The Story of my (Second) Life: Virtual Worlds and Narrative Identity

Abstract

A small but significant number of residents of Second Life (SL) insist that SL is as real to them as Real Life (RL) and that their SL avatars are as much themselves as their offscreen selves. This paper investigates whether this claim can be literally true in any philosophically interesting way. Using a narrative account of personal identity I argue that there is a way of understanding these identity claims according to which the actions and experiences of the offscreen user and the online avatar are indeed actions and experiences of a single person. In the course of describing how this is so, the paper also uncovers new insights into how a narrative approach to personal identity should be structured and developed.

Stories of doubles and alter-egos are ubiquitous throughout human culture. Such stories signal the fact that people contain unfathomed depths and present the fantasy that those depths might be expressed in an alternate life, running parallel to the life we know. But what if this fantastic possibility became actual? What if someone could live an alternate life and still remain herself? This is the way some describe the opportunities afforded by virtual worlds in which large numbers of individuals interact online. To most subscribers, these venues nothing more than a kind of game or entertainment. For a small but significant number of those engaged in them, however, virtual worlds seem to be something more. These individuals report that their online lives are as real as their offline lives and that what they do online is as much a part of their actual lives as what they do offline.

It is not clear how to interpret these kinds of claims. There is a strong temptation to view them as metaphorical, but at the same time there are insistent suggestions from both participants in virtual worlds and from those who study them that at least sometimes in at least some respects their claims should be taken more literally. What it would mean to do so—and whether we should—depends upon views about the ultimate nature of reality. But it also depends upon views of personal identity. To think clearly about these claims, we need to have an understanding of the kinds of relations that are possible between online avatars and the offline humans who animate them. It is in this issue that I will investigate in what follows. The space of virtual worlds is wide and varied,¹ and different questions about personal identity are raised in different online contexts. In what follows I will concentrate on non-game virtual worlds, with special focus on Second Life (SL). These worlds, in contrast to online game worlds, have relatively few constraints on player identity and activity and no well-defined goals or rules and so (arguably) more closely approximate life offline. In Section 1, I will formulate the particular question about identity in Second Life and similar virtual worlds that I wish to explore in what follows. Section 2 outlines an account of personal identity that can be used to shed light on this question; Section 3 applies this account to SL, and Section 4 uses what we learn from this application to sketch the basic contours of an understanding of identity in SL and beyond.

1 The Question

As of November 2010, there were more than twenty million registered users of Second Life.² Subscribers can register for a free account or, for a fee, a premium account that allows them to own land. After opening a Second Life account a subscriber generates one or more avatars who move through the virtual world interacting with other avatars. The user has control over the screen name, gender, species, and overall appearance of his or her avatar, and for the most part controls the avatar’s motion and speech. In what follows I will employ the term “user” to denote the flesh and blood human living a real life (RL) offline, “avatar” to denote the online character living a life in SL, and “resident,” the term preferred by those active in SL, as a neutral term to apply, at first vaguely, to whatever part or parts of the user-avatar constellation turn out to represent the person. For instance, when I describe a sociologist interviewing an SL “resident” (within SL) I mean to leave it open whether we should think of this as the sociologist talking to the avatar, to the user through the avatar, or to some yet-to-be-specified combination of avatar and user. In some ways, then, the central purpose of this investigation is precisely to determine what actually constitutes the resident.

The world of SL is a complex one, largely constructed by its users. There is an internal economy (the currency is Lindens, represented as L$) and avatars can use this currency for a multitude of purposes, including to acquire or rent property, and to purchase art, clothing, shoes and even skins.³ SL is a well-developed social space in which avatars interact in many of the ways in which people interact in RL, forming friendships, enmities and rivalries. Sexual encounters between avatars are frequent and varied, and there are marriage proposals, engagement parties and weddings among avatars as well. SL also includes avatars who serve as therapists, teachers, and other professionals. ⁴

SL can play very different roles in the lives of different residents but, as mentioned in the introduction, for a significant number it is more than just a game. This phenomenon is not only described by the residents themselves, but has been remarked upon by those who study SL and other virtual worlds. After extensive online interviews in various virtual worlds, ethnographer Annette Markham reports that she “found reason to destabilize a traditional idea that the experience of reality is grounded in the physical, embodied world.” To her surprise, she says, the residents she engaged with the question “What is really real?” told her “this question was of little relevance to them; rather everything that is experienced is real.”⁵ She found some residents, she said, for whom “cyberspace is a place to go to be with others,” and others for whom “online communication is integral to being and is inseparable from the performance of self both online and offline.”⁶ Anthropologist Tom Boellstorff finds similar attitudes throughout interviews he conducted in SL and quotes one of his interviewees insisting that “our virtual relationships are just as real as our rl [real life] ones.”⁷

