Skip to content

#law· 10 min readLire en français

From software to complex work: the distributive classification of video games

For a long time, the law saw video games as mere software, whose regime extended to everything else. Since the Cryo ruling, each component (code, graphics, music, script) follows its own regime. A look back at a shift with very concrete effects.

By Baptiste GODIN

Diagram: the video game, first a software work (unitary approach), becomes after the Cryo ruling of 25 June 2009 a complex work whose components (software, audiovisual, music, graphics, script, gameplay) each follow their own regime

Through legendary titles such as Pac-Man, Super Mario Bros. or Tetris, the video game has established itself as a hybrid object, both a technical program and a rich artistic creation. It contains code, pixels, music, narrative worlds, game mechanics… in short, anything but a simple work that fits into a single legal box.

Very quickly, the law came up against a central question: should a video game be treated as a unitary creation, subject to a single regime (often that of software), or as an assembly of distinct elements (code, graphics, sound, script) each to be dealt with under its own rules?

For years, the dominant answer was to look at everything through the lens of software. It was convenient, especially in the days of The Legend of Zelda (1986) or Doom (1993): a video game is a computer program, so only that regime applies. Case law long supported this reading, letting the maxim accessorium sequitur principale apply in full: software as the principal element, everything else as mere accessories.

But as works became more complex (dense narrative worlds, sophisticated art direction, original soundtracks), this unitary reading showed its limits. The rise of titles such as The Last of Us (2013), Cyberpunk 2077 (2020) or highly cinematic experiences, in the style of Quantic Dream for example, makes it hard to reduce a game to its “code” dimension alone.

In the background lies a major tension in copyright law as applied to video games: the shift from a unitary classification (every game is software, a database or an audiovisual work) to a distributive classification, where each component of the game is attached to the regime that suits it.

This shift determines how the originality of the game’s various building blocks is proven, how rights are distributed between authors, studios and publishers, and how a video game work, now more than ever a complex creation in its own right, is negotiated and legally secured.

I. The initial assimilation of video games to software: the appeal of a unitary classification

A. The pre-eminence of the software dimension in early analyses

At first, lawyers mainly looked at video games through the most obvious lens: that of the computer program that runs them. Without code, there are no images, no game mechanics and no possible interactions; the entire gaming experience rests on this technical backbone.

Following this logic, the criminal chamber of the French Court of Cassation, in its ruling of 21 June 2000, Midway (the Mortal Kombat case), established the video game as a unitary work: software is presented as the essential element, so its legal regime is applied to the whole creation. In other words, if code is the heart of the video game, then the entire work follows its legal fate.

On paper, the solution is attractive: one main object of protection, an identified regime, a clearly designated owner, and clear management of acts of reproduction, adaptation or distribution.

This view also fitted the first generations of games fairly well, as they were more modest in their artistic ambitions: Doom, Mortal Kombat (1992) or Quake (1996) could still be described above all as combinations of rules, levels and lines of code.

But this unitary classification is not just a technical reflex: it also reflects lawyers’ desire to reassure themselves with a single, stable and clear category, applying to video games a coherent copyright regime centred on protecting code, especially since software is expressly listed among protected works in Article L. 112-2 of the CPI.

B. The consequences of applying the software regime across the board

This unitary classification of video games as software has very concrete effects, which explain both its appeal… and its limits. By adopting it, the game is placed under the specific regime for computer programs, designed for code: it closely protects the program (source code and object code) but keeps underlying concepts or principles at a distance, such as the fighting mechanics of Mortal Kombat in the Midway case. The publisher then has strong control over the technical reproduction of the game, its digital distribution and its substantial adaptations.

On paper, this unification simplifies contractual life: a single regime, one main contact, centralised exploitation. For productions such as Doom or Quake, where gameplay and code formed an almost inseparable whole, this “all software” lens effectively secured publishers’ investment against wholesale copying or technical misuse.

But this global reading soon revealed its blind spots. By treating the whole game as software, it pushes into the background the graphic, musical or narrative components, which are traditionally covered by “classic” copyright. Visual worlds, music and characters legally become accessories of the code that contains them, at the risk of obscuring their own creative contribution.

Hence a real grey area: if a video game is only “software”, what about the iconic music of Super Mario Bros., the dialogue of The Last of Us or the architecture of Night City in Cyberpunk 2077? Are they merely data accompanying the program, or works in their own right, with their own authors, their own rights and their own logic of exploitation?

As games became artistically richer, notably thanks to technological progress, this tension between the functional primacy of code and the reality of a collective work became hard to ignore and began to crack the unitary construction, opening the way to the distributive classification model.

