ESG Law

Entertainment Sports
and Gaming Law

Back to Publications

Substantial similarities in copyright law: what Barbie v. Bratz can teach us about Pals v. Pokémon

By Yen-Shyang Tseng · Fri, Feb 2, 2024 7:32 AM

Substantial similarities in copyright law: what Barbie v. Bratz can teach us about Pals v. Pokémon1

Video games often take inspiration from one another. In 1992, Midway Games produced a fighting game called Mortal Kombat after Capcom’s Street Fighter II found success. Although Midway took the idea of a fighting game and produced its own, it did not copy Capcom’s expression of that idea. Mortal Kombat’s characters are different from Street Fighter II’s. The settings are different. The overall look and feel are different. In short, Midway produced Mortal Kombat without violating Capcom’s copyrights in Street Fighter II.

Allowing one to be inspired by another is good. It’s how we end up with game genres and numerous successful games within those genres. The MMORPG genre can be traced back to multi-user dungeons (MUDs) in the 1980s and early 1990s, many of which themselves take inspiration from Dungeons & Dragons, the first edition of which was released in 1974. The “Roguelike” genre comes from the 1980 game Rogue, which likewise took inspiration from earlier games. The “Soulslike” genre comes from the Demon’s Souls and Dark Souls games. Many popular esports titles today have similar origins–League of Legends took inspiration from Defense of the Ancients, which was inspired by Aeon of Strife, and Teamfight Tactics took inspiration from Dota Auto Chess.

But a distinction exists between taking an idea and taking the expression of that idea. And this brings us to Pocket Pair’s immensely popular game Palworld.2 In just the short time since its release, we have seen discussions around the game ranging from the use of AI to animal abuse in games.3 One of the many discussions sparked by Palworld is the question of copyright infringement.4 This article discusses one area of US copyright law–the substantial similarity test–as developed in the Ninth Circuit, in the context of Pals in Palworld and the titular pocket monsters of Pokémon.5 In particular, we explain how a court might use analysis from the toy industry and apply it to this type of case: it might find that the general concepts behind the designs of Pokémon are unprotected ideas, but it would likely find that specific design elements of those Pokémon are protected expressions of the ideas.

Copyright Infringement: the Basics

To show copyright infringement, a plaintiff must establish that it owns the copyright in the underlying work and that the other party copied original elements of the copyrighted work.6 A plaintiff can show the other party copied original elements of the copyrighted work by either (1) providing direct evidence of the copying or (2) showing that (a) the defendant had access to the copyright works and (b) substantial similarity exists between the two works, not of the ideas but of the expression of those ideas.7 Direct evidence may be found through admissions by a defendant or internal communications that are disclosed during the litigation discovery process8 showing the defendant copied the plaintiff’s work. This type of evidence is rare, so plaintiffs commonly rely on the second test.9

The Substantial Similarity Test Applied in Barbie v. Bratz

We assume that copyright ownership of the monsters in Pokémon is undisputed (as are Pokéballs and other expressions of ideas in the Pokémon franchise), and the only issue is whether Pocket Pair improperly copied original elements of those protected works.10 We also assume that it’s easy to establish access: Pokémon is the highest grossing media franchise in the world, and the first Pokémon games were released almost 30 years ago.11 I cannot see a plausible argument that the developers and publishers of a video game lacked access to Pokémon.

So the remaining question is substantial similarity: how similar are Pals in Palworld to certain monsters in Pokémon? The substantial similarity test has two parts, both of which must be satisfied. First, courts apply an extrinsic test that objectively examines concrete manifestations of the ideas and expressions in the two works. This examination focuses on whether “the specific details of an author’s rendering of ideas . . . standing alone, are substantially similar.”12 Specific details, or “protectible elements,” do not include “scenes a faire that necessarily result from the choice of a setting or situation,” “purely utilitarian elements,” or “elements of expression merged with the underlying idea.”13 Second, courts apply an “intrinsic test,” or a subjective evaluation – often by a factfinder (for example, a jury) – of the “total concept and feel of the works.”14

To show what these concepts mean, we can take a look at an example from the toy industry. 15Mattel, Inc. v. MGA Entertainment, Inc. – sometimes known as Barbie v. Bratz – was a sprawling litigation that lasted for more than 15 years between Mattel, the creator of the Barbie doll, and MGA, the creator of the Bratz doll.16 One of the many disputes between the two companies involved copyright infringement: Mattel alleged that it owned certain concept sketches and prototypes relating to Bratz under an employment agreement it had with Carter Bryant, the individual who developed those sketches and prototypes, and that the Bratz dolls produced by MGA infringed on Mattel’s copyrights in those sketches and prototypes.17

