OUR SECOND LIVES Social Reality and Fantasy in an Online Virtual World

Nicholas Watson 29 May 2008 Revised: 28 August 2008

Submitted in Partial Fulfillment of the Degree of Bachelor of Arts in Anthropology and Sociology, Knox College

ABSTRACT Second Life (SL), though similar to online computer games, is less a game than a platform for social interaction, entertainment and constructing virtual models of reality. Users come together to talk, drive virtual cars, shop at virtual malls, attend concerts, build homes and so on. The goal of this investigation, conducted through participant observation, interviews and document analysis, has been to analyze SL social reality and its interaction with real-world social life. The most salient SL user goals are identity-workshopping (Turkle), practical businesswork and education, socialization, entertainment and money-making. SL is an uncharted cultural space in which users mobilize cultural knowledge from an existing repertoire to collectively construct new knowledge about the virtual world. What results is an incomplete reflection of real-world cultural material and activities, and both micro- and macro-level conflicts arise as understandings of the social space compete. In-world discourse is permeated by language of empowerment and its limitations. It suggests that SL offers empowering and creative opportunities, but virtual interactions have the potential both to reverse and reinforce existing power relations.

ACKNOWLEDGEMENTS I would like to thank professors Breitborde, Eberhardt and Amor, who served as my mentors and advisors throughout the various stages of this project, for their listening, guidance and insights. I am also grateful for the contributions of professors Singer and Haslem. I would like to thank the classmates who were part of my research group for the interest they took in my work and for their valuable input. I am grateful to my in-world contacts CarolLynn, Ivy, Jill, Drewcyla, Lucina, and Tom, who have generously given of their time to share their stories with me and to answer my many questions. Finally, a special thanks goes to Jason, Devin and Alex. You three have had to listen to me prattle on about this project, day and night, for the better part of a year. I appreciate you humoring me for this long.

CONTENTS INTRODUCTION ........................................................................................................................ 1 LITERATURE REVIEW .............................................................................................................. 3 Social Control and the Limits of Virtual Reality ........................................................................ 3 The Potential of Virtual Social Worlds....................................................................................... 5 Online activities and Cultural Material....................................................................................... 6 RESEARCH DESIGN .................................................................................................................. 9 Overview..................................................................................................................................... 9 Design and Method ................................................................................................................... 11 ANALYSIS .................................................................................................................................. 12 Building a self: the virtual identity .......................................................................................... 14 Hanging Out: the nature of social interactions ........................................................................ 27 Building a Society: collective discourse about SL ................................................................... 35 CONCLUSION........................................................................................................................... 44 Revisiting Cultural Toolkits...................................................................................................... 45 GLOSSARY OF TERMS ............................................................................................................ 47 BIBLIOGRAPHY ........................................................................................................................ 50

Quotations herein that have been adapted from text-based chat or Internet forums are superficially edited for mechanics—typos and capitalization errors are usually fixed (unless they are somehow significant to the quote). In some cases, a single quote was originally typed across multiple lines of chat (i.e. the user typed a phrase, pressed the ‘enter’ key, and then typed the next phrase), but they have all been collapsed into a single contiguous utterance, with chat tags removed.

INTRODUCTION Second Life (SL) is a graphical online virtual community in which users’ avatars1 meet, chat, shop, dance, swim, watch plays, build virtual houses, go to virtual museums, fall in love, and buy and sell virtual land. Although SL is often thought of as being related to online roleplaying games like World of Warcraft, some of its participants are loathe to call it a “game.” They see it as more of a platform, world or society. Reid (1999) notes that the text-based virtual worlds of the late 90’s, called MUDs, existed in two types: adventure MUDs and social MUDs. Adventure MUDs had a game focus and were usually computerized forms of tabletop Dungeons & Dragons games, in which users wandered around the virtual world slaying monsters and collecting gold. Social MUDs, on the other hand, were not game-oriented, but were focused on social interaction between users and on creative opportunities to expand the virtual world through building and programming. If MMORPGs like World of Warcraft are the modern, graphical versions of adventure MUDs, then SL is analogous to the social worlds of the MUD era. In order to participate in SL, a user downloads the client program to her computer from the SL website. She registers an SL account2, chooses a name for her SL persona, and logs in online. Once she is connected, the virtual world is displayed on her screen and she can navigate her avatar through it using keyboard and mouse commands to walk, fly (as if with an invisible jet pack), or magically teleport between locations. She can type messages to other avatars she encounters, and can also express a range of non-verbal gestures through avatar animation.

1

See Glossary of Terms for an explanation of this and other technical terms indicated in boldface. A basic account is free, but basic users receive no stipend and cannot own land. Premium account holders pay around US $10 per month for the ability to own land, premium customer support and a weekly stipend of virtual “game money” that can be used to purchase virtual items in-world.

2

Watson 1

By the numbers, SL has about six million unique users registered for accounts (although not all of these log in regularly), and on average fifty thousand users concurrently logged into the system at any given time. One of the key differences between SL and other online role-playing games is the ability for users to create new virtual objects by programming them into the world itself. Furthermore, users retain intellectual property rights to their virtual creations, even though the objects themselves are stored in computers owned by Linden Lab, the corporate provider of SL. Users can sell their creations or buy the creations of others using an imaginary in-world currency called Linden Dollars (L$). Finally, and most remarkably, these virtual Linden Dollars can be exchanged for cold, hard, real-world cash using the LindeX exchange market, at a rate of about L$250 to US $1. In-world economic transactions total over US$1 million a day. A few virtualworld entrepreneurs even make their real-world livings from their SL virtual businesses. The goal of this investigation, conducted through the methodologies of cultural anthropology such as participant observation, interviews and document analysis, has been to construct a broad-scope analysis of social reality in SL.

Watson 2

LITERATURE REVIEW This work follows two decades of journalistic and sociological research in virtual communities. The reviewed Net-studies literature falls into roughly three categories. In the first group are those researchers who have investigated virtual communities as social worlds and seek to explicate the system of social controls and cultural constraints that govern action there (Dibbell 1993; O’Brien 1999; Smith 1999; Reid 1999; Taylor 2006). A second angle of analysis considers the potential for social worlds to be empowering and democratizing, and the limits to such potential (Turkle 1994, 1995; Webb 2001; Castronova 2005). Finally, the third body of work asks how users of virtual worlds construct understandings of online activities and cultural material (Webb 2001; Taylor 2006). For my own analysis, I also draw heavily on Swidler’s (1986) work on “cultural toolkits,” which does not consider virtual worlds, but provides a general theoretical framework that helps in developing an understanding of online culture.

SOCIAL CONTROL AND THE LIMITS OF VIRTUAL REALITY With the emergence of the Internet as a communication medium, and the increasing prevalence and complexity of online communities, a considerable amount of journalistic and popular work touted virtual worlds as idealistic spaces in which users could have complete freedom of identity, freedom of movement and freedom of expression. They would be released from the constraints imposed by cultural norms, real-world inequalities and overt social domination. Much of the scholarly work on communities in cyberspace through the late 1980s and 1990s (Dibbell 1993; O’Brien 1999; Smith 1999; Reid 1999) has found, to the contrary, that (1) real-world cultural norms and social constructs, such those associated with gender and race, are carried over into virtual worlds; (2) certain real-world inequalities are recreated in virtual

Watson 3

spaces; and (3) hierarchical social controls do exist in virtual worlds, and serve to constrain the range of acceptable speech and actions. O’Brien (1999) responds to hype that suggests that the lack of physical markers online would lead to a freedom from the constraints of socially constructed gender norms and stereotypes, and the ability to adopt fluid identities. O’Brien shows that virtual spaces— specifically chat rooms, MUDs and message boards—have instead been mapped with the same categories of gender (and of race, age, etc.) as the physical world, and that these categories have not lost their salience to users, nor their ability to constrain people’s actions. Taylor (2006) shows that existing stereotypes of race and gender have been echoed in the choices of avatar available to players in EverQuest, a swords-and-sorcery MMORPG. Elizabeth Reid (1999), drawing on the work of Allucquère Rosanne Stone, demonstrates that speech and actions online are not free and unconstrained, but are in fact governed by a complex system of hierarchical social controls. The MUDs that Reid studied have varying styles of governance, from quasi-democracies to full dictatorships, and social control is generally enforced through a hierarchical system of powers and privileges. Dibbell (1993), drawing on first-hand journalistic research into the aftermath of an infamous “virtual rape” incident on a social MUD called LambdaMOO, concludes that actions and speech online, despite being nothing more than text signals and bits of digital information in actuality, have real social consequences—that is, words are deeds. Dibbell found that maintaining a stable, safe social environment in LambdaMOO was accomplished only through a complex negotiation between parties with differing political/ideological views, including libertarians, anarchists, parliamentarian legalists, and even royalists.

Watson 4

Smith (1999) finds in MUDs systems of social control similar to those described by Reid and Dibbell. She constructs an understanding of virtual worlds in terms of conflict models of society, cites various inevitable sources of conflict in online communities, including differing ideological views, interests and expectations, as well as perhaps, in some cases, pure willfulness. Conflict exists between individual users, between groups of users, and between users and administrators. Maintaining stability in a MUD is frequently a constant struggle between users and administrators (1999: 147). Smith shows that schemes of domination and punishment are only partially effective in managing conflict, whereas mediated negotiation can yield more promising results. Taylor (2006) shows that the same systems of social order also exist in more modern, graphical virtual worlds, including MMORPGs like EverQuest. A network of internal and external organizational structures locate and contextualize the user’s role, power and status in the virtual community. The researchers above have all focused on the existence of cultural norms, social organizations and social controls in the virtual world, and all have generally found that these structures do exist in virtual societies and that they impose limits on the freedom of users to say, be, and act.

