by Ryan Lawson*
Should the copyright law doctrines of merger and scènes-à-faire be evaluated by courts as bars to copyrightability or as defenses to infringement? Ryan Lawson (’17) engages this question, based on his experience writing for the 40th Edition of the NYU Moot Court Board’s Casebook published in December 2016. There is currently a circuit split within the United States Courts of Appeals regarding whether the doctrines of merger — which exempts from copyright protection ideas that can only be expressed in a single or few ways — and scènes-à-faire — which exempts from copyright protection certain ubiquitous, foundational expressive tropes and devices — should be treated by courts either as bars to copyrightability, or as defenses to infringement. This Contribution argues that, in order to resolve this circuit split while preserving the flexibility and strength of these doctrines, courts should adopt a copyright law test that incorporates considerations from the related trademark doctrine of genericide.
There is currently a circuit split on the issue of whether merger and scènes-à-faire should be evaluated as bars to copyrightability or as defenses to infringement. The distinction determines which party bears the burden of proof during litigation and which party that burden favors. Because these defenses are important in many copyright suits and result in conflicting results among the federal courts, it is important that the circuit split be resolved.
In this article, I propose that genericide, a trademark doctrine, can be used to guide the analysis of when to apply merger and scènes-à-faire. By clarifying how a party can demonstrate that a copyright holder’s work is a product of merger or includes scènes-à-faire, parties will be better positioned to use these doctrines during litigation, and better able to adjust their behavior in accordance with expected outcomes.
The Copyright Act protects expression and not ideas, a principle known as the idea/expression dichotomy.2 Merger is a common law doctrine in copyright that “functions as an exception to the idea/expression dichotomy.”3 Merger doctrine recognizes that there are certain scenarios in which there are not enough ways of expressing an idea. The Second Circuit has explained that “expression is not protected in those instances where there is only one or so few ways of expressing an idea that protection of the expression would effectively accord protection to the idea itself.”4
Although every Court of Appeals recognizes merger as a common-law copyright doctrine, they are split over whether it is a defense to infringement or a bar to copyrightability. The Second and Federal Circuits apply merger as a defense to copyright infringement.5 By contrast, the Fifth and Sixth Circuits have determined merger stands as a bar to copyright protection in the first instance.6 The disagreement between the Courts of Appeals can be largely summarized as a difference between which party should bear the burden of proving/disproving merger and at what stage of the ligation it should be applied.7
Scènes-à-faire are tropes and other expressive devices that are pervasive or essential to certain categories of expression.8 For example, “the war between rival clans, developing love between children of the clans, [and] death and reconciliation after the rejection of revenge,” from Shakespeare’s Romeo and Juliet are considered scènes-à-faire, because the play has served as the inspiration for many other works and these elements have been used in other works countless times. Because these elements have become ubiquitous in literature, they are in the public domain and cannot be protected by copyright; they are scènes-à-faire.9
Although every Court of Appeals recognizes scènes-à-faire, they are—as in the merger debate—split over whether it is a defense to infringement or a bar to copyrightability. The Seventh Circuit recently held that “[t]he difference between the doctrine and copyright invalidity is vital to maintain.”10 By contrast, the Sixth and Tenth Circuits view scènes-à-faire as a bar to copyright protection.11
The proposed methods of dealing with merger and scènes-à-faire differ as to when courts should consider the doctrines. If they are bars to copyright protection, then the plaintiff’s choices at the time of original creation should govern whether merger or scènes-à-faire apply. For example, in Oracle v. Google,12 merger as a bar to copyright protection would depend on Oracle’s mindset regarding the inclusion of the Java programming modules. A court would need to determine what motivated the creation of the modules.13 If, on the other hand, merger and scènes-à-faire are affirmative defenses to infringement, then it is the defendant’s choices at the time of their subsequent creation that govern whether merger or scènes-à-faire should apply. Returning to Oracle, it would depend on why Google chose to use the programming modules in the Android operating system. The options open to Google, such as how standard the programming modules had become, would be controlling. Expert testimony would play an important role in determining the outcome of this analysis.
