Virtual Worlds Don't Exist: Questioning the Dichotomous Approach in MMO Studiesby Vili Lehdonvirta
Abstract:I argue that much influential scholarship on massively-multiplayer online games and virtual environments (MMO) is based on a dichotomous “real world vs. virtual world” model. The roots of this dichotomy can be traced to the magic circle concept in game studies and the cyberspace separatism of early Internet thought. The model manifests on a number of dimensions, including space, identity, social relationships, economy and law. I show a number of problems in the use of this model in research, and propose an alternative perspective based on Anselm Strauss’s concept of overlapping social worlds. The world of players does not respect the boundaries of an MMO server, as it frequently flows over to other sites and forums. At the same time, other social worlds, such as families and workplaces, penetrate the site of the MMO and are permanently tangled with the players' world. Research programs that approach MMOs as independent mini-societies are therefore flawed, but there are many other kinds of research that are quite feasible.
Keywords: reseach design, methodology, online games, magic circle, social world perspective, virtual space, identity
Introduction[T]his book is not about analogies. We won’t tell you that devising business strategies is like restoring an ecosystem, fighting a war, or making love. Business strategy is business strategy and though analogies can sometimes be helpful, they can also be misleading. Our view is that analogies can be an effective way to communicate strategies, but they are a very dangerous way to analyze strategies.
Shapiro & Varian, Information rules
In academic literature, certain online games and services are referred to as “virtual worlds” and compared to cities (Taylor, 2006, p. 21), countries (Castronova, 2006b) and most frequently, even the planet Earth (e.g. Castronova, 2002; Castronova, 2006a; Nash & Schneyer, 2004, Lastowka & Hunter, 2004). Such language is intended to communicate the scale and complexity of these systems and the activities that take place within them. But these compelling analogies also influence the conceptual framework from which researchers draw their research design. In this article, I attempt to show that the term “virtual world” invites incorrect assumptions about the entity it describes. I will instead use the less popular term MMO (massively-multiplayer online) to refer to these games and environments.
Academic interest in MMOs outside the sphere of game studies has been growing for several years. According to a “virtual law” bibliography put together by Greg Lastowka (2008), three times as many articles were published in 2007 as were in 2006. To guide and focus research efforts related to MMOs, several research agendas have been proposed. For example, Caroline Bradley and A. Michael Froomkin suggest using MMOs to conduct legal experiments that would be too costly to carry out “in the real world” (Bradley & Froomkin, 2004, p. 103). Edward Castronova argues that MMOs are the “social science equivalent of a petri dish, or a supercollider” (Castronova, 2006a, p. 170). Robert Bloomfield advocates the development of a special MMO for the purpose of education and research in business disciplines (Bloomfield, 2007). David Bray and Benn Konsynski attempt to outline opportunities for researching “intra-world and inter-world practice and behavior” from several disciplinary perspectives (Bray & Konsynski, 2007).
The problem with these agendas and various other MMO-related studies is that they are based on a dichotomous “real-virtual” perspective to MMOs, which I describe in detail in the second part of this paper. In this perspective, MMOs are painted as separate worlds, located outside “the real world”, in many ways mirroring it like a synthetic double, but carrying on independently of it like a distant planet. Marketers and mainstream media are fond of this way of viewing MMOs, as it evokes powerful images of parallel worlds from science fiction turned into reality. For researchers, however, it is a treacherous fantasy. In her extensive study on the massively multiplayer online role-playing game EverQuest (SOE, 1999), T.L. Taylor concludes that “To imagine we can segregate these things - game and non-game, […] virtual and real - not only misunderstands our relationship with technology, but our relationship with culture” (Taylor, 2006, p. 153). The danger is that social scientists, legal scholars and information systems researchers now taking interest in the area adopt an evocative but inaccurate conceptualisation of MMOs that leads them to build their research on false assumptions.
In the third part of this paper, I review some well-known problems in the “real-virtual” perspective to MMOs, and present a few new ones. In the fourth part, I suggest that instead of using a dichotomous approach, scholars should place MMOs side-by-side with spheres of activity such as family, work or golf, approaching them using the same conceptual tools. The symbolic interactionist concept of social worlds as advocated by Anselm Strauss (1978) is introduced in some detail and found particularly convenient for conceptualising the relationship between MMOs and other spheres of activity. In the final part of this paper, I discuss the implications of this social world perspective for research design, and finish with a short criticism of the way “real life” is sometimes conceptualised in MMO studies.
Virtual Worlds Versus the Real WorldVirtual worlds are now a reality. Virtual worlds allow everyone to create a digital character representing themselves and interact with other computer-generated individuals, landscapes, virtually-run global businesses, and in-world institutions in real-time. Fascinatingly, both endogenously produced economies and social orders are emerging in these virtual worlds. Political candidates are campaigning in virtual worlds, while some sales of virtual assets are producing demand in the real world for equivalent items.
(Virtual Worlds and New Realities in Commerce, Politics, and Society
Conference announcement, December 2007)
The “real-virtual” perspective in MMO studies can be seen as heritage from two earlier traditions: games and the Internet. In Internet discourse, use of terms such as “IRL” and “cyberspace” illustrates thinking where network-mediated communication is conceptualised as activity and space distinct and separate from the affairs of the “meatspace”. A very explicit and influential expression of this thinking is John Perry Barlow’s A Declaration of the Independence of Cyberspace (Barlow, 1996). The essence of Barlow’s manifesto is that a new world is emerging on the Internet, and that it is quite distinct from the old one.