II. Gradual recognition of the plural make-up of video games: the advent of distributive classification

A. The video game as a complex work made up of distinct elements

Gradually, the view of video games as mere software showed its limits. By looking closely at the video game as an object, legal scholars and courts brought out a simple idea: a game is not a technical block, but a combination of creative layers, not all of which can be confined within the regime for computer programs.

Behind every title there are several families of elements: source code falling under special software law, graphics and music calling for “classic” copyright (sometimes databases), and a script and characters that may be protected individually. The video game thus resembles a composite multimedia work more than a mere executable to which a single regime could be applied.

The example of The Legend of Zelda: Breath of the Wild (2017) speaks for itself: a program managing physics and AI, the stylised landscapes of Hyrule, a subtle soundtrack, dialogue, staging, and iconic characters such as Link and Zelda. The same is true of Cyberpunk 2077, with hours of voice acting, distinctive art direction and a particularly elaborate futuristic urban world.

At the same time, the industry has moved from 2D sprites to AAA productions with blockbuster budgets. The more artistically ambitious a game becomes, the harder it is to claim that, legally, it is “only” code with accessory elements: the music of Final Fantasy VII (1997), the animations of a God of War (2005) or the level design of Dark Souls (2011) are genuine intellectual creations, with their own value and, potentially, their own owners.

From then on, continuing to treat the video game as a homogeneous whole means missing its creative reality. Distributive classification is not a mere legal refinement: it gives legal expression to the obvious fact that a video game is a hybrid object, calling for its components to be separated in order to be protected and exploited coherently, an intuition the courts are increasingly embracing.

The real break came when the courts finally agreed to see the video game for what it is, a complex work, and not a mere software block. This is the whole contribution of the Cryo ruling of 25 June 2009, in which the first civil chamber of the French Court of Cassation held that “a video game is a complex work that cannot be reduced to its software dimension alone, however important that may be, so that each of its components is subject to the regime applicable to it according to its nature”. In other words, a single regime is no longer applied to the whole game; the reasoning is done component by component.

This decision turned its back on the unitary approaches that had prevailed until then, notably when the whole game was attached to the software regime, as in Mortal Kombat. Such a reading extended to non-software elements (script, images, music, gameplay) a regime designed for the computer program alone. It missed the creative reality of video games, made up of graphic, sound, narrative and interactive layers. With Cryo, the Court corrected this reductive view: a single game can contain several “sub-works”, each falling under its own protection regime.

This case law has a twofold merit. Technically, it avoids making the protection of the whole game depend on just one of its building blocks. In practice, it strengthens legal certainty for creators and operators: the code falls under special software law, the graphic and sound elements under ordinary copyright, and pre-existing works keep their own regime. The distributive logic thus makes it possible to recognise the plurality of contributions without denying the existence of an overall work.

Above all, Cryo lays down a real method: the courts no longer look for a single dominant classification, but accept that the video game is a hybrid creation whose elements must be isolated so that each receives its appropriate legal treatment.

This approach was extended in several decisions, such as the 2016 judgment Atari v. Raynal on Alone in the Dark, in which the video game was classified as a work of collaboration and Frédérick Raynal’s authorship of certain components was recognised, concretely illustrating the need to identify rights “block by block”. The law thus takes a more realistic view of video games: rather than imposing an artificial unitary classification, it now embraces the plurality of creative layers that make up the video game work.

III. The scope of the distributive shift: towards protection more faithful to the reality of video games

A. The fragmentation of applicable regimes according to the nature of the components

The recognition of distributive classification radically changes how the law approaches video games. Since it is recognised as a complex work, a single regime can no longer be applied to it: each component must be protected according to its own nature.

In concrete terms, this breaks the regime into several blocks: the software part remains subject to the special law on computer programs, with its objective approach to originality and the exclusion of functionality or operating principles, while the graphic, sound, audiovisual or narrative elements fall under “classic” copyright, based on the imprint of the author’s personality.

This segmentation better matches the reality of video game development, where a single game combines source code, visuals, music, script, dialogue, animations and even integrated pre-existing works, which meet neither the same criteria for protection nor the same rules of ownership.

This intuition also extends beyond France: in 2014, in Nintendo v. PC Box, the CJEU recalled that a video game is “complex matter” comprising not only a computer program but also graphic and sound elements which, although encoded in computer language, have their own creative value. Implicitly, the message is the same: it would be artificial to reduce the whole game to the single category of “software”, when some of its components fall fully under ordinary copyright.