THE POTENTIAL OF VIRTUAL SOCIAL WORLDS Despite the fact that social conflict and constraining cultural norms persist in cyberspace, thus preventing it from providing the sort of idealized utopia once imagined, numerous researchers have focused on the potential for these kinds of online virtual worlds to empower participants or enrich their lives.

Watson 5

Turkle asserts that, while it may be impossible to completely escape from social constructs like gender, virtual worlds can act as “identity workshops” for the “construction and reconstruction of self” (1994: 158, 160). Turkle notes the possible therapeutic value of such online spaces, and takes computer-mediated social interactions to be “evocative objects” (1994: 160; 1995: 22) that allow users to reflect on identity and social life by playing out certain aspects of identity and interpersonal relationships not available to them offline. Webb writes that part of the appeal of “avatar culture” is that it “binds people together temporarily and then frees them up to relocate themselves elsewhere” (2001: 587). Thus, the measure of freedom and experimentation that virtual worlds do provide can be useful and meaningful to participants. Castronova (2005) suggests that the economies of virtual worlds may have the potential to answer the ages-old “social question” about how to best deal with the fact that the distribution of resources in society is unequal. Castronova identifies a shift towards a “sheerly communicative economy” (2005: 5) in which objects that exist purely as information (that is, virtual objects) may have as much value as real, physical objects. Working from the assumption that access to imagination and communication tools is fair and equal, Castronova implies that virtual economies are inherently more democratic and meritocratic than real-world ones, and that their growing importance will lead to an overall democratization of wealth.

ONLINE ACTIVITIES AND CULTURAL MATERIAL The virtual worlds of interest in this study are those that are more than just purely communicative spaces—more than just newsgroups and chat rooms—those that produce some other cultural material in the medium. The work of Taylor (2006) and Webb (2001) considers how online activities and culture are informed by connections to offline life.

Watson 6

Taylor (2006: 18) finds that “the second wave of Net studies […] elaborated ways in which the online world was not a tidy, self-contained environment but one with deep ties to value systems, forms of identity, and social networks, and always informed by the technological structures in which it was embedded.” Webb posits a “tight fit between off- and online worlds” in which virtual worlds are “loosened from the traditional organs of social conformity but at the same time fissured by offline political and ideological values” and ultimately play out “humdrum variations of our own [offline world], with both virtual and ‘real’ being a semblance of true reality” (2001: 589-590). Webb further suggests that users in online virtual worlds create narrative content to fill the gap left by the absence of non-verbal communication and what Kiesler calls the “dramaturgical weakness of electronic media” (1984, quoted in Webb 2001: 569). Narrative that is created to fill this gap can be somewhat chaotic because “the need for communication is constant and intense […] a feature of late modernity, the notion that to be active is a normative social requirement and that boredom is the enemy of cultural life” (2001: 586).

CULTURAL TOOLKITS The work of Swidler (1986) provides a framework for understanding how SL users draw on cultural knowledge to shape their actions (and activities) online. Swidler challenges the notion that actors are motivated primarily by explicit ideology or rational self-interest. Instead, she suggests that in “settled” cultural periods, actors draw on an existing repertoire or “toolkit” of cultural knowledge and strategies available to them in order to shape action (1986: 273). In “unsettled” periods, old cultural ends are cast aside and explicitly articulated ideological ends play a key role in establishing new strategies that become part of the cultural toolkit (1986: 278).

Watson 7

Cultural lag is the notion that in a settled period, cultural strategies persist and continue to provide opportunities for action even though they have outlasted the specific ideological ends for which they were constructed.

Watson 8

RESEARCH DESIGN OVERVIEW The focus of this study is Second Life as a virtual social world. Through this project, I seek to understand virtual communities as a phenomenon of social life, the interplay between physical-world social life and virtual social life, and the significance that it has for participants. At the outset, I sought, as much as possible, to employ the tools of cultural anthropology to develop a narrative explicating local practices and meanings. I metaphorically compared SL to the far-away African village or island tribe that is often thought of as the traditional/typical subject for cultural anthropologists. In other words, I considered SL less as an activity undertaken by members of our own society and more as a bounded society and culture in its own right. Of course this analogy is quite limited: it quickly became apparent that online culture is inextricably related to understandings of offline. As Taylor (2006: 18) has suggested, it is impossible to understand one without the other. However, the analogy remained useful in developing the narrative dimension of the analysis. Moreover, the key realization is that SL can be better understood both as an activity and as its own social world (albeit one firmly enmeshed within the real-life3 social world) rather than just as an activity. This is consistent with the approaches of many of the researchers listed above, including Dibbell, Smith, Reid, Taylor and Castronova. Throughout data collection, I considered the hypotheses of the researchers reviewed in the previous section. Their theories were based on research conducted in other virtual worlds similar—but not identical—to SL. I have used these to guide my own research: one of my tasks 3

“Real life” is commonly abbreviated as “RL.” The word “real” should not be taken too literally. See the “Real life” entry in the Glossary for further explanation.

Watson 9

has been to determine the extent to which these findings are generalizable and can be applied to SL. Where my findings seem not to fit with those of previous researchers, I have noted the contradictions and used their ideas as jumping-off points to construct more generalizable theories about the phenomena under study. I began my research with a number of sub-questions that I sought, throughout the investigation, to answer: •

What is the range of personal experiences in SL, and the associated personal backgrounds?

•

In what ways have these experiences acted as evocative objects (Turkle 1995), mechanisms for discovery of self and other, or otherwise personally enriching experiences for these individuals?

•

How do SL users construct their avatars’ identities in SL? Do they engage in “identity workshopping” (Turkle 1995)?

•

What sorts of real-world inequalities, stereotypes or cultural norms do SL participants see replicated or recreated in SL, if any?

•

What kind of cultural material is (re)produced in virtual worlds, and why? o What sorts of real-world physical objects, real-world popular culture objects, and aspects of social reality are recreated/reflected in SL narratives and objects? What significance do they have for participants? o What sort of real-world activities or customs are mimicked in SL?

•

Do the personal experiences and systems of meanings indicate any conscious or unconscious goals or ideologies? Can SL, in the view of these participants, meet these goals? Is SL seen as empowering or disenfranchising?

Watson 10

•

How do SL participants perceive the relationship between the virtual and the physical, between online and offline lives? To a large extent, this project is preliminary and exploratory, working to establish the

breadth and depth of material available for analysis. The data have uncovered numerous avenues of potential inquiry that I will not address here. I have ultimately focused on three broad areas of analysis: the construction of identities through avatar interaction, the way social interaction is carried out on a micro level, and the way a larger SL society is conceptualized and constructed.

DESIGN AND METHOD Research was conducted using three strategies. First, I designed an avatar named Herschel Leakey and conducted participant observation in-world, seeking to experience the range of activities available to users. In part I explored SL on my own, taking in the sights and examining the creations that others had built. I also observed and participated directly in conversations and interactions. Secondly, I conducted in-depth interviews with six SL residents, in order to uncover salient in-world categories for further investigation, and to give locals a chance to articulate their own understandings of the topics at hand. Finally, I conducted document analysis on a number of supplemental cultural texts, including in-world “notecards” written by users (often, for example, instruction manuals distributed with products for sale), as well as external websites and blogs.

Watson 11

ANALYSIS To consider SL as a social world, one must first determine what brings people there. My findings suggest that the most salient user goals and activities in SL are identity role-playing, practical business work (meetings, etc.), education, socialization, entertainment, creative work and entrepreneurship in the hopes of making a real-world profit. In this analysis, SL is seen as a largely uncharted social space, a new medium of interaction in which modes of action and the cultural frameworks for understanding action are not well-defined. Participants in SL populate this space with virtual objects and spaces that reflect incompletely on real world objects. SL houses, for example, look like elegant real-world houses, but they often do not come with bathrooms since avatars have no need of such facilities. Participants also mobilize cultural knowledge from their “cultural toolkits” or repertoires of preexisting knowledge in order to shape understandings of SL social life and of virtuality in general. They rely on this knowledge to form strategies of action. Sometimes, explicit ideology is imported to SL from RL experience. As a result, the primary vehicle for virtual meaningmaking, formation of customs and establishment of social norms is the set of pre-existing cultural frameworks from RL. Virtual social life is mapped according to the same cultural scripts that govern real-world life. However, these cultural frameworks are not universally shared, and overlap to a varying degree, such that different users often develop different understandings of the virtual world. SL becomes a contested space in which different understandings compete for dominance and lack of consensus breeds conflict. As users develop strategies for negotiating around these conflicts, they collectively create new knowledge about the social space, and even create and enact new social norms.