In computer game parlance, developers and gamers have of course long been using the term “world” when discussing games such as Civilization (Meier, 1991), but in this use it is more of a label than a statement regarding one’s conceptualisation of reality. The conceptual dichotomy I am referring to can be found among role-playing gamers, who use the terms “in-game” and “out-of-game” to draw a line between “the game” and “the rest of the world”. In scholarly circles, this boundary is known as the “magic circle”. The concept was introduced into digital game studies by Katie Salen and Eric Zimmerman (2004), who attribute the idea to Dutch historian Johan Huizinga (1872-1945). Huizinga defined play/game as follows:
- [A] free activity standing quite consciously outside “ordinary” life as being “not serious”, but at the same time absorbing the player intensely and utterly. It is an activity connected with no material interest, and no profit can be gained by it. It proceeds within its own proper boundaries of time and space according to fixed rules and in an orderly manner. It promotes the formation of social groupings which tend to surround themselves with secrecy and to stress their difference from the common world by disguise or other means. (Huizinga, 2000, p. 13)
This “magic circle” view of games has become quite influential in contemporary game studies. According to Lin and Sun (2007), the magic circle view entails treating the game as “a world independent of the everyday real world.” The best play experience is achieved when the game is “insulated from or opposite to the utilitarian characteristics of the physical world” (Lin & Sun, 2007, p. 336). Many game scholars have since joined in the discussion to argue in support of or against the view (e.g. Copier, 2005). Malaby suggests that this setting-apart of games is actually “the largest roadblock to understanding what is powerful about them” (2007, p. 96).
Two influential authors writing on MMOs outside game studies who explicitly invoke the magic circle concept are economist Edward Castronova (2004) and legal scholar Greg Lastowka (2007; 2009). Lastowka examines the question of how law should deal with new MMORPG-related legal issues, such as real-money trading of virtual assets. Lastowka’s main argument is that play does not “conform to the reason and logic of ‘ordinary life’”, and that law must consequently treat game activities differently from “ordinary life”.
- When a sumo wrestler enters the “magic circle” of the dohyo or the professional boxer enters the space and time of the bout, the rules of what social behaviors are desirable and forbidden are suddenly, radically changed. Violent and powerful physical attacks against another person, which are normally forbidden by law and social norms, become the obligatory mode of conduct. (Lastowka, 2007, p. 8)
Inspired by Huizinga’s ideas, Lastowka makes a clear distinction between MMOs and the outside world. “[W]hile MMORPGs have some things in common with weblogs and social networks, they are also very different from other forms of online activity”, since “MMORPGs are games” (Lastowka, 2007, p. 3). According to Lastowka, courts and lawyers already understand the special status of the golf course and the baseball field, and they should now extend the same understanding to MMORPGs. Courts should set aside conventional rules in favour of the “rules of play”.
Castronova, one of the most cited authors in MMO-related scholarship, argues in The Right to Play (Castronova, 2004) that all humans have a fundamental need to play. To define what qualifies as play, he refers to Huizinga:
- For Huizinga, nothing can be a game if it involves moral consequence […] if some consequence really does matter in the end, the game is over. In fact, the only act of moral consequence that can happen within a game is the act of ending the game, denying its as-if character, spoiling the fantasy [...] (Castronova, 2004, pp. 188-189)
The urge to play is “buried very deeply in our psyches, well below rational thought and somewhat above the urge to eat and have sex”, and if the need goes unsatisfied, terrible things happen (Castronova, 2004, pp. 202-203). Virtual worlds, Castronova argues, are a great place to satisfy this need in a safe way. “They are worlds much like our world, and humans are beginning to spend many hours in them, playing games” (Castronova, 2004, p. 189). The only problem is that the real world is “seeping” into these worlds. This makes acts of “play” meaningful in the “real world”, spoiling the game and depriving humans of their right to play. Castronova concludes that impermeability of the magic circle should therefore be protected by law.
It is also common for works that do not explicitly invoke the magic circle to implicitly adopt a dichotomous perspective that resembles it. Richard Bartle frequently argues against outside interference in virtual worlds, particularly by regulators (e.g. Bartle, 2006). Castronova, in a later article titled On the Research Value of Large Games (Castronova, 2006a), sets “seeping reality” aside and claims that virtual worlds are comparable to entire societies:
- Until now, it has not been possible to take all of society as a research object [...] Thus, although we might believe theoretically, historically, and ethnographically that society operates a certain way, and we might have small-scale experiments that support our beliefs, it has generally not been possible to observe whole societies under controlled conditions. Now however, with the advent of synthetic world technology, it is indeed possible to replicate entire societies and allow them to operate in parallel. (Castronova, 2006a, p. 163)
As a demonstration of this research method, Castronova observes that players in EverQuest (SOE, 1999) and Dark Age of Camelot (Mythic, 2001) converge in certain meeting places even though no such place has been agreed upon beforehand; a result predicted by game theory. For Castronova, this “indicates that the theory of coordination games does indeed operate on a large-scale level in human societies” (Castronova, 2006a, p. 179). Castronova thus positions virtual worlds as something similar to computerised doppelgängers of Earthly societies.
The doppelgänger model is also visible in some of the first MMO-related papers in other disciplines. According to Bray and Konsynski (2007, p. 24), “Virtual worlds allow everyone to create a digital character and interact with other computer-generated individuals, landscapes, and virtually-run businesses. Both endogenously produced economies and social orders are emerging in these virtual worlds.” They suggest that virtual worlds are hosts to a great variety of virtual versions of real-world phenomena: for example, virtual business, virtual cities, virtual inhabitants, virtual citizens and virtual laws. However, Bray and Konsynski are open to the possibility of interaction between the real and the virtual world (ibid., p. 17), in contrast to the Galapagos-style isolation implied by Castronova’s virtual societies research program.
Bray and Konsynski’s work as well as other discussions referenced above suggest that the virtual world-real world dichotomy found in MMO-related scholarship can be broken down to a number of dimensions. For instance, one could identify the following distinct dimensions:
- 1. Virtual space vs. real space
2. Population of a virtual world vs. real-world population
3. Virtual identity vs. real identity
4. Relationships in a virtual world vs. relationships in the real world
5. Virtual institutions vs. real-world institutions
6. Virtual economy vs. real economy
7. Virtual law and politics vs. real law and politics
In the next part, I will examine each of these dimensions in more detail, reviewing well-known difficulties with viewing the situation as a real-virtual dichotomy, as well as pointing out a few new ones.