I propose that copyright doctrine should look to trademark law’s genericide doctrine for the answer on how to apply the merger and scènes-à-faire. One goal of trademark law is to support a system of easy identification of products so that consumers can find their preferred product with relative ease.14 However, one of the major problems with using trademarks as sources of information capital15 is that courts must protect against certain trademarks that would create unfair competition. One way courts do this is by prohibiting the registration of generic marks. An example of a generic mark is something like the phrase “apple” which “is a generic name for the edible fruit of the apple tree.”16 Accordingly, no entity could trademark the phrase “apple” with respect to the fruit.17 One simple test that courts have devised for determining if a mark is generic is the “Who are you—What are you?” test.18 Generally, if a mark answers the question “What are you?” then it is likely generic, because it is describing the product itself.19
A related problem is “genericide,” which occurs when a valid trademark becomes so ubiquitous in colloquial usage and its owner fails to police the mark adequately, with the result that the mark itself achieves generic status.20 Such a scenario is problematic for the holder of the generic trademark because they may be unable to ensure that consumers are using their mark as a source indicator rather than in its generic sense.21 In the case of “thermos,” the trademark holder, Thermos LLC, lost their trademark because consumers had begun to equate the word “thermos” with any vacuum-sealed liquid container.22 The theoretical underpinning of the prohibition of generic trademarks is fair competition: once genericide occurs, a trademark owner is precluded from continuing to utilize their mark, incentivizing them to enforce their trademark rights prior to the occurrence of genericide.23
Copyright law should adopt a fair competition analysis and apply a test similar to trademark’s genericide.24 Courts should begin by examining whether ideas and expressions had merged or scènes-à-faire existed in the relevant industry at the time of the secondary copyright owner’s creation. Instead of considering the choices that Google made, the Oracle court should consider market conditions at the time of Google’s incorporation of the modules into its own work. If at the time of Google’s creation of the Android operating system the Java programming modules had become an industry standard, then Google would not be liable for incorporating them. If, on the other hand, the modules were not an accepted industry standard, then Google’s use of them would be infringing. By adopting this approach, copyright law would give better deference to the way that industries have evolved to compete.25
A similar approach has already been applied in a seminal copyright case: Lotus v. Borland.26 In Lotus, the question was whether the menu command hierarchies in Lotus’ spreadsheet program were copyrightable. In reaching its decision that the menu command hierarchies were not copyrightable, the First Circuit reasoned: “Lotus wrote its menu command hierarchy so that people could learn it and use it. Accordingly, it falls squarely within the prohibition on copyright protection established in Baker v. Selden and codified by Congress in § 102(b).”27
Similarly, Sun—later purchased by Oracle—created Java programming modules that could be used by programmers easily and adopted widely. Although an analogue to trademark’s genericide in copyright would consider market conditions at the time of the secondary artist or defendant’s creation—and not the initial author’s intent—the Lotus case still focuses on a similar inquiry: whether the disputed mechanism has become so ubiquitous in the industry as to become a standard unworthy of copyright protection. Specifically in Lotus, the court discerned that the primary author’s intent was for the market to develop around its original creation and to incorporate it as a common element, and that the primary author was successful to that end.
Relying on author’s intent is somewhat controversial in many legal fields,28 but there are reasons for focusing on standardization within markets—to which an author’s intent or failure to take protective measures may point. Many technical markets rely on standard elements, and computer programs, which can be protected by copyright protection, are no exception. Google’s Android operating system incorporated a standard technological element by using the Java system, which “rapidly grew in popularity and now comprise[s] a large share of the United States market.” The reason that Oracle’s programming modules should lose copyright protection is because they had become widely used and their initial creators had not taken steps to protect them from infringement by others.30
There are several problems with applying the genericide doctrine in the copyright context. The first is a mechanical issue: the doctrine does not map onto copyright directly. In the trademark context, it is consumers who make a trademark generic by using it in a given market as a generic mark. One famous example is “thermos,” which became a generic mark for any vacuum insulated liquid container.31 In copyright, it is not as clear how expressive content would transform into scènes-à-faire or become the object of merger. Because authors control the content in copyrighted material, it is ultimately secondary artists who are responsible for turning something into scènes-à-faire or the object of merger. Like genericide, identifying the exact point when this occurs can be challenging. The programming modules in Oracle may be described as merged elements in the industry, but scènes-à-faire may present a trickier analysis. If secondary authors are responsible for determining when elements have become scènes-à-faire, many of these authors will be incentivized to promote works that arguably infringe on prior works but do not otherwise result in lawsuits, in order to ensure that subsequent authors have access to copying certain expressive elements through the scènes-à-faire doctrine. On the other hand, if market conditions control the analysis, then it seems unlikely that anything would ever become scènes-à-faire, because lawsuits would arise to prevent that from happening.
