Spoiler Liability Turns on Borrowed Expression, Not Revealed Facts

A studio may have good reason to dislike someone who reveals a climactic plot point before an audience has watched a show. That does not make the revelation a copyright violation. Eugene Volokh draws the central line sharply: copyright protects an author’s expression, not ideas or facts—including facts about a fictional story.

If someone says, “At the end of detective story X, the killer is Y,” that is a spoiler. It may destroy the suspense that would otherwise make people buy or watch the work. But it does not reproduce the work’s protected expression. The statement identifies a fact about the story: who the killer is.

That answer can feel unsatisfying because a spoiler can plainly reduce demand. Jane Bambauer presses a challenge to the conventional market-harm analysis: someone who learns an ending may lose interest in the original even though they were interested moments earlier. In her hypothetical, the disclosure can have something like a substitution effect because the audience receives information it wanted without paying to reach the climax through the work.

But that commercial harm does not itself create a copyright claim. The threshold question is whether the speaker copied protected expression. A bare plot revelation generally does not. The analysis changes when the speaker moves from saying what happened to borrowing dialogue, reproducing the sequence of events in expressive detail, or posting the scene itself.

The spoiler question sits inside a constitutional tension. Copyright is meant to encourage the creation and dissemination of books, music, films, television shows, and other works. Yet it does so by giving owners exclusive rights over reproduction, public performance, and certain derivative works. In practice, those rights can support an injunction against speech or damages for speech already made.

Eugene Volokh argues that this tension is not unique to copyright. Property rights generally constrain other people’s liberty: ownership of a beach can exclude people from using it, and ownership of a factory can exclude speakers from picketing on the factory’s property even when they may speak outside it. In ordinary First Amendment doctrine, people generally have no right to use someone else’s physical private property as a platform.

Copyright differs because the thing owned is expression. A right to control a literary, musical, or audiovisual work can restrain other people from reproducing it or from making new expressive works based on it. Someone writing a commercial “new Harry Potter” story might add original material, for example, while still infringing by using protected characters and details from the original. Fan fiction is one version of that problem; commercial exploitation is another.

Bambauer connects this to the ordinary economic rationale for property. Exclusive rights give people an incentive to invest in making, improving, and selling things. Intellectual property is non-rivalrous—copying a work leaves the original holder with their own copy—but the theory is similar: if creators and investors cannot reap the benefits of an exclusive right to sell, they may have less reason to bear the cost of creating the work.

The Constitution expressly authorizes Congress to promote the progress of “Science and useful Arts” by securing authors and inventors exclusive rights for limited times. Volokh notes that the phrase uses historical terminology that now sounds backwards: “science” referred broadly to knowledge and is associated with copyright, while “useful arts” referred to practical invention and is associated with patent.

That authorization does not mean copyright overrides the First Amendment. Congress’s other enumerated powers remain subject to the Bill of Rights; its commerce power, for example, would not authorize a ban on shipping particular books across state lines. The Supreme Court instead treats copyright and free-expression guarantees as provisions to be read together. Copyright’s traditional rules are understood as a means of fostering speech, even though they restrict some forms of speech along the way.

The Supreme Court’s explanation of that accommodation appears in Harper & Row v. Nation Enterprises, the 1985 case involving Gerald Ford’s memoir, A Time to Heal.

Ford’s publishers had granted Time magazine an exclusive right to publish excerpts before the book’s release. The Nation obtained an unpublished manuscript, described by the Court as “purloined,” and published an article about Ford’s account of his Nixon pardon. The article included roughly 300 to 400 words from the manuscript, and Time canceled its arrangement after being scooped.

The Court held that copyright liability was constitutionally permissible. Eugene Volokh identifies three strands in its reasoning, though he and Bambauer regard the first as weak: the suggestion that publishing Ford’s words interfered with his right not to speak, or to decide when to speak. Bambauer observes that reporting what someone has said or written ordinarily describes a fact about that person; it does not compel them to say anything. Volokh compares it to reporting something overheard from a politician, who cannot later invoke a general right not to speak.