Challenges for the Dichotomous ModelVirtual Space vs. Real Space
Perhaps the most concrete dimension of the real-virtual dichotomy is space. MMOs are designed to simulate geometrical space in one way or the other. While users must necessarily remain on Earth, all their actions are directed towards these simulated spaces. Thus users and scholars alike have come accept the simulations as tantamount to actual space for many intents and purposes, with the necessary caveat of them being computer mediated.
But even at the core of virtual space, physical space cannot be ignored. Guilds in World of Warcraft (Blizzard, 2004; from hereon, WoW) and corporations in EVE Online (CCP Games, 2003; from hereon, EVE) recruit members based on the continent and time zone in which they reside in. For WoW raiding guilds it is important that members can be online simultaneously for extended periods of time. For EVE alliances engaged in war over territory, it is vital that members are available to keep guard at all hours. It is common for corporations to advertise that new members are sought in, for instance, Western Europe or U.S East Coast. In battles over space stations, warring parties try to cause engagements at times that are inconvenient for the opponent (Combs, 2008).
A more fundamental question is, where does virtual space end and real world begin? Besides the plains and planets of WoW and EVE, players make their presence known on discussion forums, chats, voice communication servers and video sharing sites. For example, Avatars United is a social networking site for MMO characters: not part of the MMO server, yet clearly an extension of its social playfield. Another example is the propaganda war between EVE alliances, waged everywhere from forums to YouTube. Johnson and Toiskallio (2005) describe how Habbo spills over to user-maintained fan sites and fanzines. If researchers limit their observations to the MMO server only, they certainly miss a lot of the space where the action is played out. The notion of “virtual space” is useful, but also ambiguous and multi-faceted rather than a monolithic counterpart to physical space.
Virtual Population vs. Real-World Population
The population of a virtual world is an important piece of information if one wants to compare real and virtual spaces, calculate per capita macroeconomic indicators, or generalise from samples to whole virtual populations. However, the population of a virtual world has proven to be a much more elusive concept than the population of a country. Many attempts have been made to define it (Castronova, 2002; Castronova, 2006a; Castronova, 2006b; Linden, 2003), but none have been conclusive and all have been subject to criticism (Shirky, 2006; Lehtiniemi, 2009). A key problem is that the amount of play that qualifies as “active participation” in an online service depends on the specifics of the service in question.
Virtual Identity vs. Real Identity
Individual identity is a frequently cited dimension of the real-virtual dichotomy. According to Castronova (2006b), virtual worlds “give you a freedom that no one has on Earth: the freedom to be whomever you want to be.” The underlying idea is that people have two identities, one for the real world and one for the virtual world. In practice, most MMO users discuss school, work, family issues and television programs in the fantasy environment. Even those who role-play necessarily bring their attitudes, education and 21st century interpretation of knightly values into the MMO, so the two identities are never disconnected. Identity also flows in the other direction: identity-forming experiences in an online game can shape one’s character in other contexts (Fung, 2006). Celia Pearce and Artemesia (2006) describe how the once-members of a closed down MMO called Uru adopted a strong “Uru diaspora” identity that persisted as they moved on to new MMOs, forming “ethnic Uru communities”.
An opposite of the “schizophrenic” real identity vs. virtual identity view outlined above is the “mirror” view of identity: the notion that an avatar is the virtual reflection of a person, the person’s virtual body through which they express their identity in the virtual world. Many observations can be summoned to support this notion. For example, Yee (2007) reports that players’ age and gender are linked to certain MMORPG character creation choices. In Second Life, it is not uncommon for users to attempt to make their avatar literally a digital double of their body. Still, this is not the complete picture. Users frequently have more than one avatar, and some users share avatars between each other (Ducheneaut & Moore, 2004; Taylor, 2006, p. 47). Users involved in the “spy metagame” in EVE Online maintain avatars that are ostensibly in conflict with each other (Combs, 2008). In Habbo, I observed some users maintaining several avatars and discussion board handles, managing their façade towards different peer groups by selectively revealing which avatars and handles they hide behind. Neither the schizophrenic nor the mirror view alone affords a complete picture of identity in MMOs.
Relationships in a Virtual World vs. Relationships in the Real World
Since the earliest scholarship on computer-mediated communities (e.g., Rheingold, 2000; Turkle, 1995), it has been widely accepted that friendships mediated by computer networks can be as deep and meaningful as those acted out face-to-face. But it is equally well recognised that distinguishing between “real world friends” and “virtual world friends” can be difficult. Players and guilds interact with each other in gatherings and gaming cafés while simultaneously participating in games (Huhh, 2008; Lin, Sun & Tinn, 2003). Friends, family members and relatives spend time together in MMOs (Taylor, 2006, pp. 52-55; Kolo & Baur, 2004). Pinckard (2006) describes how business acquaintances gather in WoW to enjoy their free time, discuss business matters and build networks, much in the same way as some businessmen do on golf courses. Relationships move freely between offline and online environments.
One reason why participants expand MMO relationships to the offline world is to increase trust and to reduce risks associated with cooperating with strangers (Lin, Sun & Tinn, 2003; Taylor, 2006, pp. 46-47). For example, EVE is a high-stakes game where allies may want to get quite intimate before relying on each other extensively. Such relationships are not easily described as either real or virtual. Another class of relationships that can expand outside the MMO are antagonistic relations.
