regard, intolerance itself is, of course, part of the world as it is, and as children mature, the literature they read often deals directly with various forms of intolerance in books that some Americans have tried to censor. Any sincerely held belief, whether religious or irreligious, will eventually breed some measure of intolerance for the heterodox view. Some Christians believe that I am an emissary of the devil, and, in turn, I believe those people are howling-at-the-moon insane.[1] There is no reconciling our views of one another, and so, the constitutional Framers, keenly aware that such antagonisms tear nations apart, wrote the Establishment and Exercise clauses with the hope of fostering comity in the new nation. So, the big deal in this case is that at the very least, the Court should honor history and tradition by supporting public school teachers working to foster comity in the classroom.

[1] Expression by Aaron Sorkin, The West Wing.

Photo by Peopleimages

Shedding Light: Briefs Filed in Kadrey v. Meta

kadrey

The purpose of cultivating works of authorship is to shed light on human experience, and the foundational purpose of the fair use doctrine in copyright law is to shed light on works of authorship. From its 18th century, English roots to the U.S. Supreme Court’s 2023 decision in AWF v. Goldsmith, the primary rationale for fair use is to permit the unlicensed use of works in ways that critique or comment upon the works themselves. Harvesting millions books to train an LLM does not do this.

With the growth of digital technologies and copyright protection for highly utilitarian computer code, fair use doctrine expands somewhat to permit certain “non expressive” uses of works. But these uses allowed by the courts have still tended to provide information about the works used or have been held to advance purposes like software interoperability. Harvesting millions of books to train an LLM does not do this.

A pair of briefs filed in Kadrey v. Meta—one by Association of American Publishers (AAP), the other filed by a group of IP law professors—present compelling arguments against finding that Meta’s unlicensed copying of millions of books to train its generative AI product Llama is fair use. A common theme in both briefs exposes a core fallacy, and legal hypocrisy, common to AI developers in these cases—namely that they copy protected “expression,” but they don’t copy protected “expression.”

As we see in the shorthand of social media, the developers write their own dichotomy by simultaneously humanizing and dehumanizing their products. In one breath, they compare machine leaning (ML) to human learning but then drop the analogy when they seek to claim that the protected “expression” in the works used is not copied or stored by their mysterious and complex “training” models. The AAP brief argues that copying “expression” is central to training an LLM, and the professors’ brief shows why “learning like a human” is precisely why fair use does not exempt Meta from obtaining licenses.

Both AAP and the professors naturally present specific arguments as to why none of the fair use case law supports Meta’s defense, but I was intrigued by the ways in which both briefs argue from different perspectives that training Llama indeed exploits the “expressive content” of the books appropriated. In fact, if it could be shown that no protected expression is copied or stored, this would be an argument that no case for infringement exists. But considering the emphasis on fair use—and all similar cases will almost certainly turn on fair use—we can assume that this statement from AAP is correct:

Meta would have this Court believe that authors’ original expression is not preserved in or exploited by the model. But this is not so. The LLM algorithmically maps and stores authors’ original expression so it can be used to generate output—indeed, that is the very point of the training exercise.

Kadrey and all AI training lawsuits with similar facts presented will turn on fair use factors one and four. Under factor two (nature of the works used), the books in Kadrey, and the works in most other cases, are “expressive” rather than “factual” in nature, and therefore, this factor favors plaintiffs. Under factor three (amount of the work used), it is understood that whole works have been fed into the LLM models, and so, this factor also favors plaintiffs.

Under the first fair use factor (purpose of the use), the court considers 1) whether the use is transformative; and 2) whether the use is commercial. Here, Meta’s commercial purpose is undeniable, and the AAP brief soundly argues that there is nothing transformative about copying the word-for-word expression in textual works for a purpose that sheds no light on the works used. On the contrary, the intent of the LLM is to create a non-human, substitute “author,” a purpose for which there is indeed no judicial precedent.

Factor four considers potential market harm to the copyright owner(s) of the work(s) used, and factor four may be the keystone in the broader creators versus GAI battle. Meta, a trillion-dollar company run by executives whose credibility is in doubt, contends that it is not feasible to license the books they used to train Llama. In response, AAP presents substantial evidence of licensing agreements between copyright owners and several major AI developers, and it states that Meta abandoned negotiations with publishers and chose instead to harvest books from pirate repositories.

Further, AAP argues “from a policy perspective” that Meta’s accessing those pirate “libraries” of DRM-free books militates against finding fair use in contravention of Congress’s intent when it passed the Digital Millennium Copyright Act (DMCA) in 1998. “Congress sought to establish a robust digital marketplace by ensuring appropriate safeguards for works made available online, including copyright owners’ ability to rely on DRM protections in distributing electronic copies of their works.”

In this spirit, inherent to the history of the fair use doctrine is the notion of “fair dealing” or, put differently, general legality in the overall purpose and character of the use. “The compiler of the training data’s knowledge of the unlawful provenance of the source copies might well taint the ‘character’ of the defendant’s use,” writes Professor Jane Ginsburg in a paper examining the question of fair use of works for AI training.[1]

The Professors’ Brief

The brief filed by the IP professors also emphasizes that the protected “expression” in the works is copied and exploited without license, but it also rather deftly uses Meta’s own rhetoric to doom the fair use defense. In general, when the AI cheerleaders say that LLMs “learn the way humans do,” my instinct has been to sneer at this anthropomorphic sentiment. But by giving the “learning like humans” analogy weight, the professors’ brief demonstrates exactly why that claim is fatal to a defense that the developer’s purpose is fair use.

