INTELLECTUAL PROPERTY RIGHTS IN COMMUNITY BASED VIDEO GAMES
By Ren Reynolds
Ownership of player-characters in Massively Multiplayer Online Role Play games is a highly contentious issue. This paper explores the applicability of property rights from the point of view of current law and legal philosophy. The paper concludes that player-characters have an ontological status separate from either ideas or physical objects thus the implicit ontology of law would need to be expanded and codified to accommodate them. But that there is also a strong extension of self argument to suggest that property law is inappropriate for settling such disputes. Thus there is a pressing need to examine the expansion of human rights not property rights.
computer game, MMORPG, virtual, intellectual property law, ontology.
When we use a computer network it often seems as if we have entered a virtual world. We commonly talk as if we and other people exist within some virtual realm that is at once immaterial and very real. But transactions conducted on computer networks are intrinsically mediated, so what does it mean to say ‘I am on-line’?
In this paper I will address this broad question by looking specifically at virtual environments know as Massively Multiplayer Online Role Play Games (MMORPGs) and issues surrounding the ownership of game characters. I will examine current legal practice and apply Locke’s schema of private property ownership to assess whether this classic approach to property rights has any currency within the virtual realm and whether it can illuminate broader issues.
This approach to questions of virtual identity and property has been adopted for two reasons. First, game spaces and the player-characters that inhabit them, are most fully realized instantiations of what is popularly known as cyberspace (Gibson, 1984). Thus they are a good indication of the future of on-line mediated interactions. Second, player-characters are highly valued. Their ownership is an area of current dispute between players and game developer-publishers. More broadly this approach forces us to examine the law’s implicit ontological stance and whether this is adequate to deal with entities such as game characters.
Section 2 of this paper provides an overview of MMORPGs, section 3 examines the nature of player-characters, section 4 provides an overview of current law, section 5 dealing specifically with the application Intellectual Property law to MMORPGs, section 6 is a Lockean analysis of property rights in MMORPGs and section 7 concludes.
MMORPGs are online virtual world based games that have their origins in traditional role-play games such as Dungeons and Dragons, and early community computing projects such as Multi-User Dungeons (MUD’s). Popular contemporary MMORPGs like EverQuest, Ashron’s Call, Ultima Online, Anarchy Online and Lineage all operate on a roughly similar model. A player purchases software for their computer, with this software they connect to a server to play the game.
The ostensible point of an MMORPG is character progression i.e. to complete quests and to increase a player-character’s level and in-game power of ones character (Oliver, 2002). While character progression is the intended purpose of an MMOPRG they are in fact used by the millions of registered player \ participants for a wide variety of purposes. For some they are virtual communities (Rheingold, 1993) much akin to chat rooms, for others they are spaces to create works of art (Stern, 2000) or commercial spaces (Brundage, 2000; Cartner, 2002; Castronova, 2001).
Whilst technically accurate the above description of MMORPGs fails to convey the emotional value that participation in the game has for players and the economic value that the game represents to game developer-publishers. The passion and cash generated by MMORPGs has lead to a host of controversies both between players and developer-publishers, and among players themselves. A number of these controversies have gained media attention, a smaller number have gone to court. The following case studies illustrate the kind of issues that are under contest and the forms of legal remedy that are applied in an attempt to resolve them.
MMORPGs are designed such that it should take many hours of effort to build a character to a high level. A market for high-level characters has developed around the most popular games (Castronova, 2001; Muldoon,1999; Morris, 2002) where players develop characters to high levels for the express purpose of selling them to other players. In April of 2000 EverQuest’s publisher Sony On-line Entertainment (SOE) and developer Verant Interactive (a division of SOE) formed an agreement with online auctioneers eBay and Yahoo! preventing the auction of EverQuest characters (Taylor, 2000). Up to this point there was a thriving trade in both characters and in-game artefacts such as swords, fishbone earrings and magic capes earning prices of up to $1400 per item (Sandoval, 2000). Edward Castronova recently calculated that if EverQuest’s game-world Norrath were a country, then the real world trading of characters, artefacts and services would make its GDP 77th in the world putting it between Russia and Bulgaria (Castronova, 2001; Lichtarowicz, 2002).
SOE banned individual players from EverQuest for alleged acts such as Item Farming (Asher, 2000). There has subsequently been a general (but not unanimous (Taylor, 2002)) reaction against SOEs actions. Including the launch, then mysterious disappearance, of a class action against SOE, Verant, eBay et al. The text of the class action argued that banning the sale of player-characters was in fact a constraint of the right to trade “time for money” (Anon, 2001) thus constituted an unlawful restriction of trade.