Notably, the original copyright owner does not lose all protection once a court has found that merger or scènes-à-faire applies. Instead, it is only scènes-à-faire or portions subject to merger that would be unprotected—for example, the Java programming modules in Oracle.32 Although this would represent a reduction in the value of some copyrights, a benefit might be that it would help eliminate complaints about the term of copyrights, which many view as being too long.33 Additionally, if a copyrighted work has become the object of merger or scènes-à-faire, this may indicate that the copyright has enjoyed significant popularity and the author has otherwise recovered fair compensation for their copyright monopoly in spite of the subsequent reduction in value.
To some extent, the copyright doctrines of merger and scènes-à-faire already incorporate ideas underpinning genericide, because they do not contain certain groups of expressions that are fixed in time, but instead evolve as some forms of expression are repeated and move into our common lexicon. However, the Courts of Appeals have diverged in their treatment of scènes-à-faire and merger, and it is likely that at some point the Supreme Court will resolve the issue. One way to ensure that these doctrines remain flexible but powerful would be to incorporate a form of genericide analysis within them. As a result, courts would place less emphasis on whether copyright protection preceded a subsequent work, and more emphasis on the nature of the relevant market at the time of the secondary artist’s creation. This may serve as an important limitation on the scope of copyright protections, given that the Supreme Court seems hesitant to reign in expansions in copyright terms.34
* This Contribution is based on my experience writing a problem for the 40th Edition of NYU Moot Court Board’s Casebook, published in 2016. The Contribution is a commentary on the copyright doctrines of merger and scènes-à-faire. I propose to simplify both doctrines through the importation of genericide doctrine from trademark law.
2. 17 U.S.C. § 102.
3. Oracle Am., Inc. v. Google Inc., 750 F.3d 1339, 1359 (Fed. Cir. 2014).
4. Kregos v. Associated Press, 937 F.2d 700, 705 (2d Cir. 1991).
5. See Oracle, 750 F.3d at 1358 (“[C]oncepts of merger and [scènes-à-faire] are affirmative defenses to claims of infringement.”); Kregos, 937 F.2d at 705 (“[T]his so-called ‘merger’ doctrine [is applied] in determining whether actionable infringement has occurred, rather than whether a copyright is valid.”).
6. See Lexmark Int’l, Inc. v. Static Control Components, Inc., 387 F.3d 522, 537–39 (6th Cir. 2004); Mason v. Montgomery Data, Inc., 967 F.2d 135, 138–39 (5th Cir. 1992).
7. See generally Andrew B. Hebl, A Heavy Burden: Proper Application of Copyright’s Merger and Scenes A Faire Doctrines, 8 Wake Forest Intell. Prop. L.J. 128, 148–50 (2008) (explaining that the battle between the Courts of Appeals is over whether the plaintiff or the defendant should bear the burden of applying the merger doctrine). See also Sandro Ocasio, Pruning Paracopyright Protections: Why Courts Should Apply the Merger and Scènes à Faire Doctrines at the Copyrightability Stage of the Copyright Infringement Analysis, 3 Seton Hall Circuit Rev. 303, 331–37 (2006).
8. See 2 Patry on Copyright § 4:24 (quoting Judge Yankwich from the Central District of California: “The French use a very expressive phrase in dramatic literature: “scenes a faire” that is, scenes which ‘must’ be done.”).