Watson 12

A discourse of empowerment and its limitations is implied and explicitly stated in both the general study of virtual worlds and in the construction of knowledge about SL by its participants and leaders. In-world discourse, as well as theoretical work (Turkle 1994, 1995; Castronova 2005), suggests that virtual worlds like SL offer empowering opportunities to users—opportunities for therapy, discovery of self and other, transformation of social inequality, education, entrepreneurship, participation in political discourse and collective construction of knowledge. However, the data shows that while virtual interactions have the power to offer all of these opportunities and to reverse existing power relations, they also have an equal power to reinforce the status quo and to silence dissenting voices. The analysis of SL social life is split up into three sections: Building a Self considers how online identity is constructed and reconstructed through the avatar and the social persona. Identity role-playing and workshopping offer opportunities to experiment with alternate social roles and to discover aspects of the self. These activities can serve both to undermine and reinforce existing social categories and power relations—in particular, gender roles. Conflict occurs when users have fundamentally differing understandings of the social role of the avatar. Hanging Out examines the nature of social interactions between multiple individuals on a micro scale. This section considers some of the RL customs that are entrenched in SL social interaction despite the fact that they serve no practical purpose. It also addresses conflicts that can arise as a result of disinhibition, in which the user takes the virtual simulation as opaque and becomes detached from an understanding of the social costs of action. Building a Society examines the macro-level collective dimension of SL, particularly in terms of political discourse concerning SL governance, policy, and the responsibilities of its

Watson 13

leadership. In this sphere, SL users bring a range of political ideologies to the table. The language of empowerment and disenfranchisement is of utmost importance here, as it is used not only to conceive SL’s fundamental purposes but to critique the political discourse itself. Some members of the community assert that discussions of policy are collective discussions including a plurality of voices, but others feel that their voices have been systematically excluded from these discussions. Following the analysis, the conclusion discusses the overall efficacy of the cultural toolkits model based on Swidler’s work for understanding SL social life. It briefly considers the implications of inconsistencies between Swidler’s model and the findings presented here.

BUILDING A SELF:

THE VIRTUAL IDENTITY

That interacting in a virtual world via an avatar necessarily alters one’s conceptualization of the self may seem obvious, but the ways in which virtual identities are constructed and understood are more difficult to grasp. One of the appeals of worlds like SL is the ability to be— and look like—anything you want. Webb frames this ability in terms of an “avatar culture” that “binds people together temporarily and then frees them up to relocate themselves elsewhere” (2001: 587), while Turkle suggests that in virtual worlds “the self is not only decentered but multiplied without limit” (1994: 159) and that these spaces can serve as “identity workshops” for working through identity issues, playing out aspects of a multifaceted personality, and in general “the construction and reconstruction of self” (1994: 158). The data from SL seem to validate these claims, but they also make apparent that users experience the construction of the virtual self in widely varied ways. Discourse about the construction and reconstruction of self reveals opportunities for empowerment as well as sources of conflict between users in online

Watson 14

communities. Furthermore, while existing theory tends to align with contemporary understandings of postmodern, fluid identities, some SL users experience identity in essentialist terms.

Typology of the Self The “self” in SL can be thought of in terms of three components. First, there is the immutable pseudonym, chosen by the user at his SL birth as any first name he types plus a surname chosen from a limited list of available ones4. The name can contain a limited number of signals of identity, since it can sound male, female or gender-ambiguous, contain a pun or literary reference, or otherwise identify a user with a particular subculture or set of interests. For example, one SL educator and RL physics professor chose the surname “Plasma,” a word wellknown to physicists. The second component of the self is the avatar, the graphical, virtual standin for the physical body on the computer screen. Most avatars appear as male or female human beings in a variety of shapes, sizes and colors, with varying hair styles and modes of dress. However, avatars are infinitely customizable, so a user is free to be an androgynous human or even a cat, dragon, or genderless floating disembodied orange sphere. The avatar is a formalized aspect of identity in that its existence is coded directly into the grid, but unlike the pseudonym, the avatar can be changed at a moment’s notice: a quick virtual makeover can change age, sex, height, weight, and species. The final component of identity is the social self, the personality, ethos, reputation, background story and personal mannerisms that the user establishes for herself

4

Linden provides a list of about 30 available surnames to choose from. The list changes on a regular basis and usually incorporates some kind of theme. When I signed up, most of the names were conveniently related to anthropology – “Leakey,” “Afarensis,” and so on. It is unclear why surname choices are limited, but there may be a technical reason, or perhaps Linden wants to encourage real or human-sounding names as opposed to the outlandish self-descriptive and/or alphanumeric handles (e.g. “CAHockyFan2008”) that frequently appear in chatrooms.

Watson 15

through interactions with others. The social self can be fictionalized, doctored and made over, but not, as will be seen, without limit or consequence. Significantly, the typology above is somewhat artificial: different users conflate the three components of identity to varying degrees. Furthermore, users often do not have a single online self but engage in a “cycling through” of identities (Turkle 1995). There are many ways in which, to use Webb’s wording, a user can free himself to relocate elsewhere. He can make over his avatar several times in one day, present two vastly different social selves to two different online acquaintances, open an alternate account (“alt”) in SL with a different name and appearance, or participate in another virtual world altogether. Moreover, he can do many of these things concurrently, such that he is not cycling identities every day or every hour, but every handful of seconds as he switches between different program windows in which he has different virtual personae active. In short, users have the freedom to be many things in the virtual world that they are not in real life.

Identity Workshops One of the ends to which this freedom can be used is what Turkle calls identity workshopping, in which the user creates a virtual identity in order to experiment with aspects of her self, or with social roles not available to her in RL (thus, role-playing). As Turkle shows, identity workshopping on MUDs can frequently be applied to therapeutic ends—that is, the user comes to understand some aspect of himself or becomes more comfortable in a certain kind of RL social situation, as a result of his identity experimentation and role-playing in the comparative safety of the online world.

Watson 16

The experiences of residents in SL show that this is also true of SL. Jill is a high school English teacher in RL, but has been an SL resident for about a year: I read about it in the newspaper […] I first logged on just to explore and poke around...looking to see what the world was like. I was also intrigued by something the article said about people having virtual sex. I like sex… but virtual? How could THAT be fun? Wanting to see what all the fuss was about, Jill registered an account, visited virtual shops and collected a lot of freebies—virtual avatar enhancers like shapes, skins, and body parts—and started looking for places where users could have sexual encounters. She found them, but became quickly bored with these casual encounters that had “no good interaction” to go with them. After becoming disenchanted with the interaction at one such venue, she flew to a nearby mountain region for some peace and quiet while she tried to figure out how to open some of the boxes that her freebies came in. I met this guy there who seemed nice, did not have his penis hanging out, and did not open his conversation with "want to have sex?" We got to be friendly…lots of sl sex but also conversations… GOOD conversations. […] I learned with this guy that, with the right conversation, you can establish a kind of sexual link with another person here. A GOOD one!! Well, I became something of an SL slut for a while but I have backed away from that lately. Just to enjoy the good friends that I have made here. Thus, Jill’s second life began with a transformation of her understanding of what sorts of interpersonal connections were most important to her, and thus she reports, “My interest in SL has morphed from being primarily erotic release to primarily social.” But her experimentation had an even more significant result: “I have been able to explore my nascent bisexuality here. It is good for me…has allowed me to be more comfortable in my rl skin.” I asked Jill to elaborate: When I was in my sl "slut" phase, where I was hooking up with almost anyone just to see how it worked, I decided to open up a bit with the avs [avatars]...try to get beyond the surface...really connect. […] But as I got to open up with people i began to meet others through them... began linking up with avs which i came to believe really ARE female... and I talked with them openly, because if they were truly female, and were fucking other females here, then I wanted to know where their feelings were. […] What I am saying is

Watson 17

that I met women who were in the middle of the sexual continuum and were quite happy with it... who had been with other women rl... like i felt i wanted to be... the more I chatted, the more comfortable I became about being open to other women in rl... . This is identity workshopping in action. That Jill was able to experiment with as well as talk about her attraction to other women in SL enabled her to come to terms with that aspect of her real-world identity. Ivy provides a different example of identity role-playing. In-world, the owner and designer of a virtual educational science museum is embodied in the form of a woman, but the user behind the avatar is, biologically, male: I am male, and happy to be male in RL, but I decided to experiment with a female persona in SL and I started having so much fun that I keep this persona. But beyond having fun, she also notes some practical uses for gender experimentation online: I think it is a really good thing to be able to explore the role of the opposite sex. If you were to b[e] a nice looking female avatar for a while you would learn a great lesson about how crude some pickup lines are LOL5 […] so experimenting with sexuality can lead to some real appreciation…like walking a mile in a persons shoes. […] the women at work all talk to me much more openly now that they know Ivy. Later on, she adds that “every college guy should be a woman in SL for at least a day... they would then see how bad male behavior really is.”6 Ivy’s words suggest that identity workshopping has an additional use beyond a personal working-through of aspects of the self: identity workshopping has the potential to transform social relations on a much broader scale, between groups of people. Men who have spent a day as women online have a taste of some of the social realities that women have to deal with on a regular basis, and as a result are ostensibly more cognizant and sensitive of those realities in their interactions with women. By logical extension, this kind of sensitivity could help deconstruct

5 6

“LOL” is chat short-hand for “laughing out loud.” Ivy adds, “Not all males’ behavior, just the worst 10%.”