One of the many fringe phenomena that have gained popularity through the cultural development of the Internet is Fan Fiction a.k.a. FanFic, and its bizarre sub-genre Slash Fiction (which deals exclusively with gay relationship between characters). Fan Fiction is the creation of literature by either individuals or groups, based on characters and settings derived from existing popular works. Fan fiction is most notable in media genres with active fan bases e.g. Science Fiction and Fantasy. Television series such as Star Trek, Buffy the Vampire Slayer and Xena appear to be most popular loci of FanFic, though literary sources such as Harry Potter are also prominent. The name ‘Slash Fiction’ is derived from the way that the original Star Trek based Slash stories were denoted as “Kirk \ Spock” pronounced ‘Kirk slash Spock’ (Stentz, 1998). Examples of Slash can be found in almost any genre e.g. pop music - specifically ‘boy bands’, and ‘cult’ television such as Big Brother.
MMORGs also serve as the inspiration for, and setting of, FanFic. Recently a player who’s character name was ‘Mystere’ was banned from EverQuest for posting a piece of EverQuest FanFic that included references to rape (Taylor, 2002). Moreover SOE caused the story to be removed from the web site on which it was posed even though this was not a SOE or Sony owned site. SOE asserted that these acts were within its rights because, as Andrew Zaffron, general council for SEO, stated “[SOE has] the exclusive right to permit or disallow the outside use of our intellectual property” (Mulligan, 2000; see also Smedley, 2000).
In both of these cases intellectual property rights have been asserted in order to control some aspect of an MMORPG – what the rest of this paper will explore is the validity of such assertions.
3What is a player-character ?
The above case studies demonstrate MMORPGs are more than a fictional battle-ground. Since the inception of MMORPGs there has been a continual struggle over who controls the virtual world. Although these disputes are often couched in terms of property, the issue that they expose is that there is uncertainty about what MMORPGs and the elements of which they comprise actually are. For example, in the case of selling players-characters, SOE asserts that player-characters are their property as they are components of a game they own. Whereas players assert that player-characters, while existing within the game world, are in some sense separate from it and in fact are the player’s own creation, or at the very least SOE and players are equal participants in the creation of an on-going game \ narrative (Mactavish, 2002). The specific question that underlies this dispute is – what is a player-character ? After this has been clarified one can then ask whether the assumption of both game creators and players that property rights can be vested in player-characters is valid and if so, who the rightful owner is.
Player-characters are no exception from the problems of Reduction and Relativity that plague all ontological analysis (Dejnožka, 1996). Thus taking into the account the purpose of this paper, it is sufficient to derive player-characters’ ontological states by identifying each of the levels of abstraction in which they are constructed and operate, just so long as each state is self consistent within its level of abstraction.
Ostensibly player-characters exist within the virtual world of the game. From an instrumentalist point of view they are an interface into that world. At a ‘logical’ level they subsist as underlying database records (i.e. structure and attributes), and a matrix of application algorithms. At a physical level they have instantiation in electrical impulses and patterns of ferromagnetic material. Moving in the opposite metaphysical direction; socially, player-characters exist as entities within a complex web of social relationships formed through performance (Laurel, 1990) and are essential in the construction of the player \ player-character nexus that surrounds the game world.
3.2Player-character as self
Recent models of self have broken away from the essentially Cartesian view that has dominated modernist thought. Such models are particularly instructive when we seek to understand our relationship with the virtual in general and player-characters in particular - because they highlight the cybernetic relationship i.e. one of mutual-feedback (both mental and physical) that we have with virtuality. Enabling us to understand the virtual as a very connected ‘prosthetic, reality’ (Simmons, 1995) rather than the separated other that the Cartesian model forces upon us.
For example, Poster has noted that in the case of using a computer to write - “the screen-object and the writing-subject merge into an unsettling simulation of unity” and more strongly that “Identity is dispersed in the electronic network of communications and computer storage systems”(Poster, 1990). Similarly Turkle has employed ideas derived from theorists such as Lecan and Foucault to emphasise the very real physiological states behind the idea of a ‘decentred subject’ (Turkle, 1995) particular in reference to the use of online communities.
But as Žižek has pointed out, some of these arguments may be taken too far since we can fall into the metaphysical trap of asserting “there is no external reality. RL is just another window” (Žižek, 1997).
In the case of computer assisted writing, what is it that we really ‘merge into’ ? (Poster, 1990). Is it the computer itself, the word processor, or the words ? If we assert that through the act or writing we share an identity with a program such as Microsoft Word and attempt to act upon this assertion - then we might soon end up in court with Microsoft. Also, when we use Word the program remains largely unchanged. So it cannot be with this that we are merging. No, Word is clearly a tool though which we generate text, and it is with this text that our identity merges.
Applying this model to an MMORPGs we have a strong argument for player-characters to have the additional ontological state of being an extension of self, that is they share the identity of a person with the player. What’s more they are clearly separate form the game world in which they exist. Under Žižek’s schema player-characters are an example of the “many person outside a single body” model (Žižek, 1997).
