Theft of the Commons: On David Bellos and Alexandre Montagu’s “Who Owns This Sentence?”

IN 1984, THEATER DIRECTOR JoAnne Akalaitis was preparing a production of Samuel Beckett’s Endgame for Harvard’s American Repertory Theater. The stage directions describing the play’s single set read: “Bare interior. Grey Light. Left and right back, high up, two small windows, curtains drawn. Front right, a door. Hanging near door, its face to wall, a picture. Front left, touching each other, covered with an old sheet, two ashbins.” This is what the set looked like when the play was originally produced in London in 1957. It was what the set looked like when the play was first produced in the United States in 1958. It was what the set always looked like. Most everyone alive or dead who had ever seen a production of Endgame had seen it played in an empty room with two small windows.

Akalaitis thought it might be time to try something different. Dramatic texts stay relevant because new generations of artists interpret them in new ways that make them speak to new audiences. In Akalaitis’s production, she set the play in an abandoned subway station. Far from bare, the stage was full of trash, including a broken-down train car, a choice some contemporary critics read as ingeniously concretizing the play’s metaphorical setting in a bunker after a nuclear holocaust. Akalaitis also deviated from tradition by incorporating music by Philip Glass into her production and casting Black actors in two of the play’s four roles.

Beckett was not amused. He threatened to shut the production down.

The author was able to exercise this authority because he held Endgame’s copyright, which includes the right to perform the work publicly. Anyone wishing to stage Beckett’s play when he was alive had to obtain permission from Beckett himself. Anyone wishing to do so today must obtain permission from the Beckett estate. Beckett had—and now his estate has—an effective monopoly on the plays, and he could use this power to condition permission to perform them on any criteria he liked, however arbitrary or capricious. Indeed, Beckett had not objected to André Gregory’s 1973 production of Endgame, which interpolated a variety of American colloquialisms and configured the hexagonal playing space so that the audience watched the action from four-person cubicles fashioned out of chicken wire. Nor did he object to the 1984 Brussels production Marcel Delval set in a flooded warehouse so that the actors had to perform in four feet of water. Though relatively restrained by comparison, only Akalaitis’s production prompted Beckett to take legal action. While he ultimately let the show go on, Beckett insisted on inserting a note into the program that read:

Any production of Endgame which ignores my stage directions is completely unacceptable to me. My play requires an empty room and two small windows. The American Repertory Theater production which dismisses my directions is a complete parody of the play as conceived by me. Anybody who cares for the work couldn’t fail to be disgusted by this.

Beckett is an extreme example, but other playwrights and their estates have imposed peculiar restrictions as well. Some playwrights refuse to allow roles written for men to be played by women, and vice versa. Some playwrights use their copyright to advance social issues by, for example, requiring that any productions of their play featuring disabled characters be cast with disabled actors. Some playwrights turn their copyrights into instruments of sublime spite, as Thomas Bernhard did by prohibiting the posthumous production of his plays in his native Austria until they enter the public domain.

Authors should not be able to wield this power from beyond the grave. There is no guarantee that an author’s estate will fall into the hands of people with the sense to manage it in a way that appropriately balances respect for the dead author’s wishes and receptiveness to the artistic impulses of living interpreters. Literary estates often end up under the control of family members who are accountants or actors or chocolatiers and have no real idea what their illustrious relations would have deemed to be a desirable afterlife for their work. Intellectual property lawyers, who are frequently relied upon to manage the details of licensing agreements, are not exactly renowned for being trusting, creative, open-minded types. Then there are the executors who are a little too creative, like the Roald Dahl Story Company, which recently permitted expurgated versions of books like Charlie and the Chocolate Factory (1964) and The Witches (1983) to be published scrubbed of words and ideas that might hurt sensitive 21st-century children’s feelings. The gluttonous Augustus Gloop, for example, is no longer “fat” but “enormous.” After a paragraph about witches being secretly bald, a new sentence is inserted: “There are plenty of other reasons why women might wear wigs and there is certainly nothing wrong with that.”

The ways in which copyright law inhibits artistic experimentation are just some of the many “wrongs” held up for scrutiny in David Bellos and Alexandre Montagu’s new book, Who Owns This Sentence? A History of Copyrights and Wrongs. Bellos, a translator and comparative literature professor, and Montagu, an attorney, take aim at the many absurdities and injustices engendered by our current intellectual property regime. The story they tell sweeps in much of recorded history, beginning with Plato—who, like his contemporaries, treated plagiarism as violating a code of honor—and ending with Bruce Springsteen, who recently sold his catalog to Sony Music Group for over half a billion dollars.

Along the way, we are introduced to a variety of negligent, vindictive, or grasping executors, such as Alexander Pushkin’s widow Natalya. Because she suppressed the publication of so much of her late husband’s work after he died in a duel fought over a slight to her honor, critics blamed her for “having killed Pushkin twice.” We also meet James Joyce’s grandson Stephen, who destroyed about a thousand family letters, refused to allow scholars to read many others, and insisted on the bowdlerization of his grandfather’s biography. And we encounter Martin Luther King Jr.’s heirs, who once licensed a recorded passage of the civil rights icon’s “Drum Major Instinct” speech for use in a Super Bowl commercial advertising a pickup truck, despite King warning in a different section of the very same speech against being “taken by advertisers” who seek to persuade their victims that “in order to make your neighbors envious, you must drive this type of car.”

Bellos and Montagu are right to point out that none of these abuses of copyright seem remotely congruent with the “incentive effect,” the widely held but entirely erroneous belief that the purpose of intellectual property is to provide creators with the external inducement they need to create in the first place. Nor do these abuses seem to “expand public knowledge and understanding,” which one judge called “the ultimate goal of copyright.” How did we get here? Once envisioned as a way to provide limited support to living creators, copyright now runs for the duration of the life of the author plus at least an additional 50 years in most countries.