Watson 18

the unequal power relations and eliminate the mistrust that might otherwise come into play in gendered interactions between men and women, both online and offline. But if the role-played virtual self has the potential to transform the self and society in such profound ways, it also has as much potential to reinforce the status quo by playing out everyday stereotypes in mundane ways or taking them to extremes. O’Brien (1999) argues that, in spite of popular hopes that Internet communities would usher in a new age in which visual markers like gender and race would cease to matter, the virtual world is largely being mapped according to the same cultural scripts that are active in face-to-face interaction. In terms of the cultural toolkits paradigm, these scripts are being used because they are the ones that are already available to participants. Furthermore, identity workshopping doesn’t always help, as “Rather than being nullified or erased, boundary transgressions etch the boundaries deeper into the collective conscience” (1999: 84). O’Brien observes a trend of “hypergendering” in virtual worlds: Far from being a wonderland of imaginative creativity, participants tend to ‘wear’ gender features that replicate conventional gender stereotypes of sexuality and desirability. In other words, they reproduce themselves as Barbie and Ken. (1999: 87) The question becomes whether this is true of SL. My own observation suggests a definite pattern: most of the female avatars appear young and rather well-proportioned (though only occasionally ridiculously exaggerated), and many wear noticeable amounts of makeup or jewelry. They clothe themselves in anything from jeans and a T-shirt, to semi-formal attire, to large flowing dresses that echo the fashions of bygone centuries. Men also appear young, and moderately buff bodies are not uncommon, while makeup, jewelry or any other traditionally female accoutrements are. It is worth noting that the default male avatar—the one all male users have before they customize it—is a tall, broad-shouldered, muscular young man. “Wearing

Watson 19

gender” seems ubiquitous enough that being attractive may come to be experienced as a normative social requirement. In an interesting twist of events, science fiction writer William Gibson, who coined the term “cyberspace” over twenty years ago and who is often seen as a visionary who foretold the age of virtual worlds to come, paid an undercover visit to Second Life. He recounts the experience later in an interview: [W]hen you do succeed in finding a group of other avatars, people aren't very nice. […] they’re meaner than they are in the real world. […] a part of my frosty reception was that I set all of the avatar's sliders in the opposite direction than I assumed most people would do. So I wound up being this grotesquely overweight, bright blue smurf. In a tutu. Nobody thought that was cool. (Nissley 2007) However, hypergendering or otherwise constructing an idealized avatar is not a universal. One resident suggests that, In a world where everyone can be perfect you still see black, white, oriental, fat, short, tall, old, young, bald, even confined to a wheel chair. I would say most of these are reflections of their real life. However, she also adds that “even if we are modeling the SL avatar on our RL avatar it is natural to omit those aspects we dislike.” Lucina, another resident, thinks of her avatar as being based on her real self, with some tweaks: My human av… I guess would be the idealistic version of me lol. I have red hair green eyes and fair skin like my av...she just the 'me' I'd like to be in terms of looks I guess. She also suggests that this kind of relationship to the avatar is common for other users: I think there is alot of real me [“real them,” referring to other users]...but also alot of who I'd [they’d] like to be...kind of like who they would be if they wouldn't be judged...if that makes sense. Like they would wear a certain hair style but it wouldn't fly at work kind of thing. Jill, in designing her own avatar, initially started by hypergendering, but then scaled back to an only slightly touched up version of her real self: Initially, I tried to make myself a very exaggerated version of myself... red hair, HUGE boobs, Barbie body... I was looking to attract attention... […] As time went by I wanted

Watson 20

my av to look like I really look...though after a few months in the gym LOL. So I wanted to look less like Barbie and more like Jill. In general, many of the users I spoke with shied away from a wholehearted endorsement of hypergendering, suggesting that realism and authenticity are worthwhile goals, but also felt that it was natural to take advantage of the freedom to improve their looks. The data shows that users may feel a social pressure to have attractive avatars.

Role-playing and Deception Online gender-bending can become a highly charged topic. ‘Net studies history abounds with stories about the virtual lynchings experienced by those who pretended to be someone they weren’t online. Turkle (1995) and O’Brien (1999) observe that when a user develops a close, trusting relationship with another online, and it turns out that the other is “really” of the opposite sex, there is a tendency for the user to feel deceived or violated. The Linden-published book Second Life: the Official Guide encourages role-play but warns that “friends don’t like to be deceived” (77), illustrating the point with a cautionary tale quoted from a female user: I wouldn’t have minded if he told me he’s really a guy within the first few weeks. But when he told me after six months, wow, it just blew me away. I just find him impossible to trust after that. (77) There is an important distinction between playing a role and constructing a façade. Playing a role means adopting a certain online persona and acting out that character in most interactions. This is not likely to upset others. A façade exists when a user consistently, self-consciously and in no uncertain terms provides deceptive information while claiming that it is real-life fact. This is controversial at best. In face-to-face interaction, there are challenges to our notions of identity, to how we categorize others. For example, transsexuals, as well as non-transsexuals who have the ability to

Watson 21

“pass” as the opposite sex, challenge our understandings of what is “really” male or female. However, such challenges are usually few and far between. Online, they are a matter of course, something that often must be grappled with from day to day: the importance of categories such as gender, age, ethnicity, even profession comes into question when all of this information can be hidden and defies verification. Even though it is not the only relevant factor, gender is usually the “big one” here. It is seen as a very basic aspect of identity and seems to come into play in how users relate to and interact with each other. As one user put it, …if you couldnt be honest with someone about something as basic as your gender...then I don't think you should be pursuing a relationship at all. I suspect that gender is considered so important because it is one of the most visible markers of difference in real-world interaction, because it implies notions of masculine and feminine discourse (which themselves imply power relations), and finally, because romantic and sexual relationships between users are common, and it goes without saying that one’s partner’s gender is usually considered important in such contexts. Consequently, different users employ different strategies that govern the disclosure of their own “true” identities as well as how they understand their relationships with others who may or may not be, in real life, what they appear to be online. These strategies constitute and imply certain ways of relating to others online and are necessarily developed to help users navigate the challenges of a social environment that poses new challenges. While there was a range of responses among my contacts, most seemed to try to strike a balance between honesty— the desire to not upset others—and respect for others’ right to role-play. Ivy articulates one such response:

Watson 22

I am open about who I am in RL and SL, all a person has to do to find out my RL sex is to ask, and I will tell them I agonized over that decision, I want to be honest and I also am involved in SL which has a component of role playing game. I am now comfortable with my personal rule. Several times I have had long term relations with other avatars and when they asked I told them. I am in the same long term relationships now. However this is a game for many people so I think it is OK for other people to make the choice to completely separate their RL and SL identities. In general, people are not bothered by the idea of others role-playing, including genderswitching, as long as they know the facts about their close friends. As one user expressed, “I don't really think it matters...as long as you aren't 'getting involved' with someone.” Another said, It does matter to me, I know the people I am close to are gender correct… . [T]here has to be a level of trust to get close, even in sl. We share personal things, rl issues and just have real friendships here. When I asked Jill if she was concerned that her friends might not be as their avatars appear or as they claim, she had a different response: Not very [concerned]...there is a distance here...and i don't think you can demand that level of openness... I present it, and let my friends know it, but I do not demand it. Part of the sl contract with each other, I suppose […] There is only one friend whom I believe accurately portrays herself. But it doesn't matter. It's the brain behind the av that interests me. Some users also pointed out that playing an identity different from one’s own has social costs and can be a significant investment of energy: I personally wouldn't do it because well frankly I dont understand guys and I think it would totally be obvious I was a girl lol. Approaches to dealing with role-played identities also imply ways of understanding close relationships in the SL context. Some users think of their online and offline friends as contiguous groups, the only difference being that they interact with some friends face-to-face and with others online. In these situations, the real-world identity of such friends tends to be important, and once users meet in real life, online relationships can become offline relationships

Watson 23

played out in more traditional face-to-face contexts, as with one user who originally met her current boyfriend and one of her best local friends online. She explains, “I just see it as a new generation of meeting people.” However, for others, there is a sense that “what happens in SL stays in SL”—that is, there is no expectation that relationships should translate into the real world. They are fundamentally separate spheres of interaction and in some cases no inherent conflict is perceived in having, for example, two different lovers, one online and one offline. In these cases, the biographical facts about the person behind the keyboard tend to be less important, and role-played personae are the focus of the relationship. What is important to understand is that the virtual world presents those who are emotionally invested in it with more frequent challenges to how they understand the meaning of self and the meaning of what it is to truly know another person than does offline daily life, and the worldviews that these challenges undermine are deeply significant and often deeply personal. Online interactions are problematic because they are mediated and informed by technology, but they appear not to be. There is a tendency to take virtuality at face value, to believe that we have accomplished more than we have through it, to conflate the artificial with the real (Turkle 1995: 238-239). This can lead individuals to have false expectations or incompatible understandings of the same interaction or relationship. Because the virtual world supplies no cultural scripts that suggest systematic or normalized ways to deal with such challenges, the responses—the ways in which each user mobilizes cultural knowledge from her own particular repertoire—are extremely diverse and even tend to vary on an individual basis, with little community consensus. The effect is that conflict occurs in the virtual world when the ways people have come to understand and deal with identity role-playing and the status of close relationships in SL inevitably differ in fundamental ways.

Watson 24

Ideas of a Core Self and the Limits of Workshopping According to Turkle, in MUDs, “projections of self are engaged in a resolutely postmodern context. […] the self is not only decentered but multiplied without limit” (1994: 159). The idea of the online self as fluid, decentered and constructed through its actions is easy for theorists to accept given that it aligns with contemporary post-structuralist notions of the destabilized self. Furthermore, it provides an effective model for explaining how identities are constructed and reconstructed online while taken at interface value all the while. This model applies to SL as much as it did to MUDs. However, it is worth noting that some SL users themselves experience identity in essentialist terms: they deconstruct the simulated identities around them, returning to the notion of an essential or core self. For them, this turns out to be an effective—and perhaps comforting—strategy for responding to the challenge of the destabilized self, and uncovers limits to the therapeutic value of identity workshopping. When Jill says that “it’s the brain behind the av that interests me,” she is invoking this notion of a core self. For her, it represents an effective response to the uncertainty caused by the inability to be sure that others are supplying factually accurate information about themselves. Drewcyla also expresses a belief in the notion of a core self. Drewcyla’s avatar appears as a woman in leather with white hair, white kitten ears and a tail. She has two other avatars, a white tiger (non-humanoid—a literal tiger that walks on all fours) and a siren (in the mythical creature sense), and she cycles through these occasionally. Drewcyla sums up her own theory of the avatar as follows: SL allows people to get to know each other in a way that is not possible in the real world. You see, my shape, my skin, hair eyes, all of it, it’s trimmings, a package that can be changed at will, so the only constant, about anyone in sl, is them. You can change how you look, but if you’re an ass, well....nothing can help that. So it lets people really know each other.