We can now examine in detail the legal mechanisms available for settling disputes over player-characters in the light of the ontological categories established in this section. In this way we can potentially judge the applicability and adequacy of current laws when dealing with the cases at hand.
Law is a codification of a set traditions and principles many of which stretch back for centuries. Thus there is a constant tension between the historic applicability of law and contemporaneous practices and technologies. Digital technology in particular has raised many questions about the adequacy of current legal thought. However, even though the law may not explicitly keep pace with technological changes this is not an argument in itself to clutter legal ontology with strange creatures that may prove either to be weak bearers of legal principle or strong but ephemeral entities that are rapidly left behind by further technological advances.
The question that this section of the paper addresses is: what ways, if any, are player-characters recognised in law and is the present law adequate to produce sensible outcomes when used to resolve the type of issues raised in the case studies above ?
One recent advance in law, at least in the EU, is the recognition of a database i.e. the EU Database directive (Directive 96/9/EC). As one of the states ascribed to a player-character is that of a database it seems that player-characters are clearly defined in law. However if one inspects details of the directive it soon becomes clear that player-characters evade its definition and do not enjoy its protection.
That is, if we take the super-set of all player-characters to be the ‘database’ and apply the Directive, we find under Paragraph 15 that “the selection or the arrangement of the contents of the database” must be “the author's own intellectual creation”. But as the individual contents i.e. the player-characters that make up the database, exist as entries by virtue of individuals choosing to play the game, there is no database Author.
Similarly, if the ‘database’ is taken to be the set of attributes of a given player-character, there again fails to be a legal Author as database records are generated while successfully performing certain tasks within a game, which does not constitute either selection or arrangement.
Thus we are returned to traditional areas of law to seek definitions of the status of player-characters.
5Intellectual Property Law
The area of law most cited in the case studies above is Intellectual Property Law. Both sides of each dispute assume that player-characters are potential items of property and specifically intellectual property. This section examines the validity of these claims.
In summary, Intellectual Property Law codifies the ownership of products of the human mind. Intellectual Property law includes familiar legal instruments such as Copyright, Patent and Trade Mark, and less familiar instruments such as the so called. Moral Rights and Rights of Publicity.
To put this in context, broadly speaking there are two areas of property law: general property law, which deals with physical things such as land or cars or clothes; and intellectual property law, which deals with ideas. To cast this into Token-Type language one can say that property law deals with types for which there is only one token, and intellectual property law deals with types for which there can be multiple tokens. Though it should be noted that this distinction is one of application. That is, for an idea to be covered by intellectual property law it must have at least one expression. For example, the idea for a novel must actually be written down. But when the novel is written down it becomes instantiated in a particular physical object to which general property law apples. Thus if I am the Author of a novel and you steal a copy of that work from my bookshelf you have stolen my book not my novel. If then you reprint the work with your name as author, you have stolen my novel.
As we have established above that at the level of abstraction with which we are concerned, player-characters do not have a physical instantiation, and the case studies at hand specifically cite intellectual property law; so I shall deal herein with the applicability of each area of intellectual property law to player-characters.
5.1Patent and Trade Mark
Patents are the legal instruments used to protect inventions. They grant a form of monopoly to an inventor for the exploitation of that invention. Under English law an invention is patentable if: it is new; it involves and ‘inventive step’; it is industrially applicable and it is not excluded by virtue of, in the specific case of English law, the Patents Act 1977(Philipps & Firth, 2001). Now as industrial application requires a patent to be directly usable in the creation of a product, that is a physical object we can exclude patents from our discussion because player-characters are not something that have, in this sense, industrial application.
Trade Mark law provides protection for badges of trade such as a name (trading, brand of company) or logo. Thus as a player-character is not a name or a logo it cannot by protected by Trade Mark. However attributes of a player-character can. A name for instance could be Trade Marked and applied to a player-character, similarly issues may arise where pre-Trade Marked names are applied to player-characters. If for instance a player with a grudge against the publishers of EverQuest named their character ‘IhateSOE’ then SOE would be likely to have a claim under Trade Mark law. Similarly the physical appearance of character could be incorporate Trade Marked logos wherein the same type of issue would pertain.
5.2Copyright & Moral Rights
Copyright and so-called Moral Rights are both based on the concepts of an ‘Author’ and a ‘Work’. Copyright is an odd legal entity, it is a legal right to exercise legal rights of property on certain Works. In effect it is a grant of limited monopoly that an Author can use to prevent other from making copies of the Work. Importantly while Copyright originates with an Author, it can be traded (Philipps & Firth, 2001).
Moral Rights on the other hand pertain directly to the Author of a work, they provide for example for the Author to be recognized as the Author; for the Author to determining whether a work is complete, whether and how it should be made available and that it should retain its integrity as a Work.