Virtual Institutions vs. Real-World Institutions
According to Castronova (2004), an important benefit of MMOs is that they allow players to engage in behaviour that would not be possible for them outside the game. Castronova’s ideal virtual getaway is an MMO where real-world norms do not reach, and virtual behaviour is instead regulated by a new set of virtual norms and institutions. He regrets that in practice, this ideal is seldom reached: for example, secondary markets allow wealthy individuals to purchase powerful game characters. A similar case of real-world inequity seeping into the virtual can found in gender and MMO participation. Holin Lin (2008) describes a number of ways in which real-world institutions and norms regulate Taiwanese female gamers’ playing behaviour, from differential parental supervision at home to group dynamics of male and female gamer communities in cybercafés.
What of the virtual institutions that reign in the virtual world? An essential one is the MMORPG guild, which influences the behaviour of its members (and others) in many ways (e.g. Williams et al., 2006). But while a typical guild is a group of players playing the same game together, guilds also exist independently of any particular game, moving from MMO to MMO or participating in several simultaneously. According to CCP Chief China Representative Horace Xiong (personal communication, 16 May 2007), Chinese “trans-game megaguilds” can have up to a hundred thousand members. Their feuds and alliances transcend any single game, turning individual MMOs into parts of a larger metagame. According to Xiong, this metagame may even involve economic interests, such as gold farming market shares and deals with MMO operators. In summary, while “real-world” norms and institutions will always have influence on behaviour in MMOs, institutions that are ostensibly part of the virtual world can also have suprising real-world dimensions.
Virtual Economy vs. Real Economy
Virtual economies have been at the heart of the debate concerning the relationship between virtual worlds and the real world from the very beginning. Real-money trade of virtual property (RMT) is said to open a door for real-world rationalism and economic inequalities to seep into the virtual world (Castronova, 2004). A common retort is that not having RMT markets advantages those players who have more real-world time at their disposal. It can be argued that even without RMT, virtual economies would still not be isolated from the patterns of the “real economy”. Nash and Schneyer (2004) describe a situation where the release of an existing MMO to a new market created an influx of new players, which translated to a sudden rise in the number of low-level avatars, which in turn had a significant impact on the game’s internal economy in the form of supply and demand shocks. Nash and Schneyer also observed that the prices of certain goods oscillated as a function of the time of day, because Japanese players had different demand characteristics compared to North American players.
Virtual Law and Politics vs. Real-World Law and Politics
In the previous part of this paper, I summarised Lastowka’s (2007) reasoned argument that virtual worlds should be treated exceptionally by courts, because they are play spaces. But legal scholar Joshua Fairfield provides another prespective, which effectively refutes Lastowka’s dichotomy:
- At first blush, the magic circle seems to apply. The action took place in a “virtual,” and not a “real” world, and so legal liability should not follow. But upon careful examination, the distinction between “real” and “virtual” world fails. When a player enters a world in which player vs. player conflict is enabled, she consents to contact. (Fairfield, 2008, p. 18; emphasis added)
Fairfield’s point is that implied consent and community standards are (or should be) taken into account always when courts apply law; not just in the case of sports and games like sumo and WoW, but in the case of every industry and pursuit. Thus, while Fairfield agrees with Lastowka in that “rules of play” must be taken seriously by courts, he denies that there is anything out of ordinary or “virtual” about this. Moreover, rules of play are by no means unambiguous: different stakeholders have different ideas of what constitutes legitimate play in an MMO, and they support these ideas by referring to various other legal and moral arguments. The legal and moral regimes of MMO play are thus firmly part of this world.
As for virtual politics, MMO institutions such as guilds have given rise to fascinating governance issues and power struggles, which have inspired some scholars to go as far as to point to predictions of “social order independent of, and in several ways transcending, real-world governments” as well as “the waning importance of real-world governments” (Bray & Konsynski, 2007, p. 20). But the perspective afforded by Fairfield (2008) shows that at least for now, virtual world politics are strictly subordinate to real-world politics: they create rules of play that are interpreted by real-world judiciaries. When players or operators really want something, it is the national government they lobby (Yoon, 2008).
A Social World PerspectiveSummarising the above discussions, we find that any study that chooses to view an MMO as a world, a virtual one, standing apart from “the real world”, must take into account the following caveats in that model:
- 1. the space the virtual world occupies is not clearly distinguishable;
2. the population of the virtual world is ambiguous;
3. its inhabitants’ identities cannot be equated with avatars;
4. social relationships are not bounded by its limits;
5. outside norms and institutions regulate behaviour within it;
6. its economy is influenced by shifts in the real economy; and
7. its law and politics are shaped by outside processes.
Clearly, the notion of a virtual world begins to seem strained: a caveat can be found in almost every aspect of the concept. So far, the typical strategy for dealing with this difficulty has been to treat the caveats as “links” or “interaction” between the real world and the virtual world. This strategy attempts to address the issues whilst still clinging on to the dichotomous model that distinguishes between real and virtual. A few exceptions do not invalidate a model, but as the review above shows, the exceptions are beginning to pile up to such an extent that it is becoming difficult for a researcher to keep track of them. It is as if the dichotomous model did not fit together with empirical reality very well. Are there better ways to conceptualise MMO participation?
Some common themes emerge in the above discussions on the shortcomings of the dichotomous model. One is the relationship between people and space. The concept of a virtual world bundles together an MMO server and a set of social groups and institutions, expecting that their boundaries coincide and line up so perfectly that they can be subsumed into a single socio-technical entity, a society and its environment, a “world”. As the examples in the previous section illustrate, these boundaries do not necessarily line up at all. The social groupings and subcultures of an MMO frequently extend beyond the boundaries of the server to other servers, forums, platforms and physical spaces.
Another theme is the relationship between MMO users and other social groups and institutions. Families, business circles and gamer communities intersect with the MMO user base, and the MMO user base is crisscrossed by larger entities such as megaguilds, nationalities and movements in popular culture. The operator of the MMO is difficult to conceptualise in the dichotomous model: on one hand, it appears as a profit-making company providing a service to customers, while on the other, a suprememe government or god.