Watson 25

Later on, I asked Drewcyla to elaborate a bit on her ideas: Drewcyla: It is the persona that you interact with, the trimmings just give you a framework and context to interact with/in. HL: You've got "trimmings" in RL too, right... Drewcyla: clothes, cars, homes....yes HL: Does SL make it easier for people to get beyond the trimmings to the people underneath, because they know that the trimmings are so immaterial and changeable here? Drewcyla: Well it makes it easier to do that yes, but I doubt most people know or understand that. Too many people hold that it is easy to be a completely different person in sl. And they think that by changing their skin or even a new AV they are "different" when they really havent changed at all. The extent to which core self is linked to real-life biographical details varies depending on the user. For Jill, such details are as trimmings themselves: although one’s real life age or sex cannot be changed at will as the avatar can, they can be concealed at will, but this concealment does not change or restrict access to the “brain” or genuine persona behind the avatar. For Brad, by contrast, such details are important indices of what the person behind the avatar is really like. It is evident that Drewcyla, Jill and Brad all experience identity in essentialist terms, to some degree, but they have different reasons for doing so and different strategies for explaining or justifying their position. Brad’s approach is along the lines of a traditional assertion that certain aspects of identity, such as gender, are inextricably linked to notions of who a person is. Ironically, Drewcyla and Jill are using the analytical tools of postmodernism—namely, the realization that identities are fluid and physical and social markers are merely socially constructed trimmings—to deconstruct and ultimately reject the postmodern notion of the destabilized self. They find that the fluidity of identity just ends up pointing back to the existence of a stable core self. In her theory of the avatar, Drewcyla identifies something that appears to be an empowering feature of SL—the ability to “be” whoever you like—and establishes its limitations.

Watson 26

To believe that the trimmings can fundamentally change one’s identity, she suggests, is a form of false consciousness. At the same time, by emphasizing the ability to change trimmings in creative ways, SL calls attention—at least for some users—to the ways in which trimmings, online and offline, shape the ways we interact. To the extent that this encourages people to deconstruct the meaning of trimmings, to “really know each other,” it can itself serve as an empowering feature—what Turkle (1995) calls an “evocative object” for thinking about the concept of identity.

Summary Virtual worlds like SL are identity workshops in which users can experiment and cycle through different roles constructed through the graphical avatar and a set of social interactions. Identity is fluid and destabilized. Identity workshopping has the power to deconstruct existing cultural norms through boundary transgressions, but can also reinforce them when the virtual world is mapped according to pre-existing cultural categories and users exercise their limitless freedom to play out the status quo. Conflicting understandings of honesty, authenticity and roleplaying arise as people develop differing understandings of the online self. However, strategies exist to navigate these conflicts. One such strategy is to deconstruct identity workshopping by invoking the notion of a core, essential self, which seems effective even though it contradicts theoretical understandings of identity.

HANGING OUT:

THE NATURE OF SOCIAL INTERACTIONS

While building, scripting, playing games and making money are certainly the focus of SL for many users, it is primarily a social world. Users converse recreationally with close friends

Watson 27

and strangers alike, in their homes, in parks, at clubs and so on. They also engage in more practically-oriented interactions: Ivy is an SL educator who works at a real-world science museum and constructs virtual versions of those exhibits in SL in an effort to provide a multimedia interactive experience that reaches a wider audience. SL has also been used for corporate business meetings and collaborative commercial projects. Communication in SL is most often text-based7 and is carried out through the "local chat" channel, in which a user types a message that can be read on the screens of all other avatars within the (virtual) vicinity. Users can also talk in one-on-one instant message windows, which are like cell phone calls in that they are private and can be carried out regardless of the distance that separates the avatars. Aside from chatting, the single most common recreational activity in SL is, without a doubt, dancing. At social gatherings, users invoke scripts that animate their avatars in dance motions, from traditional ballroom-style partner dances to modern party dancing and even completely fantastical dance styles that would defy gravity if they were carried out in RL. Often there is music playing at these gatherings, streamed from an Internet radio station, although people do not always dance to the same tune: at one club I witnessed a couple slow-dancing while a particularly lively rock song was playing. This section considers the social construction of avatar interaction in terms of the ways in which different RL social norms are experienced to varying degrees in SL.

7

Voice chat is also available in some areas, but it is not widespread and not generally necessary to engage fully in social interaction with others. Consequently, I did no investigation related to voice chat.

Watson 28

Recreational Interaction The recreational social scene represents the sense in which SL is an extended and glorified chat room. People come together for essentially the same reason: to chat, about current events, politics, movies, art, school, things to see and do in SL, the latest fashions, and all manner of other topics. However, the interaction itself is enriched by the complexity of the virtual space. In traditional text-based chat channels such as those provided by America OnLine or on Internet Relay Chat (IRC) networks, users converse by typing messages that are conceptualized as spoken words. However, a chatter can also send another kind of message, called an emote, which can be used to indicate a gesture or state of being, or to otherwise narrate events. For example, in an IRC chat, if I type, "/me laughs at the rabbit's antics," other users will see a narrative message: "Herschel Leakey laughs at the rabbit's antics." The "/me" emote command works in SL text chat channels, but SL also provides other tools for sending such signals using the avatar. If I want to nod in response to someone's statement, instead of typing "/me nods" I can use another command to actually animate my avatar and cause it to nod. Chatters do not just talk. They also construct a narrative about the virtual/imaginary spaces in which they are talking, and about the actions of the conversational participants. IRC chatters tend to construct an imaginary physical space through emotes. For example, if I emote that "Herschel Leakey sits on the couch," in a single statement I have both established the existence of an imaginary couch and my position on it. With SL the couch is virtual but not imaginary. It can be seen and touched, and saying that I'm sitting on it does nothing, whereas right-clicking on it and selecting "sit here" will cause my avatar to actually appear seated on the couch.

Watson 29

This raises an interesting question: why do chatters construct such narratives? Why are they not satisfied simply to talk? Webb (2001: 569) notes that interaction through electronic media always entails a narrowing of the dimensions of communication (“bandwidth”) that results in what Kiesler calls the “dramaturgical weakness of electronic media.” To illustrate, there are many actions we carry out in face-to-face interaction beyond transmitting strings of words: there is intonation and body language, and in many cases the physical setting of a conversation and the objects nearby become part of the interaction in subconscious ways—whether the participants are sitting or standing, eating, fidgeting with their hands, and so on. When communicative bandwidth is narrowed, Webb suggests, participants attempt to fill the narrative gap with virtual approximations of or stand-ins for those aspects of face-to-face interaction that are missing. Thus emotes, narratives, avatar gestures and so on are all part of an apparatus of compensation for dramaturgical weakness.

Pulling Up a Seat To illustrate this point, we can consider a particular conversational space on a parcel of land owned by New Citizens Incorporated (NCI), an organization that exists to provide support for new SL users. The following fieldnote excerpt describes the setting (see also Figure 1): The NCI campus is laid out along a beach-front. A large boardwalk runs parallel to the water, some thirty feet back, and is open on one side to the beach, and bounded on the other side by a wall displaying billboards and class schedules. About halfway along the boardwalk, four benches are arranged in a rectangular formation, two with their back right up against the beach-ward edge of the deck, and the two others arranged perpendicular to these, facing inward. Two other benches bench are further away, against the wall, separated by a wide corridor that has been left open so that avatars can walk back and forth along the boardwalk. Even though NCI is expansive and there are plenty of benches and tables elsewhere, this area is usually the center of lively interaction. Conversation ebbs and flows, as users come and go over several hours. When a small group of people is sitting and chatting, others wander by and sit on one of the four close-grouped benches to join the

Watson 30

conversation. Each bench seats two avatars side-by-side, so as more people come, the benches fill up quickly, leaving some avatars standing—the other two benches are generally not popular sitting spots because they are awkwardly separated from the group. However, on numerous occasions, avatars who were standing were invited to "pull up a seat." This is a simple matter of rezzing a box to sit on, but in one case, an NCI officer who had the power to move NCI-owned furniture moved the two further benches up close to the rest of the group to provide a place for newcomers to sit.

Inviting others to sit down when they join the conversation is not a foreign concept. It is a simple rule of politeness, and we tend to feel bad if a participant in our conversation is left standing because there are no seats available. That this same rule applies online is not insignificant. At NCI I was invited to sit on a bench, at Drewcyla's forest retreat a rock or a pool float, in Tom's home a large black leather couch. None of this sitting was practically necessary—avatars do not need to sit because their legs don't get tired. Yet I was usually invited to take a seat, and glad to do so because standing while others were seated seemed awkward. Pulling up a seat in the virtual world is for social, not physical, comfort, and this demonstrates the extent to which, in order to fill the communicative gap in online interaction, the avatar has been fundamentally conceived as the social stand-in for the physical body, and must obey the same rules that constrain the physical body in face-to-face social interactions. The RL cultural framework that governs social interactions persists within the virtual world.