Thus Copyright and Moral Rights are well understood when one can clearly identify an Author and a Work. But Copyright has its origins in the world of print where Authors and Works were simple to identify and categorise. As technology has changed Copyright has sprinted to keep up. At present there is good precedent for the protection of works such as literary works, broadcasts and performance, however there are many grey areas in the digital field most of which come down to an understanding or either what is a work and who (if anyone) is the Author is (or both).
Author & Work
The legal definition of these two terms is best taken in reverse order. A Work is a product of human endeavour that is original, recorded and has an identifiable Author. An Author is one who creates a Work. Thus for Copyright and \ or Moral Rights to apply to player-characters there must be an identified Work and an associated Author.
In the case studies above it’s assumed that a player-character is a Work and either the player or the developer-publisher is the Author of that Work. However there is another way to interpret the state of affairs in property rights terms. If a player-character is an extension of self, then potentially the player \ player-character is the Author and the Work, if any, is not the player-character but, for example, its performance.
Developer-publisher as Author
So is there a legal argument to support the position a developer-publisher as the Author of a player-character Work ? This position can be explored through an examination of Fan Fiction. Just as in the case of Mystere, many authors, publishers or broadcasters have attempted to assert their rights of Copyright to restrict or prohibit Fan Fiction. In practice the weight and expense of lawyers tends to be on the side of the company rather than the individual FanFic creator. However some legal theories have challenged the prevailing notion that copyright in ‘original work’ extends to FanFic. For example Rebecca Tushnet argues that Fan Fiction is a transformative use of a Work that takes time and effort; that they are not duplications of an original text; that where FanFic is parody or departs greatly from the tone of the primary text clearly separates itself from that work (i.e. there is no claim of substitution to be made); that Moral Rights of Authors who’s works are for the mass media do not enjoy the same extent of rights, specifically that characters in some way belong to the audience; moreover that unless a character ‘constitutes’ the story being told (in a copyrighted literary or similar work) they are within the scope of that copyright (Tushnet, 1997) - thus Copyright is likely to reside with the Author or Authors of the piece of Fan Fiction.
Moreover, in the specific case of EverQuest, Taylor has pointed out SOEs position as definitive Author is problematic as EverQuest itself draws not only on the estate of Tolkien in its use of “orcs and “halflings” but also has a command structure that is “strikingly similar to a flavour of MUD called DIKU’ (Taylor, 2002). What’s more the very essence of Role Play Games from the point of view of the developer-publisher and a large number of the player \ participants is the generation of community fictions. A position that seems to defeat the developer-publisher’s own claims of authoritative Authorship, a claim that the arguments above render tenuous at best.
Player as Author
The alternative position – that of Player as Author, is similarly difficult to sustain. If one is to assert that a player is the Author of a player-character, then under the present codification of Intellectual Property law, the player characters must be a Work.
Player-characters require human endeavour in the their creation and development, but while endeavour is a necessary condition of a Work, it is not sufficient. A Work must also be original and recorded in an appropriate form.
But the originality of any player-character is difficult to sustain because while player-characters do have unique elements e.g., their name. Other elements are simply field with a set and finite range of values, such that it is more than likely that for any give character there exists a digital-doppelganger in all but name.
Thus there seems no defence in law for the position of a player as the Author of a player-character as player-characters are not Works.
Player-character as Author
What then if we attempt to establish the player-character as Author ? It may appear that this would flow directly the assertion that a player-character is an extension of self. But the legal status of Author is not a metaphysical claim. Rather it is the narrow claim that something is the identified creator of a Work. Thus for a player-character to be an Author we must be able to identify a corresponding work. As we have established that a player-character is not itself a Work, we cannot make some self referential claim for Authorial status (though this would be legally dubious anyway as while many artists have claimed that they are living works of art (Jahn, 1989) this has no status in law as a Work has to be recorded and a person is not recognised as a legitimate recording device).
Two candidates for a Work are any form of art created by the player-character in the virtual world the performance of the player-character itself. The first of these – works of art, seems to pass the recording test as we could take a screen shot or photograph of the work Also given the permutations of virtual-object and arrangement there should be no problem with the test of originality. The only issue that remains is one of labour i.e. is a player-character something that can perform labour ? Here we are thrown back to metaphysics as players certainly can perform labour. Thus if we make the strong claim that a player \ player-character is a single person then it seems we can establish Author status for art Works within the game-world – though in practice this would likely generate a secondary dispute over copy right in the art work. Deriving Author status from performance gives us the same argument but with a slightly weaker claim as forms of recording a performance may be moot.
The arguments above demonstrate that under current Intellectual Property Law (the law applied most often in disputes over player-characters) player-character are not a Work, are not subject to property rights and are thus not owned by either the developer-publisher of a game or any individual player.
What we can demonstrate however is that there is an argument to support the view that the status of Author may accrue to player-characters, but only in the case where there are Works i.e. original, recorded, products of labour.