The first step towards a better model is to uncouple the technological platform from the user groups and institutions. This means stepping back and re-evaluating the relevant boundaries of the phenomenon under scrutiny. Do the boundaries of an entire computer-mediated society coincide with the edges of an EverQuest server, as suggested by Castronova (2006a)? Or should we also look at, for instance, guild servers and forum discussions as parallel modes of the same social world? The point is not to give up on boundaries altogether and let research lose its focus, but to avoid drawing artificial boundaries based on technological distinctions.
The second step is to conceptually reembed the MMO in the rest of society, from which it became detached when rhetoric turned it into a world of its own. Discussions above place MMO user groups alongside social groups such as golf circles and gamer communities. Indeed, scholars and marketers frequently refer to the user base of an MMO as “a community”, recalling Rheingold’s (2000) notion of a virtual community. This is an ill-fitting term for the user base, however. Rheingold’s virtual community as well as the classical sociological notion of community are characterised by familiarity, unity and even intimacy. A group of million or more individuals who happened to buy the same box or create an account on the same website will clearly not share those characteristics. An MMO server can act as an arena where communities gather, but some other concept must be used for theorising the larger user base.
To find a more fitting abstraction, I suggest looking at sociologist Anselm Strauss’s classical social world perspective (1978; 1984; 1997). Strauss was contributing to an interactionist stream of thought that had historically focused on either macro-level group encounters, e.g. those involving ethnic, racial and nationalistic groups, or the enormous proliferation of micro-level social groups which are not necessarily clearly boundaried or organised. His aim was to unify these approaches in a perspective that is holistic yet captures the fluidity of the social reality.
According to Strauss, social worlds are “universes of discourse” (Strauss, 1978, p. 121), the boundaries of which are set “neither by territory nor formal membership but by the limits of effective communication” (Strauss, 1978, p. 199). In the social world perspective, social reality as a whole is seen as consisting of numerous social worlds of varying size, which overlap, intersect and segment into subworlds. A typical individual belongs into several: their family, workplace, profession, hobby, religious community, drinking buddies, and so on. Worlds can be international or local, emergent or established, public or hidden, hierarchical or anarchic. The boundaries of the social worlds and the authenticity of one’s membership and activities are under constant debate and negotiation (Strauss, 1978, p. 123).
Though an MMO involves discourses and people, we cannot ignore concrete aspects such as technology, space and action. Strauss emphasises these tangible, observable qualities in the concept of social worlds:
- In each social world, at least one primary activity (along with related clusters of activity) is strikingly evident; i.e., climbing mountains, researching, collecting. There are sites where activities occur: hence space and a shaped landscape are relevant. Technology (inherited or innovative modes of carrying out the social world’s activities) is always involved. […] In social worlds at their outset, there may be only temporary divisions of labor, but once under way, organizations inevitably evolve to further one aspect or another of the world’s activities. (Strauss, 1978, p. 122; emphasis in original)
In MMOs such has EVE Online, participating can be a very complex and involved activity, giving rise to sub-activities, organisations and even new technologies arranged around the central activity. 
While the participants cannot all share unity and intimacy, they form a sphere of “effective communication” that connects them together. Their communications are mediated not only by EVE’s servers and discussion forums, but by third-party sites, online radio broadcasts, video streams and a quarterly print magazine covering events and persons on 84 full-color pages, engaging EVE users in a “universe of discourses” centered around the play activity.
From this perspective, EVE can be seen as a complete and independent social world that moreover hosts constellations of subworlds. This may not sound like a particularly significant claim, until we realise that it places EVE in the same category as London’s world of finance, the Judo world, or the world of game studies. From an individual’s point of view, EVE’s role in their life would thus be comparable to their professional world, neighbourhood or even extended family. Some of these other social worlds will involve some of the same people, activities, sites and technologies as the MMO, representing instances where the MMO social world intersects with another social world.
On the other hand, it is obvious that not every MMO platform will necessarily spawn a distinct social world that encompasses its user base; many services do not even aspire to do this as EVE does. For example, Habbo and Second Life are designed as open-ended arenas where participation does not center around a particular activity to the same extent. Instead, user groups are invited to use the service as a resource for their own activities. Some of these groups and activities are rooted in established social worlds (e.g. school friends in Habbo, academic meetings in Second Life), while others are newly formed online (see Ishii & Ogasahara, 2007, on “real-group-based online communities” vs. “virtual-network-based online communities”). This type of MMO is thus more accurately analysed as an intersection site of multiple social worlds rather than as a cradle of a single monolithic social world. The same MMO artefact occupies a different role in different worlds: it is a “boundary object” (Duchenaut, 2007). One social world to which the MMO artefact is tremendously important are the developers of that MMO. To them, it represents a target of common activity and a matter of constant negotiation with the world of users.
Implications for ResearchRichard Bartle defines “virtual” as “that which isn’t, having the form or effect of that which is” (Bartle, 2003, p. 1). By this definition, the proposition “virtual worlds don’t exist” is a truism, a pun. But I hope to have demonstrated that the proposition is also true in the second sense of the definition: that there are no “virtual worlds” that achieve the form or effect of the real world. 
Substituting the dichotomous “one server, one people” virtual world perspective with a social world perspective where people are uncoupled from technology allows us to point out certain shortcomings in the current MMO research agendas. The most obvious point is that an MMO server does not enclose a “world” or “society” separate from the Earth. An MMO server may be at the center of a social world, a universe of activity and discourse so deep that it feels like a different reality. But to a greater or smaller extent, the same individuals are simultaneously part of numerous other social worlds, which shape their identity and regulate their behaviour.
To point out one concrete target for my criticism, consider Castronova’s conclusion that observing apparent coordination effects in an MMO server indicates that coordination effects take place in human societies (Castronova, 2006a, p. 179). In principle, is it not possible that players voted for the meeting place on a discussion forum or at a guild meeting? Or that players who have avatars on multiple servers brought the custom from another server? The social world of the MMO server is tangled with and crisscrossed by other social worlds, and its boundaries are never definite. The experimental world and other worlds are simply not separate in a way that would justify considering the MMO server an independent human society. This is fatal for research programs that make such an assumption.