Noting that humans indeed use protected works for “learning” all the time, the professors make plain that this exact relationship between author and reader (the basis for copyright) does not exempt the human from obtaining works legally. Thus, by Meta’s own analogy, the “machines learn like humans” claim is both an affirmation that the “expression” is being exploited and proof that that there is nothing transformative about using works for “learning.”

Further, the professors have a bit of fun emphasizing that Meta et al. strain to make the machine leaning process sound as technically complex as possible to obscure the fact that only by copying “expression” could the LLM actually “learn” anything. Here, a tip of the hat is deserved for the brief’s description of a human being reading a book thus:

… many billions of photons hit the book’s surface; some of those billions reached a lens, which focused them onto a retina, which converted them into electronic signals, which then resulted in electronic and chemical changes in some portion of over 100 billion neurons with over 100 trillion connections, some of those changes being transitory, and others more permanent.

The technical description of human processing and learning is even more mysterious because not even expert specialists in neuroscience know how the brain works at the neuronal level.

Well done! If that needlessly technical description of human reading requires legal access to the book, then so does the far less complex process of machine learning for AI development. Moreover, even if Meta were the vanguard developer and there were no examples of licensing deals being made, there is no rationale anywhere in commerce that a necessary resource must be free because it is essential. Meta et al. need electricity, engineers, and probably a computer or two to develop Lllama, and not one of these resources is free. Yet, somehow the most essential resource—the work of millions of authors—should be free.

On that note, there has never been a more important time to protect the rights and economic value of authors who shed light on the world we inhabit. I remain more than skeptical that it will ever be desirable to create literary works without authors, musical works without composers, etc. And certainly, licensing deals alone do not address all the potential hazards of unethical or questionable uses of generative AI. How products like Llama are used will provoke discussions that are cultural as well as legal. But for the moment, fair training of all AI models is the only rule that is both ethical and consistent with copyright’s purpose.

[1] Prof. Ginsburg is not one of the professors in the brief cited for this post.

Photo by Busko

Heroes and Villains in Copyright Fights

heroes

After Internet Archive (IA) lost its copyright infringement suit with major publishers this week, the organization wasted no time alleging that great harm has been done to society. As if it had the posts ready to go, IA alleged that research itself was in peril and even went so far as to shamelessly post on X that works by Orwell and Bradbury are now “no longer available”—as if its unlicensed repository provided the only access to 1984 or Fahrenheit 451.

If you don’t see the hypocrisy in citing those titles for the purpose of propaganda, you might as well not read the books. Funny, though, that literature is the subject because it is only human nature to consider even complex matters of law and policy through narrative. And because narrative requires heroes and villains, IA presumes to play the Rebel Alliance to the publishers Empire. One problem with this perspective is that if one insists the publishers are “villains,” then one must assume the authors are as well. Because here’s how things work in reality…

Every author owns the copyright rights in her book the moment the manuscript is finished. Whether she signs a deal with Random House or a small, independent imprint, she transfers at least part of her claim of copyright to the publisher in exchange for the publisher’s investment in producing, distributing, and marketing the book. Publishing agreements vary greatly, and sometimes, authors are disappointed. Nevertheless, most authors seek publishing deals rather than self-publish, and nobody commenting on the IA lawsuit should presume to tell authors that they are wrong to work with publishers.

Among the rights owned by the author is the right to “prepare derivative works.” With books, this means derivatives like translations, serials, motion picture adaptations, eBooks, and audiobooks. Typically, the author will transfer the right to prepare eBooks and audiobooks to the same publisher who produces and distributes the hardbound and paperback copies. Consequently, the author’s interest is aligned with the publisher’s interest in selling these electronic versions of the book. And quite often, the author’s share of audio and eBook sales is a higher percentage than her share in the sales of physical copies.

One way in which authors receive compensation from eBooks is through a variety of licensing regimes used by libraries around the country. Although there is no evidence that these licensing models hamper a library’s ability to serve its community, certain individuals, including IA founder Brewster Kahle, object to these licensing regimes as a matter of some principle they invented. That “principle” really comes down to the fact that they simply don’t like copyright rights, and Kahle et al. have made that very clear in public statements.

In fact, despite all the good IA provides with its database of hard-to-find works long in the public domain, Kahle’s fervent anti-copyright ideology seems to drive him to risk that worthy enterprise just to prove a point about which he is entirely wrong. In simple terms, IA et al. were determined to prove that libraries are entitled to prepare derivative works (i.e., make and distribute their own eBooks) instead of the publishers to whom those rights were assigned by the authors.

IA’s legal theories were so unfounded that the district court issued a judgment less than a week after oral arguments. And now that the appeals court has affirmed the obviousness of that judgment, IA is playing the victim on social media, like Icarus blaming the sun and gravity for his fate. In literary terms, we might recognize Kahle’s persistence against reason as that fatal flaw which can turn heroes into anti-heroes or villains. And wouldn’t it be a classic tragedy if the guy who wants to build the “New Library of Alexandria” managed to burn it all down in a grand display of hubris?

Illustration