This raises the issue of whether property law in general and intellectual property law in particular is the appropriate forum for settling disputes over player-characters. But the fact that the current codification of Intellectual Property law does not accommodate player-characters is not sufficient justification to abandon this area of law completely. Rather we should look to the philosophical underpinnings of property \ intellectual property law to discover whether a coherent property based argument can be established around player-characters that could provide a guide to the further development of law in practice.
There are commonly thought to be four areas of theoretical thought underlying Intellectual Property law (Fisher, 2001). These are: Utilitarianism (Landes & Posner, 1989); Labour Theory (Shiffrin, 2001); Personality Theory (Hughes, 1988), and Social Planning Theory (Netanel, 1996). For considerations of space on Labour theory will be examined here.
The reason for selecting Labour theory is that labour most cited motivation for player’s challenge to the property claims of developer-publishers. Moreover there is currently an active debate about the scope and applicability of Locke to contemporary Intellectual Property law.
6.1Labour theory 101
Locke’s theory is predicated on a state of nature where all is held in common until labour is applied to it. Then, within certain limits, things may become private exclusive property in virtue of an extension of self-ownership argument.
The theory assumes a world where physical labour is being applied to physical objects. Player-characters, at least at the level of abstraction with which we are dealing, are non-physical. Thus, to find a theoretical underpinning for their status as property, we need to seek areas of theory that apply across the physical and non-physical or at least good theoretical parallels which can be applied to the non-physical.
Hence, in the following section Locke’s labour theory shall be applied a generic MMORPG at a single level of abstraction where we have a Game, Player-characters, a game Developer-publisher and Players.
A perennial issue with the application of Locke to the ownership of non-physical items is the concept of the ‘common’ i.e. the state of the world or some part therein before labour is applied.
Sean Valentine Shiffrin defines the three possible states of the ‘initial intellectual common’ as: (a) underlying ideas are in the initial common but not works, (b) the common is empty, and (c) all intellectual works are in the initial common (Shiffrin, 2001). For the moment we shall assume that the intellectual common is the same as the non-physical common i.e. the set of non-physical items of putative property is the same as the set of intellectual items.
So applying these states to the question of whether player-characters are potential items of property we get: (a’) if underlying ideas are in common but not works, then the status of a player-character must be a work i.e. a work distinct from the game itself; (b’) if the common is empty then again a player-character must be a work distinct from the game itself and (c’) if all intellectual works are in common then for a player-character to be privately owned it must have at least one unique property from other products of intellectual work.
6.3(a) Underlying ideas are initial common but not works
Shiffrin’s first state of the initial intellectual common is that ideas are common but works are not i.e. there is some Platonic realm of ideas e.g. 2 + 2 = 4 or the very notion of number and addition, which is common, but products and potentially combinations of these ideas e.g. the idea of a calculator, are not. And thus can be owned. Applying this to our generic MMORPG we have four possible states (i) game and player-characters are common, (ii) game is common but player-characters are not, (iii) player-characters are common but the game is not, and (iv) game and player-characters are not common.
(a) (i & ii) The Game is Common
Assuming the Game is common, does Locke’s theory provide support for non-common status of player-characters and thus a basis for their private, exclusive ownership ? Applying Locke’s theory we have:
Mixing of Labour argument
Locke’s first condition of property is “Whatsoever then he removes of the State that Nature hath provided, and life in, he hath mixed his Labour with, and joined to it something that is his own, and thereby makes it his property”. In the original ’state of nature’ of an MMORPG, player-characters do not exist. Rather they enjoy a neo-Platonic state, where every possible character is defined by the finite set of potential states established by the game rules and configuration. Actual characters are only called into existence through an individual player obtaining a copy of the Game and creating a character through their labour. Hence there is an argument, parallel to of Locke’s example of picking an acorn, that player-characters meet Locke’s first basic requirement of property.
Locke applies boundary conditions to property ownership. The first of these is that property can be private if and only if the putative owner takes “As much as any one can make use of to any advantage of life before it spoils.” Obliviously digital artefacts do not literally rot. However Locke’s condition is still applicable in the digital world as many digital things can have time based qualities e.g. the URL www.worldcup2002.com would in a sense ‘spoil’ if it were not utilized during 2002 as after this time its utility (and value) would be markedly reduced.
In the case of a game player-character the notion of spoiling exists in so much as, if a character is created and not used by a player during the lifetime of a game then it has been underused and the resources of the system have been squandered. What’s more a player-characters must have a unique name, thus every character depletes the finite name space associated with that game. If not used then that area of name space has gone to waste.