This does not mean that MMOs are useless for research purposes. MMO participants can be studied like any large group of people, and since their activities are largely though not wholly computer-mediated, obtaining data on them can be comparatively easy. For example, MMOs can be useful for studying economic decision making (Nicklisch & Salz, 2008), social psychology (Igarashi, Nagashima & Baba, 2008) and online consumer behaviour (Lehdonvirta, 2009). These research streams do not attempt to use the MMO as a metaphor for a society, or assume that the world ends where the boundaries of the server lie. Instead, they take the MMO as one particular kind of site where people interact, with its own features and peculiarities. 
Another necessary target for criticism is the way in which MMO related studies tend to make statements about a general category called “virtual worlds” and its two subcategories, game worlds and open-ended worlds (Bray & Konsynski, 2007, p. 19; Castronova, 2004). The social world perspective reminds us that universes of activity and discourse similar to those found around MMOs may also be found around other online arenas that do not necessarily meet the technical definition of a “virtual world”. In some aspects they may be less complex than MMO-worlds, but virtual economies, for instance, can be found in many kinds of services, from social networking sites to instant messaging systems. In fact, for many purposes, the closest comparison to a given MMO-world might not be another MMO-based world at all, but some other computer-mediated social world. For example, in some ways Second Life might be closer to Cyworld or even Flickr than to Entropia Universe or WoW. Therefore, researchers should not automatically adopt “virtual worlds” as a category to make generalisations to, unless their research somehow justifies it.
As a way of summarising the above discussion, I suggest that social scientists ask themselves the following questions to ensure their work is in line with their aims:
1) Out of all social world sites and technologies, what is the reason I am focusing on MMOs?
2) Out of all possible interaction modalities used by the social world under scrutiny, am I justified in limiting my observations to the MMO server only?
3) Do my results concern MMOs in general, a specific MMO, or some completely different category?
There are good answers to all three questions, but “MMOs are like virtual versions of the real world” is not among them.
Finally, the social world perspective highlights the fact that while notions such as goods, institutions, norms and politics can be identified over MMOs, they should not be considered “virtual versions” or “simulations” of “real-world” phenomena. Firstly, they are not doppelgängers but entities in their own right, sui generis. Secondly, parallel phenomena exist in several social worlds, computer-mediated and otherwise. For instance, virtual goods can be seen as a new category of commodities, parallel to but not derivative of clubs on a golf course or clothes in a mall. The “virtual” prefix in this case should be understood as signifying that they are computer-mediated, not that they are unreal or derivative. In most cases, the prefix is not even necessary: instead of “virtual politics”, it might be more instructive to talk about the politics among users and developers of an MMO.
Hopefully this realisation will allow scholars to begin retiring unnecessary references to “virtual” from their language - again. T. L. Taylor notes that “[i]n much the same way we now see the relationship between on- and offline life as not a bounded one, in many ways a game/not-game dichotomy does not hold” (Taylor, 2006, p. 19). Cyberspace isolationism is considered antiquated now; few people around me use the term “IRL” anymore. Dichotomous either-or views have been replaced with more nuanced understandings of how the Internet meshes into our lives (Dimaggio et al., 2001; Wellman & Haythornthwaite, 2002). Though Castronova explicitly denies connection with Barlow’s ancient declaration (Castronova, 2004, p. 208), it is hard not to see kinship in their ideas. Both were undeniably pioneers.
The Real World Doesn’t ExistThis paper could have been titled “The real world doesn’t exist”. If there are problems with the concept of the “virtual world”, so are there problems in the way “real world” is implicitly conceptualised in many MMO studies: as a uniform, monolithic reality, where people lead a rational “real life” with their unitary “real identity”. Such a view is in stark contrast to the views prevalent in contemporary sociology, which emphasise the multiplicity, fluidity and even fragmentation of identities (e.g. Turkle, 1995; Slater, 1997) and the often arational, constructed and “aestheticized” character of everyday life (e.g. Featherstone, 1991; Giddens; 1991).
Moreover, structuralist and post-structuralist theorists such as Pierre Bourdieu see rules and structures in all fields of life that are not unlike the written and unwritten rules of MMO-worlds. “Interest […] is to admit that the game is worth playing and that the stakes created in and through the fact are worth pursuing; it is to recognize the game and to recognize its stakes” (Bourdieu, 1998, p. 77). This game-like character of everyday life has not gone unnoticed in MMO studies: Castronova equates society with a large game, although he only sees one game instead of a multiplicity (Castronova, 2006a, p. 171). Conversely, MMO gameplay has in some instances come to resemble work: laborious, tedious and occasionally lucrative (Yee, 2006; Grimes, 2006, pp. 982-985). Malaby (2007) takes this line of thought furthest. He suggests that if we look at games as domains of artificial outcomes, of “contrived contingency”, we find that society is full of games: ones associated with business risk, others associated with political risk, and still others that relate to family matters. Increasingly, these old games are being played out online, sometimes even on the same arenas as the newer games, like EVE alliance warfare.
AcknowledgementsI would like to thank my colleagues Vilma Lehtinen and Mikael Johnson for inspiration and advice, and the three anonymous reviewers for useful comments on an earlier version of this paper.
Notes These new technologies range from simple calculators to elaborate organisational management tools such as human resource and enterprise planning systems. “[M]ost of the tasks that are handled by FCs [Fleet Commanders] are handled using tools and processes that are external to the game client software.” (Combs, 2008)
 Bartle actually seems to have considered a social world perspective, as indicated by the following quote: “a world […] doesn’t have to mean an entire planet: it’s used in the same sense as ‘the Roman world’ or ‘the world of high finance.’” (Bartle, 2003, p. 1)
 Consider also this example regarding survey research design: In an as yet unpublished survey of WoW users, sociologists asked respondents how much WoW interferes with their real life. The question contains the assumption that there is a “real life” which can be “interfered” by some separate set of “virtual” activities. But many WoW users play in order to spend time with their friends, family members and colleagues. For such users, the question is misconstrued and will not yield valid data. A better survey might ask about WoW’s role in various social worlds of which the respondent is part of.