9. Reed-Union Corp. v. Turtle Wax, Inc., 77 F.3d 909, 914 (7th Cir. 1996).
11. See Lexmark Int’l, Inc. v. Static Control Components, Inc., 387 F.3d 522, 535 (6th Cir. 2004) (explaining scènes-à-faire does not play any role in the infringement analysis but is instead a bar to copyright protection); Gates Rubber Co. v. Bando Chem. Indus., Ltd., 9 F.3d 823, 838 (10th Cir. 1993).
12. See Oracle, 750 F.3d 1339 (Google wanted to incorporate the well-known and frequently used Java programming modules—which Oracle owned—into its Android operating system).
13. See Oracle, 750 F.3d at 1361 (“The focus is, therefore, on the options that were available to Sun/Oracle at the time it created the API packages.”)
14. William M. Landes & Richard A. Posner, Trademark Law: An Economic Perspective, 30 J.L. & Econ. 265, 268–69 (1987).
16. 2 McCarthy on Trademarks and Unfair Competition § 12:1 (4th ed.).
17. Of course, the phrase “Apple” can serve as a trademark in other commercial realms, as it does in the personal computing market. 2 McCarthy on Trademarks and Unfair Competition § 12:1 (4th ed.).
19. See U. S. Jaycees v. San Francisco Jr. Chamber of Commerce, 513 F.2d 1226, 1228–29 (9th Cir. 1975) (explaining the application of the “Who are you—What are you?” test with regards to the Junior Chamber of Commerce, which was found to be too generic to register as a trademark).
20. See 2 McCarthy on Trademarks and Unfair Competition § 12:1 (4th ed.) (explaining the process of genericide using examples such as thermos and aspirin).
23. See Wayne F. Osoba, The Legislative Response to Anti-Monopoly: A Missed Opportunity to Clarify the Genericness Doctrine, 1985 U. Ill. L. Rev. 197, 198 (1985) (explaining that the prohibition against generic trademarks represents one boundary beyond which trademarks would become too anti-competitive to be granted); Rebecca Tushnet, Why the Customer Isn’t Always Right: Producer-Based Limits on Rights Accretion in Trademark, 116 Yale L.J. Pocket Part 352, 353 (2007) (explaining that the goal of prohibiting generic marks is to improve competition). See generally Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976) (finding that depriving certain competitors of generic names would give too much competitive advantage to whoever held the generic mark).
24. See generally Jamie Lund, Copyright Genericide, 42 Creighton L. Rev. 131 (2009) (explaining how to apply genericide to copyright’s scènes-à-faire and merger doctrines).
25. This is slightly different than accusing an author of sleeping on his or her rights. An author whose copyright becomes the standard is not being punished because they have slept on their rights, because they have no right to assert against their copyright becoming scènes-à-faire in whatever area their copyright exists in. Contra Lund, supra note 24, at 136 (arguing that a copyright owner’s protected expression becoming scènes-à-faire is the result of sitting on their copyright and not asserting it).
26. 49 F.3d 807, 817–18 (1st Cir. 1995), aff’d, 516 U.S. 233 (1996) (holding that the copyright at issue was not allowed as a “method of operation”).
27. Id. at 817.
28. See generally Amy M. Adler, Against Moral Rights, 97 Cal. L. Rev. 263, 275 (2009) (explaining the relationship between artists’ intent and moral rights, and why disregarding the former strengthens arguments against the latter).
29. 750 F.3d at 1351.
30. Lund, supra note 24, at 147–49 (suggesting several steps Lotus could have taken to protect its program, including using patents to protect their software).
31. Am. Thermos Prod. Co. v. Aladdin Indus., Inc., 207 F. Supp. 9, 14 (D. Conn. 1962), aff’d sub nom. King-Seeley Thermos Co. v. Aladdin Indus., Inc., 321 F.2d 577 (2d Cir. 1963), and decree modified sub nom. King-Seeley Thermos Co. v. Aladdin Indus., Inc., 320 F. Supp. 1156 (D. Conn. 1970).
32. See Oracle, 750 F.3d at 1349.
33. 17 U.S.C. §§ 302–305.
34. See Eldred v. Ashcroft, 537 U.S. 186, 208–10 (2003) (explaining how Congress’ extensions of the copyright term do not constitute perpetual copyright).