‘enough, and as good’ Limit
Locke’s next test is that something may only be private property if there is “enough, and as good left ”. This rule can also be applied to player-characters as the creation of a character uses both system resources and, as noted above, a unique name. Naming depletes the name space and leaves a finite amount remaining. This does not appear to be a relevant issue because with 26 standards letters, number and symbols, even restricting character names to 6 alphanumerics would produce a name space larger than number of players that a current MMORPG could technically sustain. But this assumes that a name spaces are value-neutral. They are not.
Name spaces are highly contoured in terms of utility and value. For instance if a player John Smith wanted to name the character after themself they would have a number of options: John, Smith, John Smith, John_Smith, Jon etc. Again, there seems to be no problem, but as anyone one who has tried to create an AOL identity recently will know, it is certainly conceivable that all of these options will have been depleted. Similarly, ignoring for a moment trademark rights, one would have to be a very early user of just about any game system to be able get the character names Xena or Frodo. Thus enough and as good is a meaningful limit in any system with a finite name space where each instance of name must be unique within that space. This again adds weight to the Lockean justification of player-characters as capable of being items of property.
Locke’s minimal criteria & intellect vs non-physical property
On a first pass analysis, and given the assumptions: (a) ideas are common but works are not, and, that the state of the common is such that the game is common (i & ii). It seems that Locke’s tests make sense when applied to an MMORPG and that player-characters pass them. However, what this argument actually demonstrates is that with the given assumptions, player-characters pass Locke’s maximal limit tests. But one argument against Locke’s support of Intellectual Property is that all ideas pass these maximal tests. As, by their metaphysical nature they cannot either waste or be depleted. What we further have to demonstrate is that player-characters also pass the minimal test, i.e. it must be explicitly demonstrated that player-characters are not common. Or to put this in Platonic terms they are created not discovered.
In fact the necessary argument is implicit in the discussion of the limit tests above. MMORPGs rely on name spaces with discrete unique elements i.e. character names. So just like other name space elements e.g. IP addresses or domain names, it is intrinsic to the being and functioning of player-characters that they have some necessary element of exclusivity. Now it might be argued that brands exist within a brand name spaces i.e. that no two brands can have the same name. But separate products with the same names have existed at the same time, moreover the expression of a brand name on a physical product is distinct from the brand itself. Player-characters differ from this by virtue of the state in which they exist. Virtual name spaces are simplistic. They are comprised only of the set of potential attributes that their members can have. Thus, in a name space, whether two things have the same attributes is in fact an identity criteria i.e. if they do they are the very same thing. Hence, player-characters can be demonstrated not to be part of the intellectual common as they require some particular, though virtual, existence.
The stark conclusion that we must draw from this is that while player-characters are not physical, neither are they pure ideas. That is, we find that in terms of property, the virtual world shares characteristics with both the physical and non-physical worlds – a point that I shall return to in the conclusion.
Player-characters meet Locke’s basic test for ownership criteria, and the maximal and minimal tests. For completeness it remains to be demonstrated explicitly that Locke supports their private ownership as this is not a necessary a corollary of the limit test.
Support for exclusive ownership can be derived by recognizing that Locke’s argument for private and exclusive ownership is largely pragmatic rather than principled. This is demonstrated by Locke’s example of an apple. His main line of argument is that by mixing labour with an item such as an apple it becomes exclusive property because one is ‘self owning’ and has a duty to self preservation, thus one has a duty to use the apple to this end i.e. in the most basic form - gather and eat the apple. To fulfil this purpose the apple must be exclusively owned – as obviously the very same apple cannot be digested by ones own stomach and another’s.
But what if the apple is cut in half, and half again, and again. Logically the apple could be divided such that all held the tiniest fraction of it. In this way it would it seem become common again. But practically speaking at some point the apple would cease to have nutritional value. Thus one can interpret Locke’s underlying motivation as the wish to ascribe exclusive rights in the smallest unit of a thing that retains practical value. Given our level of abstraction, player-characters are the smallest unit to bear practical value then we can temporarily conclude that under the given assumptions Locke does indeed provide support for their private exclusive ownership.
6.4(a) (iii & iv) The Game is not Common, (b) The Common is empty &(c) All intellectual works are in initial common
If we now return to Shiffrins schema and retain the assumption (a) ideas are common but works are not, but now assume (iii & iv) that the state of the common is such that the game is not common, we in fact have no impact on the arguments above for the status of player-characters as putative items of property. What is altered is the weight of argument for players being the owners of player-characters (see below).
Next, if we assume state (b), that the common is empty, this resolves into assumption (iv) above i.e. nether the Game or player-characters are common, hence the argument in the above can be applied.
Lastly we can now discount the applicability of argument (c) as we have demonstrated that a player-character is not part of either the intellectual common or some wider non-physical common.
Would the rightful owner please stand up ?
Thus far we seem to have a Lockean justification for player-characters as potential items of private exclusive ownership, but no conclusive view as to owner.
What the above illuminates is that the property arguments of developer-publishers stem from dependency rather than identity. That is, the arguments demonstrate that player-characters are in a strong sense distinct from the game world although they are dependent upon it.