ReferencesBarlow, J. P. (1996). A Declaration of the Independence of Cyberspace. Retrieved July 24, 2008 from http://homes.eff.org/~barlow/Declaration-Final.html
Bartle, R. (2003). Designing Virtual Worlds. Indianapolis: New Riders.
Bartle, R. (2006). Virtual Worldliness. In J. Balkin, & B.S. Noveck (Eds), The State of Play: Law, Games and Virtual Worlds (pp. 31-54). New York: New York University Press
Blizzard (2004). World of Warcraft. Blizzard Entertainment.
Bloomfield, R. J. (2007). Worlds for Study: Invitation - Virtual Worlds for Studying Real-World Business (and Law, and Politics, and Sociology, and….). Retrieved July 24, 2008 from http://ssrn.com/abstract=988984
Bourdieu, P. (1998). Practical Reason: On the Theory of Action. Stanford: Stanford University Press.
Bray, D. A., & Konsynski, B. R. (2007). Virtual worlds: multi-disciplinary research opportunities. ACM SIGMIS Database 38(4), 17-25.
Castronova, E. (2002). On Virtual Economies. CESifo Working Paper Series No. 752. Retrieved July 24, 2008 from http://ssrn.com/abstract=338500
Castronova, E. (2004). The Right To Play. New York Law School Law Review 49, 185-210.
Castronova, E. (2006a). On the Research Value of Large Games: Natural Experiments in Norrath and Camelot. Games and Culture 1(2), 163-186.
Castronova, E. (2006b). Virtual Worlds: A First-Hand Account of Market and Society on the Cyberian Frontier. In K. Salen, & E. Zimmerman (Eds), The Game Design Reader: A Rules of Play Anthology (pp. 814-863). Cambridge, MA: MIT Press.
CCP Games (2003). EVE Online. CCP Games.
Combs, N. (2008). A View of Politics and Morale in Eve-Online, Draft, Version 1.3X, Updated 5/14/08 (http://scratchpad.roaringshrimp.com/). Retrieved July 24, 2008 from www.roaringshrimp.com/scratchpad_docs/EvePoliticsMorale_1.3X_Research.pdf
Copier, M. (2005). Connecting Worlds. Fantasy Role-Playing Games, Ritual Acts and the Magic Circle. Proceedings of DiGRA 2005. Retrieved July 24, 2008 from http://www.digra.org/dl/db/06278.50594.pdf
DiMaggio, P., Hargittai, E., Neuman, W. R., and Robinson, J. P. (2001). Social Implications of the Internet. Annual Review of Sociology 27, 307-336.
Ducheneaut, N. (2007). Habbo Hotel as a boundary object (blog entry in Terra Nova, 12 October). Retrieved July 24, 2008 from http://terranova.blogs.com/terra_nova/2007/10/habbo-hotel-as-.html
Ducheneaut, N., Moore, R. J. (2004). Let me get my alt: digital identiti(es) in multiplayer games (position paper for The CSCW 2004 Workshop on Representation of Digital Identities). Retrieved July 24, 2008 from http://www.parc.com/research/publications/files/5551.pdf
EA (1997). Ultima Online. Electronic Arts.
Fairfield, J. (2008). Anti-Social Contracts: The Contractual Governance of Online Communities. Indiana Legal Studies Research Paper Series, no. 89. Retrieved July 24, 2008 from http://ssrn.com/abstract=1002997
Featherstone, M. (1991) Consumer culture & postmodernism. London: Sage.
Fung, A. (2006). Bridging Cyberlife and Real Life. A Study of Online Communities in Hong Kong. In D. Silver, & A. Massanari (Eds), Critical Cyberculture Studies (pp. 129-139). New York: New York University Press.
Giddens, A. (1991) Modernity and self-identity. Cambridge: Polity.
Grimes, S. M. (2006). Online multiplayer games: a virtual space for intellectual property debates? New Media & Society 8(6): 969-990.
Huhh, J.-S. (2008). Culture and Business of PC Bangs in Korea. Games and Culture 3(1), 26-37.
Huizinga, J. (2000). Homo Ludens: A Study of the Play-Element in Culture. Routledge: London.
Igarashi, M., Nagashima, Y. & Baba, A. (2008). Are Tendencies in Real-World Social Behavior Reproduced in the Virtual World? Investigation and Implications. Proceedings of the 39th ISAGA Conference, Kaunas, Lithuania, 7-11 July 2008.
Ishii, K., & Ogasahara, M. (2007). Links between Real and Virtual Networks: A Comparative Study of Online Communities in Japan and Korea. CyberPsychology & Behavior 10(2), 252-257
Johnson, M., & Toiskallio, K. (2005). Fansites as Sources for User Research: Case Habbo Hotel. Proceedings of the 28th Conference on Information Systems Research in Scandinavia (IRIS’28). Retrieved July 24, 2008 from http://www.soberit.hut.fi/johnson/Johnson_IRIS_2005.pdf
Kolo, C., & Baur, T. (2004). Living a Virtual Life: Social Dynamics of Online Gaming. Game Studies 4(1). Retrieved July 24, 2008 from http://www.gamestudies.org/0401/kolo/
Lastowka, F. G., & Hunter, D. (2004). The Laws of the Virtual Worlds. California Law Review 92, 1-73.