According to Bellos and Montagu, copyright law assumed its present, compromised form through random fits and starts of industry lobbying, rather than through the coherent, principled public debate that shapes sound policy. While it is not clear in which of the world’s great deliberative bodies Bellos and Montagu are imagining all this principled debate to be taking place today, they give a compelling account of a legal framework arrived at through the shrewd, self-interested improvisations of people like George Garrard, an 18th-century English landscape painter who somehow convinced Parliament to pass a law rendering his diminutive plaster artist’s models of cows copyright-eligible. Bellos and Montagu see such “copyright creep and capture” as preposterous, and it is easy to understand why when they offer such hyperspecific examples. But the dynamics they describe are not really antidemocratic. True, not a lot of people cared about rival casters and molders ripping off their dummy livestock. If Garrard hadn’t taken it upon himself to petition lawmakers, there likely would have been no 1798 Act for Encouraging the Art of Making New Models and Casts of Busts. But formal and informal petitioning processes have long served important lawmaking functions by giving individuals and minorities a way of making their voices heard when they lack the numbers to do so meaningfully via the franchise alone. Of course, Garrard’s tiny cows were never going to be front-page news, but to the people to whom they mattered, they mattered a whole lot; Garrard’s success in obtaining protection for them is evidence of democracy in action, not dysfunction.

But individual interests do sometimes run counter to broader social considerations, and Bellos and Montagu convincingly argue that today’s excessively lengthy copyright terms constitute a kind of “theft of the commons,” a harm for which no one has the standing to seek redress. Work locked up in what Bellos and Montagu call “copyright prison” cannot be reproduced, or even confidently remixed or quoted, without a license, given the regrettably muddled state of the fair use doctrine. The risk of litigation has a chilling effect: artists refrain from using material they are (probably) entitled to use without a license out of fear of a costly infringement suit.

The fees associated with obtaining these licenses are what give rise to Bellos and Montagu’s real bugbear: inequality. They call intellectual property “the biggest money machine the world has seen,” one that makes rich countries richer and poor countries poorer through cross-border licensing. While the United States, they tell us, earns about $80 billion per year from licenses for its intellectual property, Afghanistan raised just $335 from licensing fees in 2020. These numbers are meant to shock and outrage, but it’s just silly to suggest that copyright is to blame for the desperate poverty of a country that has been ravaged by war and repression for centuries. Bellos and Montagu go on to call it “unbelievable” that foreign sales of US copyright licenses generate more revenue than the rockets we export. This fact may surprise us, but it is hard to see why it should outrage us. On the contrary, reading this passage, I felt an unfamiliar flicker of patriotism—why should thinking Americans not be proud that more of their country’s wealth is generated by music and storytelling than by peddling weapons of mass destruction?

More understandable is Bellos and Montagu’s complaint that, in practical terms, copyright today chiefly benefits corporations and not the artists responsible for creating the work. Since 1909 in the US, under the “work made for hire” doctrine, it is the author’s employer who often holds the copyright to the employee’s creations. Given the unequal bargaining power between most individual authors and their employers, individuals are seldom in a position to resist yielding their copyright. Musicians need record labels to produce, promote, and distribute their music. Film directors need studios to finance and wrangle the thousands of moving parts required to make a major motion picture. But as intellectual properties are bought and sold and traded among wealthy, rent-seeking corporations for years after the author’s death, the work-for-hire doctrine starts to look less and less justifiable.

Lively and intelligent, if intermittently unpersuasive, Who Owns This Sentence? should be of interest to attorneys, artists, and everyone in between. Those of us who were taught in law school that “the law is a seamless web” will enjoy watching Bellos and Montagu demolish that cliché. Those who pick the book up without knowing the first thing about copyright will receive a first-rate education on a range of topics, including significant US case law, Soviet copyright, and the “copyleft” movement.

But as flawed as our current system is, Bellos, Montagu, and other critics may soon find themselves nostalgic for a time when authors—individual or corporate—had any enforceable rights to their work at all. In the year or so since the release of OpenAI’s ChatGPT, Google’s Bard, and similarly ambitious, generative artificial intelligence systems, authors have been sounding alarms about the damage these technologies stand to do to their livelihoods and reputations. A watershed lawsuit filed by The New York Times against OpenAI in December is the latest in a series of cases claiming that the large language models powering these systems are illegally trained on copyrighted material. Existing copyright regimes will make it difficult to hold the makers of these technologies liable for infringement.

Jonathan Franzen, Sarah Silverman, and David Henry Hwang are among the authors who have filed class actions against tech companies for misappropriating their work for AI training purposes, but the Times case is the most significant challenge yet. OpenAI trained ChatGPT on millions of Times articles, according to the media company’s complaint, making the Times one of the most highly represented single sources of text in the system’s dataset. ChatGPT can easily be coaxed into helping users skirt the Times paywall by reproducing entire articles verbatim upon request. Perhaps worse, it is also prone to “hallucinating” Times articles that were never published, like one cited in the complaint that describes an authoritative-sounding (but nonexistent) scientific study linking orange juice consumption with cancer.

OpenAI will undoubtedly advance a fair use defense to escape liability, and it may well succeed. Our laws were not written with pseudo-sentient robot writers in mind, and commentators are fretting that AI could mean the end of copyright once and for all. The Times case and others like it may soon force a reimagining of intellectual property more radical than anything Bellos and Montagu contemplate.