Watson 31

Figure 1: The conversation benches at NCI beach. The avatar in the yellow hardhat is me, Herschel Leakey. The bench on which Herschel is seated is one of the benches that is usually separated from the main group. It is usually next to the other bench against the “SL INFO” wall, but had been moved some time shortly before this picture was taken.

Disinhibition and the Simulation Although users populate the virtual space with certain social norms and conventions imported from RL, the medium itself prevents some normative forces from being felt as strongly in SL. Smith (1999: 156) and Reid (1996: 170; 1999: 111) observe a phenomenon of disinhibition among users of virtual worlds, by which users are more likely to act in a socially disinhibited and potentially transgressive manner online than they otherwise would in real life. In face-to-face interaction, we are constantly (consciously or unconsciously) calculating the potential outcomes of our actions, considering how others will react and what consequences we

Watson 32

might face. Online, however, people are likely to be less fully aware of the social costs of action, and they tend to feel more disconnected from the consequences. Disinhibition is an important part of some of the empowering dimensions of virtual life: identity workshopping, addressed in the previous section, hinges in part on the extent to which individuals are willing to experiment online in ways that they would be unwilling to try in RL. However, disinhibition can also be a source of danger or conflict. In her MUD studies, Reid found that users were all too eager to provide her with personal information without considering the consequences, whereas Smith found that people were more likely to act in inappropriate and disruptive ways online than they would be in analogous situations offline. This is true also of SL: for example, Ivy's story above has already indicated the ways in which attractive female avatars are harassed more intensely online than they would be in RL. The disinhibition effect is probably the result of online anonymity, the perceived sense of security that results from the physical distance that separates Internet users, and the perceived sense of privacy that comes from participating in a mediated interaction from the comfort of one’s own home. There can also be a tendency to experience the world as opaque, to accept the simulations at face value. This can result in seeing it as a game, and avatars in it as being objects or artificial intelligences like the kind that populate single-player video games. Drewcyla suggests that: People forget that the other avatars walking around are also people. Its easy to tie those pixels to a bot or AI and ignore it. Jill relates another experience in which the opposite effect seems to have occurred. Here, the danger was that taking the world at face value—accepting the experience as more real than it is—hid the real social consequences of actions:

Watson 33

There were times at the beginning when I let myself get too emotionally wrapped up... Not anymore. I am ok with myself. Sometimes [in the past] i let the "game" rule me. I do not let that happen. It is a danger in this environment...you are dealing with real people here. Disinhibition is a fact of SL social life, and while it can be a facilitator of interaction and identity-building, it also creates conflict and carries a social risk. SL is not a risk-less medium: social actions have social consequences, and for the emotionally involved participant, such consequences are not so simple that they can be escaped by logging out. Some participants remain unaware of the social costs of disinhibition, but others, like Jill in the above example, successfully negotiate solutions by constructing new, self-conscious knowledge about the risks of the medium. Such knowledge leads to new, and ostensibly healthier, understandings of SL social life.

Summary Social interaction in Second Life can be recreational (chat, dancing, etc.) or practical (education, business). Linguistic communication is supplemented by a variety of narrative tools such as emote commands, avatar animations, and scripts that allow interaction with virtual objects. These narrative tools respond to the dramaturgical weakness that results from the narrowing of communicative bandwidth. Because it is conceived according to existing RL cultural frameworks, social interaction is constrained by certain social conventions that apply in face-to-face interaction, such as inviting newcomers to a conversation to sit down. However, some social imperatives are felt more strongly than others: some users will feel disinhibited online, making them less attuned to the social costs of action and therefore more likely to break with norms or transgress across barriers. This can be a source of conflict that necessitates the negotiation of new strategies for understanding social life.

Watson 34

BUILDING A SOCIETY: COLLECTIVE DISCOURSE ABOUT SL The collective dimension of SL represents a contested, negotiated space in which different ideologies compete for recognition and a part in building societal infrastructure and shaping policy. Many of these ideologies are imported from the RL worldviews brought to the table by participants, but they are also informed by experiences within SL, including encounters with structural deficiencies and inequalities. At issue are both the quality of social and technical infrastructure and the extent to which rule-making is responsive to residents’ concerns. Rulemaking governs what activities are permissible in SL and how they are carried out, and also how inter-user disputes are resolved. Sociopolitical discourse in SL is often framed using the language of empowerment and disenfranchisement. One branch of in-world discourse—that adopted by official Linden statements—suggests that discussions about policy and direction are collective negotiations in creating new knowledge about the virtual world and how best to run it. However, opposing discourse makes it clear that not everyone feels included in such collective discussions. First, it is worth noting that none of my on-the-ground contacts shared any particularly strong opinions about these public issues. At least half of them expressed little interest in what they perceive as SL politics. Others, when asked, said that they feel that Linden Lab is doing an admirable job of managing SL given the daunting challenges they have faced. The overtly political discourse I encountered was largely confined to public spaces like the Linden official SL blog and accompanying comments forum (see Linden Research, Inc. 2007). While I am not suggesting that my sample of non-politically-active contacts is representative of the SL community at large, the sample of users who post on public forums is not necessarily

Watson 35

representative either. This shows, at the very least, that political issues are not universally salient for users.

Bringing Political Ideologies to the Table Collective discussions of virtual world politics are not new or unique to SL. As early as 1993, Dibbell (1998: 19-20) observed several competing ideologies of virtual world governance in LambdaMOO, a popular MUD of that era. Political orientations ran the gamut from parliamentary legalists, who believed in the democratic rule of representatives elected by a majority vote, to royalists, who believed in the right of those who owned and maintained the MUD’s server to dictate the terms of governance. There were even techno-libertarians, who felt that each individual was responsible for looking after his own virtual well-being, and that the less government involvement there was in the everyday affairs of users and in inter-user conflicts, the better. A similar spectrum of political views exists for interpreting today’s virtual worlds, which tend to be commercial enterprises, whereas LambdaMOO was (and still is) a non-profit community space maintained by volunteers and supported by donations. Most MMORPGs like World of Warcraft are conceived as corporate spaces—analogous to the company-owned mining town of the mid-20th century8—in which the maintainers of the game have legal ownership over all of the virtual space, including users’ avatars and their virtual “possessions” (this most closely parallels the royalists of LambdaMOO). Significantly, this has been the traditional paradigm for understanding ownership and governance in corporation-owned virtual worlds, and thus SL broke with tradition when it conceived a new world in which a user’s creations are legally her

8

This metaphor was original proposed by Jenkins (2004).

Watson 36

own intellectual property and a formal system exists through which she can turn her virtual assets into real-world wealth.

Social and Economic Dimensions: Promises and Inequalities In fact, putting ownership of at least part of the virtual world back in the hands of the individual user is one of SL’s main selling points. In a published interview as part of a public relations podcast9, Linden founder Phillip Rosedale called virtual worlds “fundamentally empowering,” saying specifically of SL that …the key enabling power that it has for people, is that you can create anything you like, and you own it, and there’s an economy that allows you to monetize it to make money on it if you want to. (Rosedale 2007) The possibility of making real money from virtual creative labor is attractive, which is why Rosedale says that SL is fundamentally a world for innovators, entrepreneurs and artists. However, some thinkers question whether SL really does offer egalitarian financial opportunities, given that most of the wealth is controlled by a handful of elite businesspersons like “land baron” Anshe Chung who nets an estimated US $150,000 per year from her in-world real estate business—an accomplishment remarkable enough that she made the cover of Business Week magazine in 2006 (Rymaszewski et al., 2007: 253). Some land barons set up “landbot” programs designed to automatically search for and snap up parcels of virtual land being offered at below-market prices, thus preventing the average user from ever getting a good deal on land. In 2007, one prominent blogger and financial analyst went as far as to call SL a pyramid scheme, claiming that the economy is structured such that those at the top continue to rake in cash while the average user cannot realistically expect to make any significant profits by conducting 9

A podcast is an episodic series of electronic media broadcasts distributed via the Internet. The word can also refer to one specific episode of the series. In this case, the series is called Inside the Lab, and it serves as a vehicle for official Linden public relations releases.

Watson 37

business in SL (Harrison 2007). One SL user responded by suggesting that the RL economy is no less unfair or disenfranchising than SL’s, while another wrote that making profits is only part of the package: I make and sell virtual swords, in an online world. It just may be geeky, but it feels great that I can actually say I do that as a job. I'm sorry if all SL boils down to some people is making money. To me it's alot more. (Denton 2007) Even so, critics note another serious obstacle to entrepreneurs, one that undermines the language of empowerment and perhaps even users’ ability to take pleasure in economic activity: counterparty risk. Put simply, you can seldom trust those with whom you’re doing business in SecondLife. Even supposedly well established, well regarded business citizens are prone to defaulting on any obligations which prove inconvenient. Whole banks will disappear over night, along with your L$ balance. Private businesses will simply refuse to make good on financial contracts. And individuals, pretty much all of whose real world identities are carefully guarded anonymous secrets, sometimes even will openly default, without recourse. In an similar vein, early on in my fieldwork I attended an NCI class on land ownership, which warned newbies of crooked land owners who would scam unwary customers and noted that Linden Lab rarely involves itself in any kind of mediation or enforcement involving business transactions between individual users, even when such transactions involve unethical or illegal activities. Some SL participants respond by asserting that Linden rule-makers should establish measures to protect users from being cheated. Others—libertarians, perhaps, or strong believers in the free market—maintain that Linden should not be involved in such issues.

