The copyright merger doctrine is the rule that when an idea can be expressed in only one way or a small handful of ways, copyright will not protect that expression. The reasoning is straightforward: copyright never protects ideas, only the creative choices an author makes in expressing them, so letting someone own the only possible expression of an idea would hand them ownership of the idea by proxy. Rather than allow that, courts treat the expression as inseparable from the idea and leave both free for anyone to use.
The Idea-Expression Line the Doctrine Enforces
Federal copyright law draws a hard line between ideas and the way ideas are expressed. Under 17 U.S.C. § 102(b), copyright does not extend to any idea, process, system, method of operation, concept, principle, or discovery, no matter how an author describes or illustrates it.1Office of the Law Revision Counsel. 17 USC 102 – Subject Matter of Copyright: In General What the law protects is the specific creative choices an author makes when putting an idea into words, images, or sound. A novelist who writes about a detective owns that manuscript, not the concept of detective fiction. A photographer owns that specific sunset photograph, not sunsets.
For most creative work, this line is easy to hold. The same idea can be expressed in countless ways, and thousands of detective novels coexist without infringing each other. Merger enters at the opposite extreme, where the idea is so narrow or so constrained that there is essentially only one way to say it.
What “Merger” Actually Means
When an idea can only be expressed in one way or a very small number of ways, the expression is said to merge with the idea. Protecting that expression would effectively give the copyright holder ownership of the idea itself. Because the statute categorically forbids owning ideas, the law resolves the conflict by withdrawing protection from the expression. The first person to write down the only way to describe a concept does not get to block everyone else from describing it.
The Supreme Court laid the groundwork in 1879 in Baker v. Selden, holding that a copyright in a book explaining a bookkeeping system did not give the author exclusive rights over the system or the ruled forms needed to use it. Granting that kind of control, the Court said, “would be a surprise and a fraud upon the public,” because it would confer a patent-like monopoly without the examination patent law requires.2Justia. Baker v Selden, 101 US 99 (1879) That patent-versus-copyright concern runs through nearly every merger case since. If a particular arrangement of words is the only practical way to communicate a concept, treating it as copyrightable expression gives the author something closer to a patent monopoly without meeting any of patent law’s requirements for novelty or non-obviousness.
When Courts Find Merger
Courts do not apply merger as a yes-or-no checklist. The analysis usually starts by asking how many ways the idea could have been expressed. If the subject is complex and open-ended, expressive options are plentiful and merger is unlikely. If the subject is narrow, technical, or constrained by regulation, external standards, or the nature of the medium, the range shrinks and merger becomes plausible.
The most cited case is Morrissey v. Procter & Gamble Co., decided by the First Circuit in 1967. Morrissey held a copyright on a set of sweepstakes contest rules and sued Procter & Gamble for using similar language. The court held that the subject matter of the rules was so simple that the possible ways to express them were extremely limited, and warned that allowing copyright in such constrained expression would let “a party or parties, by copyrighting a mere handful of forms, could exhaust all possibilities of future use of the substance.” Copyright, the court said, should not become “a game of chess in which the public can be checkmated.”3Justia. Frank Morrissey v The Procter and Gamble Company, 379 F2d 675 (1st Cir 1967) The important point in Morrissey is that a court does not have to wait until every possible variation has been copyrighted before finding merger. If the range of expression is narrow enough, none of the variations get protection, including the first one written.
Physical objects can trigger the same result. In Herbert Rosenthal Jewelry Corp. v. Kalpakian, the Ninth Circuit held in 1971 that the idea of a jewel-encrusted bee pin and the expression of that idea were inseparable, because there are only so many ways to make a small piece of jewelry shaped like a bee. Granting copyright over one design would effectively confer a monopoly on bee-shaped jewelry, and, the court said, “a copyright must not be treated as an equivalent to a patent lest long continuing private monopolies be conferred over areas of gainful activity.”4Justia. Herbert Rosenthal Jewelry Corp v Kalpakian, 446 F2d 738 (9th Cir 1971)
Outside these examples, courts often examine the constraints the creator faced. Standardized forms used in insurance, banking, or medical intake frequently trigger merger analysis because their content is driven by regulatory requirements or industry practice rather than creative choice. When the format and language of a form are dictated by what the form needs to accomplish, the expression is not protectable.
Thin Copyright: A Middle Ground
Merger does not always work as an on-off switch. When a work contains some original creative choices but also relies heavily on elements dictated by subject matter or medium, courts sometimes grant “thin” copyright: protection so narrow that only exact or near-exact copying counts as infringement.
The Ninth Circuit took this approach in Satava v. Lowry in 2003. An artist who made glass-in-glass jellyfish sculptures sued a competitor over similar work. The court found that many elements, such as clear glass, a bell-shaped body, and flowing tentacles, either merged with the idea of depicting a jellyfish or were standard features anyone would use for that subject. Rather than deny protection entirely, the court gave Satava thin copyright, meaning he could sue only over work “virtually identical” to his own. The competitor’s sculptures were similar but not virtually identical, so no infringement was found.5Wikimedia Commons. Satava v Lowry, 323 F3d 805 (9th Cir 2003)
Thin copyright comes up frequently with works depicting things found in nature, functional designs, and compilations where the selection and arrangement are somewhat creative but heavily constrained. The work is still registered and technically protected, but the owner cannot use it to block anyone who makes something merely similar. The practical distance from having no protection at all is small, though it does still stop line-for-line copying.
Merger in Software
Software cases push merger into its most complex territory because computer programs are fundamentally functional. The code has to make the machine do something, and there are often only a few efficient ways to accomplish a given task.