On this last point, the consequences of distributive classification are significant: depending on the component concerned, the author will not be the same (developers for the software, writers for the story, composers for the music, artists for the graphic assets, etc.). The publisher can only claim rights through the applicable assignments or legal mechanisms, which requires a “block by block” analysis to secure the game’s economic exploitation.

Ultimately, this fragmentation is not a mere technical refinement, but the legal expression of what a video game really is: it is more faithful to the reality of video games than earlier attempts at unification.

B. Practical effects on ownership, exploitation and protection of creations

The distributive classification of video games is not just a theoretical debate: it very concretely changes the way rights are identified and exploited. Since the game is no longer seen as a single block, one must reason component by component, which affects the identification of authors, contractual arrangements and the extent of protection.

On ownership, it can no longer be presumed that a single player holds “all” the rights. Depending on the element concerned, there may be multiple owners: developers for the software, writers for the story, composers for the music, graphic artists for the visuals, not to mention the authors of pre-existing works integrated into the game. Following the Cryo ruling, each building block must be attached to its own regime, which requires identifying precisely who holds what.

This logic directly affects exploitation: the publisher can no longer simply reason globally; it must contractually secure each protected contribution (code, graphic assets, music, cutscenes, characters, trademarks, third-party content, etc.). Otherwise, exploitation may be weakened by an infringement action or a challenge to an assignment of rights, hence the central role of contracts in organising assignments, licences, warranties and the sharing of remuneration.

Protection itself becomes more nuanced: the video game is no longer covered by a single regime that would absorb all its dimensions, but by a series of partial protections, each with its own conditions (originality, duration, exceptions…). This avoids, for example, inferring the originality of software from mere audiovisual effects, in favour of a separate demonstration of originality for each component.

Ultimately, the practical effects of distributive classification are ambivalent: it strengthens legal coherence by recognising the plurality of the video game, but it also raises the bar when it comes to securing rights. Creators and operators alike must think of the video game not as a monolithic work, but as a bundle of interlocking rights, to be mapped and organised from the production stage onwards.

Bibliography

Legislation and European texts

  • French Intellectual Property Code
  • Directive 2009/24/EC of 23 April 2009 on the legal protection of computer programs
  • Directive 2001/29/EC of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society
  • Directive 2006/115/EC of 12 December 2006 on rental right and lending right and on certain rights related to copyright in the field of intellectual property
  • Directive 96/9/EC of 11 March 1996 on the legal protection of databases
  • Directive (EU) 2019/790 of 17 April 2019 on copyright and related rights in the Digital Single Market
  • Directive 2004/48/EC of 29 April 2004 on the enforcement of intellectual property rights

Case law

  • Cass. crim., 21 June 2000, Midway / Mortal Kombat
  • Paris Court of Appeal, 20 Sept. 2007, Cryo
  • Cass. 1re civ., 25 June 2009, Sesam v. Cryo
  • CJEU, 4th chamber, 23 Jan. 2014, case C-355/12, Nintendo and Others v. PC Box SARL and Others
  • Lyon Judicial Court, 8 Sept. 2016, Atari v. Raynal
  • Cass. 1re civ., 23 Oct. 2024, Valve v. UFC-Que Choisir
  • P. Sirinelli, comments on Cass. crim., 21 June 2000, Mortal Kombat, D. 2001, p. 2552 (critique of unitary classification as software)
  • E. Treppoz, note on Cass. 1re civ., 25 June 2009, Cryo, JCP 2009, No. 328 (in-depth analysis of distributive classification)
  • J.-M. Bruguière, comments on Cryo, Propriétés intellectuelles 2009, p. 366
  • « Le jeu vidéo, à qui appartient-il ? Propriété intellectuelle », Culture Games, 2014
  • M. Cahen, « Quelle protection pour le jeu vidéo ? », 5 Oct. 2022
  • APP, « Focus : la protection juridique d’un jeu vidéo », 17 Jan. 2019
  • Eurojuris, « La qualification juridique du jeu vidéo », 2024
  • J. Douillard, « Fin de partie pour l’épuisement immatériel du jeu vidéo », comments on Cass. 1re civ., 23 Oct. 2024
  • IREDIC, « Le jeu vidéo Alone in the Dark est une œuvre de collaboration », 3 March 2017

Further resources

  • CSPLA, Le régime juridique des œuvres multimédia : droit des auteurs et sécurité juridique des investisseurs, report, 2005
  • French Ministry of Culture, Projet de statut juridique du jeu vidéo, 27 February 2013
Baptiste GODINMaster's student in digital law

https://lexludica.com/en/articles/from-software-to-complex-work

Subscribe to our newsletter

By subscribing, you accept our privacy policy.