Collective Discussion, Responsive Government The upshot of all this ideological competition is that SL becomes a space into which participants import their pre-formulated political ideologies, which in turn shape their

Watson 38

understanding of how virtual society should be run. In a sense, people try to play out their utopian visions in the virtual world, although they may not experience it in these terms. Thus, SL is empowering not just because we can make virtual toys, money and friends there, but also because we can have a part in rebuilding society as it should be. Naturally, notions of society as it should be are varied and engaged in competition. To return briefly to Swidler’s analysis of ideology, she says that “structural opportunities for action” determine which of a set of competing ideologies survives in the long run (1986: 273). However, such structural factors are not transparent in SL and no particular political orientation seems to have the upper hand at this point. Rosedale frames these competing ideologies in terms of a collective discourse between groups coming from different cultural backgrounds. Discussions about what should and should not go on in SL are … collective discussions that include people from, well, just about every country in the world […] huge numbers of people from dozens of different countries that have very different perspectives on what people should and shouldn’t be doing on the Internet, so, there’s a clash I think and an interesting challenge… . (Rosedale 2007) This statement was made in response to a question framing the 2007 SL gambling ban10 as the “encroachment […] of the real world on the virtual world.” Clearly, not all SL users believe that these discussions are truly collective or that (referring to another of Rosedale’s statements) SL users are “all in this together.” Writes one (presumably European) user in response, “We weren’t in this together when Linden Lab decided that ALL of us needed to follow the antigambling laws, and we weren’t in it together when only the EU-Europeans had to be

10

Prior to the gambling ban, SL participants were able to bet Linden dollars on games of chance in-world. The ban was enacted because given that Linden dollars can be exchanged directly for real world money, SL gambling would constitute real gambling by most legal standards, thus making it an activity of questionable legality depending on who was gambling, who was hosting the games, and where both were geographically located.

Watson 39

V.A.T.’ed11… .” These examples show that a Linden-endorsed discourse of empowering, collective discussion is opposed by some sections of the public who respond by expressing their sense of disenfranchisement.

Governance That Responds To Users’ Needs Another common criticism of Linden Lab is that it is not responding to the participants’ need for a stable, reliable platform. In early 2007, faced with exponential growth of its user base and inadequate resources to accommodate such growth, SL suffered from numerous technical difficulties involving server downtime, buggy software and glitches that prevented large numbers of users from being able to connect (Rosedale 2007; Linden Research, Inc 2007). These events frustrated many users. At the end of 2007, Rosedale said in his podcast that one of Linden’s goals in the coming year would be to improve reliability: Obviously we need to make Second Life more stable, and more reliable […] I think of Second Life as needing to become a kind of public utility. I think it needs to be something that people can rely on being there […] a million dollars a day in the economy there’s a lot of people making their real lives better in Second Life and we’ve got to make it stable and reliable for them. In spite of his promises, for some users, the damage is already done. Between reliability issues and continuing bitterness over the gambling ban and the VAT, some participants seem to have lost faith in SL’s leadership. One response to Rosedale’s podcast reads: Glad to see LL is working on what’s important, reviewing the year that they progressively screwed up SL, and made most of the community angry with so many different things, man I just can’t wait to see this laugh-riot. Another, latching onto Rosedale’s statement that SL is facing “growing competition,” suggests, “ummm, maybe because they [the competition] listen to their users.” These users are expressing

11

Also in 2007, land-owning users located in the European Union became subject to a Value Added Tax that Linden was required to attach to land use fees paid to the corporation.

Watson 40

two forms of disenfranchisement: on the one hand, they feel that their need for a stable, reliable environment has been ignored; on the other hand, they feel shut out of the policy-making process. It should not be taken as given, based on this evidence, that Linden is a dictatorship indifferent to its users’ needs, attempting to placate an angry public with empty promises. About half of the user responses to the 2007 year-end podcast have something positive to say. Some applaud Linden for managing the 2007 challenges as well as they did considering their limited resources, while others are critical but remain optimistic. One respondent in particular congratulates Linden on its support for and involvement with educational programs in Second Life—programs like Ivy’s virtual science museum, which is actually a project undertaken by a real-world science museum to recreate virtual versions of some of its exhibits. Admission is offered free of charge, and Rosedale notes that “you’ve got a broader audience now, you’ve got people that are from all over the world. I mean [in] how many parts of the world are there science museums?” Education programs like this one become part of the discourse of empowerment because they bring education to audiences who otherwise have limited access to it, and they bring it into what is otherwise a recreational and commercial space12.

Taking Action Without The Government It should be noted that not all collective action is engaged on the level of government and political ideology. As previously mentioned, Linden does not generally involve itself in disputes between users. Sometimes, however, when faced with “on-the-ground” concerns not tied 12

The medium itself, from a technical perspective, also offers benefits for education. Ivy says that virtuality makes it easier to model scientific concepts and build interactive teaching tools, although she also notes that it is no substitute for the real, physical hands-on experience.

Watson 41

directly to ideology, groups of users are able to take collective action against a transgressor themselves. Drewcyla, who owns about four acres of virtual land, tells of how she and the neighboring landowners in her region worked together to buy up the “ad farms” in the area in order to prevent advertisements from being placed there. Sometimes, because of the way land is divided, small parcels of land (on the order of 16 square meters) remain in between the larger properties. Advertisers buy up this land and build large billboards that loom over neighboring properties, plugging some SL business or other (“sex, shops, phone sex, etc.” in Drewcyla’s words). Such advertising sites are known as “ad farms” and because they are unsightly, obstruct views, and may even display content that some users would find unpleasant or objectionable, they are regarded as a detriment to nearby property and overall ambiance, but until recently were not regulated by SL policy13. Thus it was advantageous for Drewcyla and her neighbors to be able to, through their own collective, strategic action, limit the incursion of ad farms in their space. Summary SL political discourse consists of the competing ideologies imported by users from RL worldviews. The virtual-political scene provides a space in which participants can attempt to construct or advocate construction of their own ideological vision within the virtual society. Political discourse is also reshaped by on-the-ground concerns like platform reliability, and the sense that SL’s leadership should be responsive to these concerns. Finally, discussions about policy are conceived by some as collective discussions that negotiate through conflicts to construct new solutions and new knowledge, but others claim that some groups are

13

In February 2008, the SL community standards were updated to forbid “content, particularly advertising, to deliberately and negatively affect another resident’s view so as to sell a parcel for an unreasonable price” (Linden Research, Inc. 2008). Constructing billboards on small parcels remains legal if not used in the manner described here.

Watson 42

systematically excluded from these discussions. The entire sphere of discourse is permeated by the language of empowerment and disenfranchisement. Opportunities for creative expression, economic entrepreneurship, access to education and involvement in socio-political discussions are seen as empowering. Structural inequalities that negatively affect the entrepreneur, and exclusion of some interests from socio-political discussions, are examples of disenfranchisement. I have not given the collective dimension of SL an exhaustive treatment here. This section has focused on macro-level political discourse of how SL society should be managed, but has not addressed other spheres of collective action, including special interest groups, businesses, non-profit organizations, informal networks of friends, and residential neighborhoods, all of which are relevant to understanding SL social life. A superficial treatment of these other spheres of collective action is given in the account above concerning ad farms.

Watson 43

CONCLUSION Second Life is best understood from an analytical perspective when it is considered not just as a computer-based activity undertaken by participants, but as a social world in its own right in which participants are actors, meaning-makers and creators of cultural knowledge. SL is conceived as an uncharted social space that does not supply participants with all the necessary cultural frameworks for understanding social life. Instead, actors rely on their own pre-existing repertoires of cultural tools to construct new understandings of how social life is and should be structured. As a result, some familiar RL social mores are felt more strongly than others inworld. Conflict arises when different users understand the same social situations according to fundamentally different worldviews or cultural frameworks. Such conflict can be resolved by a collective construction of new knowledge as each participant dislocates from his previously held views to create new modes of understanding based on in-world experience. It can also be resolved without a cooperative effort as individual users negotiate new strategies for dealing with the conflicts they habitually encounter. Discourse on virtual worlds in general, and SL in particular, suggests multiple empowering dimensions, but structural factors place limits on the potential of virtual worlds. Some counter-discourses suggest also that certain parties experience disenfranchisement as they are actively excluded from structural opportunities and collective discussions. Three dimensions of SL social life have been addressed here: how virtual selves are (re)constructed through role-playing and identity workshopping, how micro-level recreational social interaction is understood and structured, and how socio-political ideologies and discourses engage in macro-level discussions about governance, empowerment and inequality. Many other avenues of inquiry exist that have not been treated here. Future work could benefit greatly, for

Watson 44

example, by conducting a more in-depth analysis in a number of categories including real-estate markets, the actual and perceived value of land, how SL landowners interact with their neighbors, how SL homeowners decorate their homes, the SL cultural markets in which users sell their virtual creations, and the overall structure of SL economy and the LindeX.