The Second Circuit’s 1992 decision in Computer Associates International v. Altai created the abstraction-filtration-comparison test that most courts now use to analyze software copyright disputes. The court breaks the program into levels of abstraction, from general purpose down to specific code, and then filters out unprotectable elements at each level: ideas, merged expressions, elements dictated by efficiency or external factors like hardware requirements, and material from the public domain. Only what survives filtration gets compared to the allegedly infringing work. The court recognized that in programming, “the more efficient a set of modules are, the more closely they approximate the idea or process embodied in that particular aspect of the program’s structure,” making merger especially likely for optimized code.6University of California, Berkeley. Computer Associates International Inc v Altai Inc, 982 F2d 693 (2d Cir 1992)
In Lotus Development Corp. v. Borland International, the First Circuit held in 1995 that the Lotus 1-2-3 spreadsheet menu command hierarchy was an uncopyrightable “method of operation” under § 102(b). Users had to type specific command terms like “Copy” and “Print” to operate the software, and without those commands they could not access the program’s functionality at all. The court noted that merger reinforced the conclusion: when the ways to express a function are limited, the expression merges with the idea and even identical copying is permitted.7MIT OpenCourseWare. Lotus Development Corp v Borland International Inc, 49 F3d 807 (1st Cir 1995)
The Supreme Court had a chance to address merger directly in Google LLC v. Oracle America, Inc. in 2021 but sidestepped it. The Court assumed for the sake of argument that the Java API declaring code Google copied could be copyrighted and decided the case on fair use instead. The majority observed that “the fact that computer programs are primarily functional makes it difficult to apply traditional copyright concepts in that technological world,” while the dissent argued that merger, “merely an application of § 102(b),” should not apply because Oracle had innumerable ways to write the declaring code originally, even if Google had no other way to copy it for compatibility.8Supreme Court of the United States. Google LLC v Oracle America Inc, 593 US 1 (2021) The unresolved copyrightability question leaves merger’s application to software interfaces somewhat unsettled. Lower courts continue to filter out code sequences dictated by compatibility requirements and external specifications, but the Supreme Court has not endorsed or rejected that approach.
Merger Versus Scènes à Faire
Merger is often confused with a related doctrine called scènes à faire, and some courts blur the two. Both remove elements from copyright protection, but for different reasons.
Merger applies when the idea itself is so narrow that it can only be expressed in one or a few ways. The constraint comes from the nature of the idea. Scènes à faire applies when the author’s choice of a particular setting, genre, or situation makes certain elements practically inevitable. A courtroom drama will include a judge, a witness stand, and lawyers making objections, not because there is only one way to express the idea of a trial, but because those elements flow naturally from the chosen setting.
The practical distinction matters in litigation. A merger argument says the client had to use this expression because there was no other way to convey the idea. A scènes à faire argument says the client used these elements because they are stock or expected given the type of work. Both get filtered out before a court compares two works for substantial similarity, but they target different kinds of unoriginal content.
Merger at the Registration Stage
The doctrine does not only come up in lawsuits. Under Section 313.3(B) of the Compendium of Copyright Office Practices, the U.S. Copyright Office may refuse to register a work if protecting the author’s expression would effectively grant a monopoly over the underlying idea, process, or concept.9U.S. Copyright Office. Compendium of US Copyright Office Practices, Third Edition – Section 313.3(B)
The categories the Office typically refuses on merger grounds include bare story ideas built on familiar scenarios, such as a family quarrel followed by a reconciliation, without enough original expression added on top; code that uses only routine programming techniques any competent programmer would write essentially the same way; and works where expression naturally follows from the idea with little room for variation, such as a sculpture of a brightly colored jellyfish in a vertical orientation. A refusal on merger grounds is not necessarily the final word, and applicants can correspond with the registration specialist and argue that their work contains enough original expression to warrant protection.9U.S. Copyright Office. Compendium of US Copyright Office Practices, Third Edition – Section 313.3(B)
Who Has to Prove Merger in Court
Who carries the burden on merger is less settled than you might expect. Neither the Copyright Act nor most case law clearly assigns the burden of proving or disproving merger to one side. Some courts treat it as something the defendant raises; others view it as a built-in weakness in the plaintiff’s case that the plaintiff must overcome.
The stronger analytical view is that merger is not a true affirmative defense but a gap in the plaintiff’s case. A plaintiff suing for infringement has to show that the defendant copied protectable expression, and if the expression has merged with the underlying idea, it was never protectable to begin with. Under that view, the defendant can raise merger, but the plaintiff always carries the underlying burden of proving that what was copied qualifies for protection. The Ninth Circuit is the only federal circuit to have explicitly labeled merger a “defense” to infringement, though even that label may not shift the actual burden. In practice, a strong merger argument forces the plaintiff to demonstrate that meaningful creative alternatives existed, which is often the real fight.
Merger and AI-Generated Works
The doctrine has taken on new relevance as AI-generated content raises questions about copyrightability. In January 2025, the U.S. Copyright Office published a report on copyright and artificial intelligence concluding that prompts alone do not give users enough creative control to qualify as authors of AI output. The Office characterized prompts as “instructions that convey unprotectible ideas” that do not control how the AI system processes them.10Congress.gov. Generative Artificial Intelligence and Copyright Law
The reasoning echoes merger’s logic. If a prompt is essentially an idea, and the AI’s output is one of only a few possible expressions of that idea, then the output may be unprotectable for the same reason contest rules and standardized forms are. The Copyright Office has not framed its AI guidance in merger terms explicitly, but the principle is the same, and the boundaries of the doctrine are likely to be tested in this context in ways courts have not yet addressed.