The more durable rationale is economic. Copyright, the Court said, creates a marketable right in expression and thereby gives authors and publishers incentive to create and disseminate ideas. The Court also emphasized two “built-in” protections for speech: the distinction between ideas and expression, and fair use. In Volokh’s formulation, nobody has a monopoly in ideas or facts, and the fair-use doctrine is the other essential limit.

Those limits do not produce the kind of searching review that might apply to an ordinary speech restriction. Bambauer emphasizes that fair use operates as a defense, so a speaker may bear litigation costs before establishing that a use was lawful. The statute identifies criticism, comment, news reporting, teaching, scholarship, and research as examples of purposes that can support fair use, while requiring courts to consider four factors rather than applying a categorical exemption.^†

The Court has also been deferential to Congress on conventional copyright policy. In Eldred, it upheld a 20-year extension of copyright terms, from life plus 50 years to life plus 70 years, while leaving open only the possibility that a term extended effectively to infinity might pose a different problem.

For Volokh, that deference has limits at the edges. A copyright rule denying protection to anti-government or racist works, for example, might depart from traditional copyright principles in a way that raises viewpoint-discrimination concerns. But the Court has generally treated the idea-expression distinction and fair use as the mechanisms that reconcile traditional copyright with speech interests.

Copyright’s fact-expression line matters far beyond television spoilers. Eugene Volokh gives the example of a historian who spends years uncovering facts that reshape public understanding of an event. Later writers cannot copy the historian’s prose, and may not be able to closely imitate a protected selection or arrangement of material. But they can report the facts. Scholarship would be severely constrained if the first person to publish a fact could monopolize it.

The same principle applies to fiction. The proposition that a particular character committed a murder is not a fact about the real world, but it is a fact about the story. A reader can communicate it without appropriating the novel’s expression.

That is what separates a plot summary from the use at issue in Harper & Row. Had The Nation paraphrased Ford’s most newsworthy explanation for pardoning Nixon, Volokh says, it would not have infringed. It could have reported the bombshell fact in its own words. What made the case a copyright dispute was the reproduction of Ford’s original language.

Jane Bambauer questions whether the damages in Harper & Row were really driven by copied wording rather than by disclosure of the material readers most wanted to know. Volokh responds that quotation supplied verisimilitude: an article that merely summarized Ford’s account would give readers less confidence that Ford had really said what it claimed. That could make it less of a substitute for the authorized excerpt.

The distinction also keeps copyright separate from plagiarism. Plagiarism is an attribution error. Someone may reproduce substantial portions of another writer’s work with full attribution and avoid plagiarism while still infringing copyright. Conversely, someone may take uncopyrightable facts without attribution and violate academic or professional norms without committing copyright infringement. Bambauer notes that the distinction bears on debates over AI outputs, though they do not pursue that question.

Some appropriations of facts may still seem unfair. Volokh points to International News Service v. Associated Press, associated with the narrow “hot news” misappropriation theory: a limited doctrine concerning direct competition built on time-sensitive news gathered at expense. It does not create a general right in facts. But it shows why someone might imagine a court creating a narrowly framed spoiler tort even though copyright does not reach a simple revelation. The discussion identifies no existing general tort for interference with viewing pleasure.

Once a speaker copies expression rather than merely states a plot fact, fair use becomes the central inquiry. The statute identifies favored purposes including criticism, comment, news reporting, teaching, scholarship, and research, but the list is not exhaustive and no label automatically resolves the case.

Eugene Volokh describes fair use as a fact-sensitive balancing inquiry. Courts consider the use’s purpose and character; the nature of the copyrighted work; the amount and substantiality of what was taken; and the effect on the potential market for or value of the work. The factors do not operate mechanically, and Volokh stresses that judges must weigh them without a formula that tells them how much each should count.

A conventional review illustrates why copying is not always infringement. A reviewer may need to quote a poem, book, film, or other work to explain its language or artistic choices. The work reviewed may be creative, but the purpose is criticism or comment and the excerpts will usually be limited. A commercial review in a newspaper or magazine does not lose fair-use protection merely because the publication earns money.

Market harm is also not simply lost sales. A devastating review may persuade readers not to buy a book, but that does not count as the relevant harm merely because it changes their assessment of quality. Bambauer puts the distinction as substitution rather than a reduction in perceived quality.