In one of the case’s many appeals, the Ninth Circuit applied the substantial similarity test to Bryant’s sketches and sculpts. Under the extrinsic test, the court found that unprotectable elements of those sketches and sculpts included “(1) the dolls’ resemblance to humans; (2) the presence of hair, head, two eyes[,] and other human features; (3) human clothes, shoes[,] and accessories; (4) age, race, ethnicity, and “urban” or “rural” appearances; (5) standard features relative to others (like a thin body); and (6) other standard treatments of the subject matter.”18 In addition, the Ninth Circuit court noted that “fashion dolls with a bratty look or attitude, or dolls sporting trendy clothing” were unprotectable ideas.19

Applying the Ninth Circuit’s standard, the district court found that the depiction of “a young, female fashion doll with exaggerated proportions” was an unprotectable idea, and the “large heads, thick lips, high cheekbones, slim arms, long legs, and slim torsos” of the dolls were unprotected features of the underlying idea.20 On the other hand, the court found certain protectable expressions of those underlying ideas, including: (1) the “precise shape, size, and placement of the ears”; (2) the “precise design of the nose”; (3) the specific “face shape”; (4) the “precise angles, measurements, and shapes of the sculpts’ midsections”; (5) the “uniquely rounded shoulders”; (6) the “way in which both sets of lower arms flare out from the body in an unnatural and unrelaxed states”; (7) the “noticeable indentation that exists between both sculpts’ shoulder blades”; and (8) the “dramatic S-shaped curve in both sculpts’ spines.”21 And, applying the intrinsic test, the court found genuine issues of material fact as to whether the protectable expressions of the Bratz production sculpts and several Bratz dolls were substantially similar to the protectable expressions of Bryant’s sketches and sculpts.22

Applying the Substantial Similarity Test to Palworld

This brings us back to the similarities between the Pals in Palworld and the monsters in Pokémon. @CeciliaFae on Twitter/X has gone through the entire list of 111 Pals existing as of the game’s release and compared them with certain Pokémon that the Pals most closely resemble, and identified specific designs that appear similar.23 For example, she notes that one Pal, Verdash, has “ears and theme of [L]eafeon and everything else of [C]inderace. The small hands, the fluff at the pants, all of it.” Others have analyzed the 3D models of Pals and Pokémon to determine whether similarities exist in the models.24

The Mattel v. MGA analysis relating to doll sketches and sculpts provides a clear framework for the way in which a court might undertake this type of analysis. Applying the extrinsic test, the general idea of Cinderace – a “bi-pedal, rabbit-like” monster with elements of soccer players (as generally described by Bulbapedia, a community-driven Pokémon encyclopedia) – might be an unprotected idea, as might be the fact that Cinderace has general features resembling soccer players such as hands and feet and general features resembling rabbits such as long ears. But the expression of that idea, including the specific design of Cinderace’s hands and the fluff at the pants (as pointed out by @CeciliaFae), are protected elements.

Under the intrinsic test, a court might find genuine issues of material fact and ask a jury to determine whether substantial similarities exist. This could involve direct side-by-side comparisons as well as 3D modeling to address similarities in the precise shapes, sizes, placements, angles, measurements, and other elements of Pals and Pokémon. From my perspective, reviewing the comparisons others have made, it appears that the design of Verdash has elements substantially similar to the designs of Cinderace and Leafeon. The designs of many other Pals also bear substantial similarities to the designs of certain Pokémon; compare, for example, Azurobe to Serperior and Primarina, Jetragon to Latios and Latias, Lamball to Wooloo, Boltmane to Luxray, and Direhowl to Lycanroc. Testimony from experts in graphic and game design, who can make these comparisons at a professional level, would be key to convincing the jury in either direction.

Conclusion

Ultimately, it’s up to The Pokémon Company to investigate the issue and the strength of a legal claim against Pocket Pair and determine what it will do with the results of that investigation. On January 25, 2024, The Pokémon Company (Japan) released a statement regarding its intent to “investigate and take appropriate measures to address any acts that infringe on intellectual property rights related to the Pokémon.” As of this article, any action remains to be seen.


  • 1.

    Disclosure: As of the date of this article, The Pokémon Company International is an active client of ESG Law LLP.

  • 2.

    See Miles Klee, ‘Palworld’ Is a Smash Hit. Gamers Claim It’s ‘Pokémon With Guns’ (Jan. 22, 2024), https://www.rollingstone.com/culture/culture-news/palworld-pokemon-steam-game-accused-copying-1234951508/.

  • 3.

    See Harvey Randall, AI Palworld accusations are currently based on vibes, a social deduction game, and the CEO's past tweets about machine-generated fakémon (Jan. 23, 2024), https://www.pcgamer.com/ai-palworld-accusations-are-currently-based-on-vibes-a-social-deduction-game-and-the-ceos-past-tweets-about-machine-generated-fakemon/ (AI discussion); Lincoln Carpenter, Palworld could be a delight if it wasn’t so invested in being awful (Jan. 18, 2024), https://www.pcgamer.com/palworld-could-be-a-delight-if-it-wasnt-so-invested-in-being-awful/ (animal abuse discussion).