Lastowka, F. G. (2007). Rules of Play (paper presented at AoIR 8, October 17). Retrieved July 24, 2008 from http://terranova.blogs.com/RulesofPlay.pdf
Lastowka, F. G. (2008). Virtual Law Bibliography (blog entry at Terra Nova, March 17). Retrieved July 24, 2008 from http://terranova.blogs.com/terra_nova/2008/03/virtual-law-b-1.html
Lastowka, F. G. (2009). Planes of Power: EverQuest as Text, Game and Community. Game Studies 9(1). Retrieved September 11, 2009 from http://gamestudies.org/0901/articles/lastowka
Lehdonvirta, V. (2009). Virtual Consumption. Publications of the Turku School of Economics A-11:2009. Turku: Turku School of Economics. http://info.tse.fi/julkaisut/vk/Ae11_2009.pdf
Lehtiniemi, T. (2009). Measuring Aggregate Production in a Virtual Economy Using Log Data. Journal of Virtual Worlds Research 2(3). https://journals.tdl.org/jvwr/article/view/631/512
Lin, H. (2008). Body, Space and Gendered Gaming Experiences: A Cultural Geography of Homes, Cybercafes and Dormitories. In Y. B. Kafai, C. Heeter, J. Denner, & J. Y. Sun (Eds), Beyond Barbie and Mortal Kombat: New Perspectives on Gender and Gaming (pp. 54-67). Cambridge, MA: MIT Press.
Lin, H., & Sun, C.-T. (2007). Cash Trade Within the Magic Circle: Free-to-Play Game Challenges and Massively Multiplayer Online Game Player Responses. Proceedings of DiGRA 2007. Retrieved July 24, 2008 from http://www.digra.org/dl/db/07312.38207.pdf
Lin, H., Sun, C.-T., & Tinn, H.-H. (2003). Exploring clan culture: social enclaves and cooperation in online games. Proceedings of DiGRA 2003 (pp. 288-299). Utrecht: University of Utrecht.
Linden (2003). Second Life. Linden Research.
Malaby, T. M. (2006). Parlaying Value: Capital In and Beyond Virtual Worlds, Games and Culture, 1(2), 141-162.
Malaby, T. M. (2007). Beyond Play A New Approach to Games. Games and Culture 2(2), 95-113.
Meier, S. (1991). Sid Meier's Civilization. Microprose.
MindArk (2003). Entropia Universe. MindArk.
Mythic (2001). Dark Age of Camelot. Mythic Entertainment.
Nash, J., & Schneyer, E. (2004). Virtual Economies: An In-Depth Look at the Virtual World of Final Fantasy XI: Online. Retrieved July 24, 2008 from http://lgst.wharton.upenn.edu/hunterd/VirtualEconomies.pdf
Nicklisch, A., & Salz, T. (2008). Reciprocity and status in a virtual field experiment. Preprints of the Max Planck Institute for Research on Collective Goods 2008/37. Retrieved December 12, 2008 from http://www.coll.mpg.de/pdf_dat/2008_37online.pdf
Pearce, C., & Artemesia (2006). Communities of Play: The Social Construction of Identity in Persistent Online Game Worlds. In N. Wardrip-Fruin, & P. Harrigan (Eds), Second Person: Role-Playing and Story in Games and Playable Media (pp. 311-318). Cambridge, Mass.: MIT Press.
Pinckard, J. (2006). Is World of Warcraft the New Golf? (article in 1UP, February 8) Retrieved July 24, 2008 from http://www.1up.com/do/newsStory?cId=3147826
Rheingold, H. (2000). The Virtual Community: Homesteading on the Electronic Frontier. London: MIT Press.
Salen, K., & Zimmerman, E. (2004). The Rules of Play: Game Design Fundamentals. Cambridge, MA: MIT Press.
Shirky, C. (2006). The tech reporters who flack for Second Life (blog entry at Valleywag, December 26). Retrieved July 24, 2008 from http://valleywag.com/tech/naming-names/the-tech-reporters-who-flack-for-second-life-224092.php
Slater, Don (1997). Consumer Culture & Modernity. Cambridge: Polity.
SOE (1999). EverQuest. Sony Online Entertainment.
Strauss, A. (1978). A Social World Perspective. In N. K. Denzin (Ed.), Studies in Symbolic Interaction 1 (pp. 119-128). Greenwich, CT: JAI Press.
Strauss, A. (1984). A Social World Perspective. In N. K. Denzin (Ed.), Studies in Symbolic Interaction 5 (pp. 123-139). Greenwich, CT: JAI Press.
Strauss, A. (1997). Mirrors & Masks: The Search for Identity. New York: Transaction Publishers.
Sulake (2000). Habbo. Sulake Corporation.
Taylor, T.L. (2006). Play Between Worlds: Exploring Online Game Culture. Cambridge, MA: MIT Press.
Turkle, S. (1995). Life on the Screen: Identity in the Age of Internet. New York: Simon and Schuster.
Wellman, B., & Haythornthwaite, C. A. (eds.) (2002) The Internet in Everyday Life. Oxford: Blackwell.
Williams, D., Ducheneaut, N., Xiong, L., Zhang, Y., Yee, N., & Nickell, E. (2006). From Tree Houses to Barracks: The Social Life of Guilds in World of Warcraft. Games and Culture 1(4), 338-361.
Yee, N. (2006). The Labor of Fun. Games and Culture 1(1), 68-71.
Yee, N. (2007). The Demographics of Game Choices (blog entry at The Daedalus Project, March 20). Retrieved July 24, 2008 from http://www.nickyee.com/daedalus/archives/001558.php
Yee, N., Bailenson, J.N., Urbanek, M., Chang, F., & Merget, D. (2007). The Unbearable Likeness of Being Digital: The Persistence of Nonverbal Social Norms in Online Virtual Environments. The Journal of CyberPsychology and Behavior 10, 115-121.
Yoon, U. (2008). A Quest for the Legal Identity of MMORPGs - From a Computer Game, Back to a Play Association. Journal of Korean Judicature 1, 418-477. http://ssrn.com/abstract=1113327