But ownership does not flow from dependency. To illustrate, take the example of an apartment on the 13th floor of a building. It and the 11 apartments below are not owned by the ground floor in virtue of the fact that it must exist to support all the floors above it.
Hence it seems that there are Lockean justifications for both the status of a player-character as an item of property and the player as the owner apply whether the Game is common or not i.e. in all cases.
6.6Ownership vs Identity
The arguments above seem to provide a conclusive Lockean justification for the private, exclusive ownership of player-characters by players. One of this argument’s predicates is that player-characters are not pure intellectual objects in either legal or metaphysical terms, indeed this dichotomy has been forced upon us by the metaphysical assumptions of law and property which were formed before the creation of virtual worlds. But we have also argued that player-characters are aspects of self.
Returning to Locke we discover that his fundamental motivation for the existence of property is also, in a sense, based on an extension of self argument i.e. “he hath mixed his Labour with, and joyned to it something this his own, and there by makes it his Property”.
This raises two further issues. First, the concept of self that Locke’s theory is based upon (Mackie, 1976) is very different from the decentred self we have from Poster, Turkle et al as Locke recognized a separation between the self and ones property i.e. after ‘mixing’ there is separation. Moreover it is fundamental a player-character that is it s work in progress. So it is not the case the player “hath mixed” their labour but they are continually in a state of mixing. Second, as Shiffrin notes, it is not clear that using strong self-identity claims can “sensibly be extended to extend protect individual’s ability to disperse aspects of themselves into the public domain while simultaneously retaining complete control over them.” (Shiffrin, 2001)
Hence making a strong self-identity claim for player-characters undermines much of the previous argument for their status as property.
7Conclusion – Duties and EULAs
Intellectual Property law does not currently provide protection for player-characters because metaphysically they are outside the idea \ object \ Cartesian-self ontology on which it is based. To integrate player-characters and other non-physical \ virtual objects into law its ontology wound need to be expanded, and the nature of these objects codified. But, taking seriously the idea of a player-character as an extension of self such a modification of law would be inappropriate. Because selling a player-character would literally mean selling oneself –which is disallowed under most legal jurisdictions and is metaphysically problematic at best.
This renders much of the argument in the case studies above meaningless, because property law is an inappropriate vehicle for settling such disputes. What’s more the developer-publishers’ End User Licensing Agreement (EULA) that is often cited as the basis of their property claim is, in this respect of property rights over player-characters, invalidated. This is because, although property rights may be altered by the application of contract law, basic rights cannot i.e. it is not possible for example to form a legal contract to either buy or kill someone.
The extension of self argument has a number of further pertinent consequences. First it effectively negates the trade in player-characters (unless these are created by bots or some other automated means where Locke’s argument may apply due to the lack of an identified self). Similarly as developer-publishers do not and cannot own player-characters it is questionable whether they would be entitled to sell their game world (Scheurer, 2002). As far from having rights over player-characters, developer-publishers in fact have (probably unexpected) duties. For example a developer-publisher should have some duty of care over player-characters i.e. just as the owner of a fun fair may intend to create a space for innocent fun in fact they must abide by stringent rules of maintenance and safety. An idea that has significant implications for the development and maintenance any virtual environment be it a game world, community space or part of a wider system such as a banking network – in fact any virtual space where a strong self identity claim can be made on virtual entities.
This conclusion raises more questions than it answers. It demonstrates that taking seriously the notion of a decentred self and an online entity as an aspect of that self (for which there exists sound argument) exposes wide gaps in the law. Fundamentally, the extension of and codification of human rights into the virtual is far from science fiction but an issue that should be of pressing and immediate concern.
I wish to thank: K Conroy, L Dubbeld and V Scheurer for their help in the development of this paper.
Anon. (2001), Everquest class action threat over auction spat, online at
http://www.theregister.co.uk/content/6/16355.html accessed on 10.07.2002.
Asher, M. (2000), Two Players Banned for EverQuest Auctions, online at http://www.gamecenter.com/News/Item/0,3,0-4036,00.html?st.gc.fd.lgn.gn accessed on 28.01.2001.
Beard. J. J., (2001), Clones, Bones and Twilight Zones: Protecting the digital persona of the quick, the dead and the imaginary, Berkeley Technology Law Journal, 16:3.
Brundage, S (2000), Making EverQuest Easier to Play: Cheating or Not?, gamers.com, available online at http://www.gamers.com/news/322928, accessed on 15.07.2002.
Carter, G. (2002), It’s My Time and I’ll Sell if I Want To, The Adrenalin Vault, available online at http://www.avault.com/articles/getarticle.asp?name=mmogsell, accessed on 15.07.2002.
Castronova, E. (2001), Virtual worlds: A first-hand account of market and society on the cyberian frontier, Gruter Institute Working Papers on Law, Economics, and Evolutionary Biology, Volume 2.