  • 4.

    See Andres Guadamuz, Palworld, Pokémon, and copyright infringement (Jan. 27, 2024), https://www.technollama.co.uk/palworld-pokemon-and-copyright-infringement. Laypeople online typically use the terms “plagiarism,” “stealing,” or simply “copying.” The legal concept is whether the copying infringes on another’s copyright (or other intellectual property right).

  • 5.

    We have narrowed and simplified the issue for the purposes of this article. We note several limitations of our discussion. First, we discuss US copyright law, but the laws of other countries, such as Japan, might be more applicable to the parties involved. Second, the substantial similarity test can differ between jurisdictions; for example, the Ninth Circuit applies a different test than the Second Circuit. Third, we discuss only the designs of Pals and Pokémon and do not address other elements of Palworld.

  • 6.

    See Mattel, Inc. v. MGA Entertainment, Inc., 782 F.Supp.2d 911, 947 (C.D. Cal. 2011) (citing Funky Films, Inc. v. Time Warner Entm’t Co., L.P., 462 F.3d 1072, 1076 (9th Cir. 2006)).

  • 7.

    Id. (quoting Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157, 1164 (9th Cir. 1977) and citing Feist Pubs., Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 361 (1991)).

  • 8.

    Discovery is a process in litigation that allows each side to ask the other for relevant information and documents such as internal emails related to the case. Parties must provide responsive information and documents; they may not withhold something just because it harms their position.

  • 9.

    See Clark D. Asay, An Empirical Study of Copyright’s Substantial Similarity Test, 13 UC Irvine L. Rev. 35, 37 (2022).

  • 10.

    For a case discussing copyrightability of characters, see, e.g., Blizzard Ent’mt v. Lilith Games (Shanghai) Co. Ltd., 149 F.Supp.3d 1167 (N.D. Cal. 2015).

  • 11.

    See Jenna Romaine, The top 10 media franchises (Oct. 7, 2021), https://thehill.com/changing-america/enrichment/arts-culture/575813-the-top-10-media-franchises/.

  • 12.

    782 F.Supp.2d at 949 (citing Funky Films, 462 F.3d at 1077).

  • 13.

    Id. (citing Well-Made Toy Mfg. Corp. v. Goffa Intern. Corp., 210 F.Supp.2d 147, 160-161 (E.D.N.Y. 2002) and Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1444 (9th Cir. 1994)).

  • 14.

    Id. at 948 (citing Cavalier v. Random House, 297 F.3d 815, 822 (9th Cir. 2002)).

  • 15.

    The comparison is helpful here because the toy industry relies on designs and models. The industry also commonly faces infringement-related issues; one successful toy design often leads to many similar designs produced by competing companies.

  • 16.

    I represented MGA Entertainment in this dispute for several years while working as an associate attorney at Keller/Anderle LLP.

  • 17.

    This is an extremely high-level summary of one claim in Mattel v. MGA. The dispute began in 2004 and involved numerous claims and counterclaims in several courts, touching not just on copyright infringement but also contract law, employment law, trade secret law, antitrust law, and many other areas. This dispute resulted in two jury verdicts and multiple appeals. One professor has published an entire book about the dispute. See Orly Lobel, You Don’t Own Me: How Mattel v. MGA Entertainment Exposed Barbie’s Dark Side (2017).

  • 18.

    Mattel, Inc. v. MGA Entertainment, Inc., 616 F.3d 904, 916 (9th Cir. 2010).

  • 19.

    Id. at 916.

  • 20.

    782 F.Supp.2d at 949.

  • 21.

    Id at 950.

  • 22.

    The district court applied this standard in a summary judgment opinion, where the analysis focused on whether genuine issues of material fact exist such that those issues should be decided by a jury. And for those curious, the case later proceeded to a second jury trial, where the jury ultimately found no copyright infringement by MGA. See Mattel, Inc. v. MGA Entertainment, Inc., 705 F.3d 1108, 1109 (9th Cir. 2013).

  • 23.

    CeciliaFae (@CeciliaFae), Twitter/X (Jan. 21, 2024), https://twitter.com/CeciliaFae/status/1749183059877085396; see also Joshua Yehl, Palworld vs. Pokémon COmparison: Just How Similar Are The Designs? (Jan. 29, 2024), https://www.ign.com/articles/palworld-vs-pokmon-comparison-just-how-similar-are-the-designs (another comparison).

  • 24.

    byofrog (@byofrog), Twitter/X (Jan. 20, 2024), https://twitter.com/byofrog/status/1748890830495010870; ProjectAlpha22 (@ProjectAlpha22), Twitter/X (Jan. 27, 2024), https://twitter.com/ProjectAlpha22/status/1751395263799783682; Gogo, YouTube (Jan. 24, 2024), https://www.youtube.com/watch?v=UcmCuimN6YY&t=3s.

  • Contact
  • Legal