REVISITING CULTURAL TOOLKITS The analysis above relies extensively on Swidler’s notion of cultural toolkits. However, Swidler’s theory is situated in a broader context concerning what factors shape social action. As such, an invocation of “cultural toolkits” implies an invocation of settled versus unsettled lives and the role that ideology plays in shaping action. Swidler says cultural lag is observed in settled cultural periods as cultural strategies and behaviours outlast the ideas that originally conceived them (1986: 276, 281). Cultural traditions have weak control over short-term action but provide the resources from which long-term action is constructed. In unsettled lives, old cultural ends are easily jettisoned and explicit ideology has a key role in shaping new action (1986: 278). Swidler probably did not have SL in mind when she penned these ideas in 1986. SL presents a unique situation because its users are thrown into an unsettled, uncharted space while at the same time remaining firmly situated in settled RL social life. I hypothesize that SL represents neither a settled nor an unsettled life: rather, it is a vehicle through which lives become unsettled through a cyclic process of cultural construction and reconstruction. The cycle starts when users import their cultural toolkits to SL. They use this pre-existing cultural knowledge to form their initial understandings of the social world. Where such understandings are uncontested, they tend to become entrenched social norms, such as the imperative to invite friends to sit down when they join a conversation. If understandings are contested, social life

Watson 45

becomes unsettled and participants may undergo a dislocation from previous cultural knowledge. At the moment of dislocation, explicit ideologies begin to play a larger role in determining how new understandings of virtual social life are constructed. Swidler says that, during a period of unsettled life, structural opportunities for action ultimately determine which of competing ideologies survives to become tradition in the next settled period (1986: 273). As previously mentioned, such structural opportunities are not transparent in SL, and at this point it is difficult to say which particular worldviews seem most likely to survive. Virtual worlds are still young, still strange and novel even to those who participate regularly within them. It seems likely that, in the coming decades, we will as a culture become more intimately familiar and comfortable with virtual worlds, which suggests that virtual social life will eventually “settle” and begin to provide us with some systematic cultural frameworks for understanding virtuality.

Watson 46

GLOSSARY OF TERMS •

Avatar – the representation of the physical self in the virtual world. In SL, the avatar is an animated, three-dimensional graphical image of an individual. A user can design his/her avatar to look like just about anything he/she wants.

•

L$, Linden Dollar – the SL in-world unit of virtual currency. This is the unit that measures the value of virtual property and merchandise that is exchanged in-world. Linden Dollars can also be exchanged on the LindeX for real-world U.S. Dollars. The exchange rate hovers between L$250 and L$300 to $1 USD. “Linden Dollars” is also sometimes abbreviated as “Lindens,” not to be confused with “The Lindens” (see below).

•

Linden Lab, “The Lindens” – the corporate provider/designer/maintainer of Second Life. Phillip Rosedale is the CEO/founder of Linden Lab. The senior Linden staff is sometimes referred to as “The Lindens” because all staffers’ avatars bear the surname “Linden.”

•

MMORPG – Massively Multiplayer Online Role Playing Game: a genre of computer game in which users role-play adventurers in an online virtual world. Generally, these have a combat swords-and-sorcery theme. Classic examples include Ultima Online, EverQuest, The Sims Online and World of Warcraft.

•

MUD – Multi-User Domain: a type of text-based online virtual world popular in the 80s and 90s. The world is presented to the user through text descriptions stored in a database on the service provider’s computer. Users interact with the world by typing commands. For example, to move from one location to another, a user might type “go north.” Communication is enriched by commands that invoke narrative gestures. For example, a user can display a message to others to the effect that said user is laughing by typing “laugh.” A user can also interact with virtual objects through such commands. “Sit on couch” can be

Watson 47

expected to generate a message to the effect that the user takes a seat on the couch, but only if there is a virtual couch present. MUDs are generally divided into adventure MUDs (roleplaying games, usually with a swords-and-sorcery theme) and social MUDs (spaces where socialization and world-building are the focus). •

Real Life (RL) – refers to the offline social world, the range of interactions that occur within the physical space. RL interaction does not involve the use of avatars or self-conscious roleplay. Users tend to use the term “RL” to refer to physicality and to their activities that occur outside of SL—eating, sleeping, going to work, interacting face-to-face with friends, and so on. Certain electronic interactions, such as telephone calls and email, are considered part of RL to the extent that these forms of electronic communication are used in the context of relationships that exist in the RL social world, and generally carry purely verbal content (i.e. no narrative dimension). The use of the word “real” should not be taken to imply that SL is “fake life,” given that real, physical beings sit at the controls of every avatar and interactions can be perceived as real actions mediated through simulation. In other words, a user’s virtual palatial home in SL might be seen as part of the unreal, but if he sees his interactions with friends in SL as glorified phone calls, then those interactions are no less real than a conventional phone call would be. Furthermore, there is no tidy boundary between first life and Second Life. Thus, the term “real life” should not be taken too literally. Nevertheless, most users invoke the term “RL” to refer to offline life as described above.

•

Resident – a participant in Second Life.

•

Rez/Rezzing – the verb “to rez,” probably derived from “resolve,” means to make an object physically manifest in the virtual world. To illustrate: Herschel owns a car, a nice black virtual imitation Mercedes. However, most of the time it exists in a prototype form in his

Watson 48

inventory. It is not an actual virtual car, but the information needed for the program to create a virtual car. When Herschel “rezzes” the car, he actually makes it appear in the world. It becomes visible, other users can see it, and most importantly, Herschel can get behind the wheel and drive around. •

Second Life (SL) – a fully graphical, three-dimensional online virtual world, complete with its own economy and system of currency. SL is hard to characterize (hence part of the reason for undertaking this project in the first place), but it is perhaps best thought of as an MMORPG that focuses on socialization instead of dragon-slaying, or a social MUD with 3D graphics.

•

Virtual world – generally, any online space that is used for group communication, including real-time chat rooms, newsgroups, mailing lists and web forums. However, the types of virtual worlds of interest are those that, like MUDs, modern MMORPGs, and SL, include some other form of narrative tools – e.g. a database of interactive objects, graphics, etc. In some discourses, “virtual world” is also used to refer to offline, single-user virtual realities like video games, but this connotation is not explored here. Generally, “virtual world” is used synonymously with “virtual community” and “cyberspace community.”

Watson 49

BIBLIOGRAPHY Castronova, Edward. 2005. “Synthetic economies and the social question.” Working Paper Series. Available: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=782826 Denton, Nick. 2007. “Virtual world’s supposed economy is ‘a pyramid scheme’.” Valleywag. Retrieved 26 May 2008. http://valleywag.com/tech/second-life/virtual-worlds-supposedeconomy-is-a-pyramid-scheme-230813.php Dibbell, Julian. 1998. “A rape in cyberspace.” In My Tiny Life: Crime and Passion in a Virtual World, 11-30. New York: Henry Holt and Company. Originally published in The Village Voice 38:51. Harrison, Randolf. 2006. “SecondLife: Revolutionary virtual market or Ponzi scheme?” Capitalism 2.0 (Weblog). Retrieved 26 May 2008. http://randolfe.typepad.com/randolfe/2007/01/secondlife_revo.html Jenkins, Peter S. 2004. “The Virtual World as a Company Town - Freedom of Speech in Massively Multiple Online Role Playing Games.” Journal of Internet Law 8(1). Available: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=565181 Kollock, Peter and Mark A. Smith. 1999. “Communities in cyberspace.” In Communities in Cyberspace, ed. Mark A. Smith and Peter Kollock. London: Routledge. Layton, Julia. 2007. “Can I make my living in second life?” HowStuffWorks. Retrieved November 14, 2007. (http://computer.howstuffworks.com/second-life-job.htm). Linden Research, Inc. 2007. “’Inside the Lab’ podcast with Phillip Rosedale.” Official Second Life Blog (Weblog). Retrieved 27 January 2008. http://blog.secondlife.com/2007/12/21/inside-the-lab-podcast-with-philip-rosedale/ Linden Research, Inc. 2008. “Mainland and the ad farm problem.” Official Second Life Blog (Weblog). Retrieved 26 May 2008. http://blog.secondlife.com/2008/02/13/mainlandand-the-ad-farm-problem/ Nissley, Tom. 2007. “On Second Life and In Second Life: William Gibson Q&A.” Omnivoracious’ Amazon Blog (Weblog). Retrieved 29 April 2008. http://www.amazon.com/gp/blog/post/PLNKKANW2CY4P1GE O’Brien, Jodi. 1999. “Writing in the body: Gender (re)production in online interaction.” In Communities in Cyberspace, ed. Mark A. Smith and Peter Kollock. London: Routledge. Reid, Elizabeth. 1996. “Informed consent in the study of on-line communities: A reflection on the effects of computer-mediated social research.” The Information Society 12(2):169174.

Watson 50

Reid, Elizabeth. 1999. “Hierarchy and power: Social control in cyberspace.” In Communities in Cyberspace, ed. Mark A. Smith and Peter Kollock. London: Routledge. Rosedale, Philip. 2007. Interview. Inside the Lab Podcast. Official Second Life Blog (Linden Research, Inc.). Retrieved 27 January 2008. Available: http://blog.secondlife.com/2007/12/21/inside-the-lab-podcast-with-philip-rosedale/ Rymaszewski et. al. 2007. Second Life: The Official Guide. Indianapolis: Wiley Publishing, Inc. Smith, Anna D. 1999. “Problems of conflict management in virtual communities.” In Communities in Cyberspace, ed. Mark A. Smith and Peter Kollock. London: Routledge. Swidler, Ann. 1986. “Culture in action: Symbols and strategies.” American Sociological Review 51(2):273-286. Taylor, T.L. 2006. Play Between Worlds. Boston: MIT Press.

Turkle, Sherry. 1994. “Constructions and reconstructions of self in virtual reality: Playing in the MUDs.” Mind, Culture and Activity 1(3):158-167. Available: http://web.mit.edu/sturkle/www/pdfsforstwebpage/ ST_Construc%20and%20reconstruc%20of%20self.pdf Turkle, Sherry. 1995. Life on the Screen. New York: Simon & Schuster. Webb, Stephen A. 2001. “Avatar culture: Narrative, power and identity in virtual world environments.” Information, Communication and Society 4(4):560-594.

Watson 51