A film adaptation of a novel points the other way. It is ordinarily commercial, takes substantial creative material, and occupies a conventional potential market for the book: film rights. The fact that an unauthorized adaptation might increase book sales does not excuse it. The author can choose whether and on what terms to license that market.

Parody provides a different example. In Campbell v. Acuff-Rose, 2 Live Crew’s rap takeoff on Roy Orbison’s “Oh, Pretty Woman” used recognizable elements of the original while changing its tone and message. The Court treated parody as closely related to criticism and comment because it comments on the original work. Commerciality was not decisive, nor was the use of substantial material, because a parody may need enough recognizable material to work.

Bambauer observes that a genuine parody often does not compete in the original market precisely because it mocks or rejects the original. Volokh adds that not every humorous use qualifies: a work is more plausibly parody when it pokes fun at the original itself.

The framework does not make every spoiler dispute easy. It does establish a sequence of questions: has the speaker merely conveyed a story fact, or copied expression; and, if expression was copied, does the use serve a purpose and take an amount that fair use may tolerate?

A one-sentence revelation of the ending generally remains outside copyright because it reports a fact about the story. A concise paraphrase that uses little or no dialogue is likewise generally reporting plot information in new language. The fact that an account is enough to ruin a viewer’s suspense does not by itself turn it into copied expression.

Quoted dialogue is different. The wording belongs to the work’s expression, so the inquiry shifts to fair use. The amount may be small, but the language could be important to the work, particularly if it delivers the reveal. The source’s publication status can matter as well.

A raw climax clip is more clearly expressive still. A final 30 seconds—or even five seconds—can reproduce performance, dialogue, image, sound, pacing, and the creator’s presentation of the climax. Its duration may be brief, but its importance may be great because it captures the payoff.

A clip posted without commentary has a weak claim to transformative purpose. It is literal copying of a creative work for much the same purpose as the original: allowing viewers to see the scene. It may not replace the full experience of a 45-minute program, but Volokh suggests it can substitute for the need to watch and pay to reach that particular conclusion.

A genuine review program that displays a twist and analyzes its framing or effectiveness is harder. It could argue that showing the surprise is necessary to explain whether the surprise works. It could also spoil the original for viewers who might otherwise have watched it. Volokh does not offer a definitive answer and says the hypothetical apparently has not been litigated. That uncertainty is a feature of a doctrine built from case-specific precedents rather than a rule that resolves every imaginable use.

The timing and source of a spoiler can change the analysis. A plot summary based on an unaired episode remains, in principle, a disclosure of facts rather than a copyright infringement if it does not copy protected expression. But unpublished status matters once dialogue, clips, or other expressive details are involved.

Eugene Volokh points to Harper & Row, where pre-publication use mattered because The Nation had scooped the authorized release of Ford’s memoir excerpts. The fair-use statute says unpublished status does not itself bar a finding of fair use; the factors still must be considered together. But Volokh suggests that someone borrowing dialogue from an unaired work may be entitled to borrow less than someone using material that has already aired. He treats the line as uncertain and fact-dependent, not as a categorical rule.

Bambauer raises an example involving spoilers for unaired Game of Thrones episodes that prompted takedown requests to a platform. Some disclosures reportedly described plot facts, while others came close to reproducing dialogue. That is the hard edge of the distinction: a plot description alone does not infringe, while even a few borrowed lines from an unaired episode may be more difficult to defend as fair use.

Volokh also suggests that a coordinated leak can raise theories that do not depend on copyright. Imagine that an employee at a production company leaks material to a prominent broadcaster under an understanding that the broadcaster will publicize it, share proceeds, or otherwise reward the leak. Depending on the facts, Volokh suggests the recipient could face a claim connected to the leaker’s breach of fiduciary duty, disclosure of trade secrets if the material qualifies as a trade secret, or other wrongdoing. Taking a tangible manuscript or other item could also raise theft or civil conversion issues.

That is different from arm’s-length reporting. If someone posts material and another speaker later reports on it without participating in the underlying breach, Volokh suggests those separate tort theories would be less clearly available. The remaining copyright question is the familiar one: what protected expression, if any, was copied, and can fair use justify it?