Dejnožka, J (1996), The ontology of the analytic traditions and its origins, Littleford Adams Books.
Dibble, J. (1993), A Rape in Cyberspace, The Village Voice, December 21 1996. p36-42.
Fisher, W. (2001), Theories of Intellectual Property, in Munzer (ed), New Essays in the legal and political theory of property, Cambridge.
Gibson, W (1984), Nuromancer, HarperCollins.
Hughes, J.(1988), The Philosophy of Intellectual Property, The Georgetown Law Review , Vol. 77, 287-366.
Harris, J.W. (1997), Legal Philosophies, Butterworths, p64-82.
Jahn, W. (1989), The Art of Gilbert and George, Thames and Hudson.
Laurel, B (1990), Computers as theatre, The Art of Human-Computer Interface Design, Adison-Wesley
Landes, W.M & Posner R.A. (1989), Economic Analysis of Copyright Law, Journal of Legal Studies, 18.
Locke, J (1988), Two Treaties of Government, Laslett, P. (ed) Cambridge
Lyon, D.(1997), Cyberspace sociality, in Loader, B. D. (ed), The Governance of Cyberspace, Routledge, p 23-37.
Lichtarowicz, A. (2002), Virtual kingdom richer than Bulgaria, online at
http://news.bbc.co.uk/hi/english/sci/tech/newsid_1899000/1899420.stm accessed 10.07.2002.
Mackie, J. L. (1967), The Problems from Locke, Oxford.
Mactavish, A. (2002), Technological Pleasure: The Performance and Narrative of Technology in Half-Life and other High-Tech Video Games, in (Forthcoming) King, G and Krzywinska, T (eds), Screenplay: Cinema/ Videogames/ Interfaces. Wallflower
Morris, C. (2002), Imaginary worlds. Real cash, online at
http://money.cnn.com/2002/01/16/technology/column_gaming/ accessed on 10.07.2002.
Muldoon, M. (1999), They got game, online at
http://www.salon.com/tech/feature/1999/07/13/ultima_ebay/index.html accessed on 10.07.2002.
Mulligan, J, (2000), Mommy, she talked dirty!, happypuppy.com, online at
http://www.happypuppy.com/features/bth/bth-vol9-28.html accessed on 15.07.2002.
Netanel, N. W. (1996), Copyright and a Democratic Civil Society, 106 Yale Law Journal 283.
Oliver, J. H. (2002), The similar eye: Proxy life and public spaces in MMORPG, CGDC Conference Proceedings, Tampere University Press, p171-184.
Phillips, J. & Firth, A. (2001), Introduction to Intellectual Property Law, Butterworths, 4th ed.
Poster, M. (1990), The mode of information, Polity, p99-128
Rheingold, H. (1993), The Virtual Community: Homesteading on the Electronic Frontier, Addison-Wesley p1-10.
Sandoval, G. (2000), Sony Bans EverQuest Auctions, gamecenter.com, online at
http://www.gamecenter.com/News/Item/0,3,0-4036,00.html?st.gc.fd.lgn.gn access on 28.01.2001.
Scheurer, . (2002), For Sale – the concept of ownership of an online community, Developer Magazine.
Shiffrin S.V., (2001), Lockean Arguments for Private Intellectual Property, in Munzer (ed), New Essays in the legal and political theory of property, Cambridge.
Simmons, J. Sade and Cybersapce (1995) in: Brook, J, Boal, I. A. (eds), Resisting the virtual life: The culture and Politics of Information, Brook and Boal,145-159.
Smedley, J. (2000), Posting on eqvault.com, online at http://eqvault.ign.com/archive/arc73.shtml access on 15.07.2002.
Stentz, Z. (1998), Vulcan Love Slave, metroactive.com, available online at http://www.metroactive.com/papers/metro/05.07.98/tv-9818.html accessed on 15.07.2002.
Stern, E (2000), Summons to Surrender, available online at http://www.summonstosurrender.com accessed on 15.07.2002.
Taylor, T.L. (2002), “Whose game is this anyway?”: Negotiating corporate ownership in a virtual world, CGDC Conference Proceedings, Tampere University Press, p227-242.
Turkle, S. (1995), Life on the Screen : Identity in the Age of the Internet, Simon and Schuster.
Tushnet, R. (1997), Legal Fictions: Copyright, Fan Fiction, and a New Common Law, Loyola of Los Angeles Entertainment Law Journal, (17 Loy. L.A. Ent. L.J. 651)
Žižek, S. (1997), The plague of Fantasies, Verso, p127-167.
Wrinn, G (2000) post on EverQuest game forum (http://eq.castersrealm.com/viewarticle.asp?Article=1613)
© Ren Reynolds 2002 Page of +44 (0) 7778 285 273