Markham sums up the overall idea behind these types of assertions nicely when she explains that over the course of her study of virtual worlds she realized that while she had assumed that her interviewees’ use of the terms real, RL (Real Life), and IRL (in real life) was to distinguish “their ‘not real’ or virtual online identities and lives from their real offline identities and lives,”⁸ this was not exactly right. “The participants were not using these terms the same way I was (or as I was wanting them to),” she says, “…for almost all of these users, real does not so much indicate a state of authenticity or genuineness as it indicates both ‘that which is experienced’ and demarcates ‘those experiences that occur offline.’”⁹ In other words, while RL is used as a synonym for “offline,” the word “real” does not have any deep ontological implications in this context. “Even the term virtual,” she continues “doesn’t encapsulate online experience. For these participants every experience is as real as another.”¹⁰

It is important to reiterate that not all residents have this kind of attitude. Nevertheless, it does seem that there is a robust and widespread sense that for some residents in some instances, virtual worlds are not mere fantasies or fictions but are instead a genuine part of their lives, involving experiences and actions as real as any others. This possibility raises important questions about personal identity. If SL and (at least some) other virtual worlds are (for at least some people) more than games and what is done there is not just play, it seems to follow that the avatars of these people must be in some important sense themselves, and the actions and experiences of these avatar their own actions and experiences. The avatar and user must then somehow be the same person, even though it is also clear that in obvious ways they are not, and this requires explanation.

One obvious way of understanding this phenomenon involves making a distinction between merely playing at something and expressing something. In many gaming contexts there is occasion to engage in pretend or make-believe. Playing a single-player video game one might be told “You’re Harry Potter and need to walk around Hogwarts collecting jelly beans” or “You’re Mario and you have to rescue Princess Peach from Browser.” In each of these cases the player is supposed, in one way or another, to act as the assigned character would act in the circumstances depicted. There is, however, no suggestion that there is any deep or serious connection between the real-life player and the character he portrays. He might, of course, call upon inner funds of aggression, valor or avarice in his portrayal, but there is a clear understanding that what he says and does as the character is not deeply expressive of his real-life self. He is merely playing at being the character.

There are other contexts, however, in which we have reason to think that a character someone develops or portrays in a fictional context or game is an expression of something that is authentically part of her. This may be for the straightforward reason that the person plays an only slightly fictionalized version of herself or, in the cases that tend to get more attention, because she uses a fictional context to give voice to a part of her personality, a trait, or set of desires that is for one reason or another suppressed in real life. This phenomenon can be seen outside of gaming contexts when authors create characters that can be read as representing features of their personalities that are not given expression in the author’s own life. Components of one’s sexuality, aggressive impulses, fears, and desires can be acted out in characters that pursue courses of action the author does not.

It is natural to interpret claims that one’s SL avatar is “as real as” or “more real than” one’s RL self along these lines. Unlike the examples of mere play given above where players are assigned characters with well-defined and rather limited traits SL users have a great deal of authorship over their avatars—in some ways more than over their RL selves—and many degrees of freedom in operating them.¹¹ While some may use this control to create characters through which to engage in mere play, others undoubtedly take advantage of the opportunity to explore and express elements of personality that they cannot or do not express in RL. The anonymity of the online experience, together with the relatively low risks of behaviors that would be dangerous in RL is disinhibiting for many subscribers, and there is little doubt that SL is sometimes used for this purpose.

One way of unpacking the idea that someone’s SL identity is as real as her RL identity is thus in terms of the opportunity to experience vicariously an authentic part of her personality through the actions of her avatar. On this interpretation, the sense in which SL is more than a game and avatars more than mere characters is that the attributes and actions of avatars represent authentic traits and desires of the subscriber. In these cases avatars are still seen as fictional characters in a fictional setting. There is no claim that the avatars are in any strong sense really their users, only that the traits of the avatars are genuine traits of their users. Nor is there any claim that the experiences of the avatars are real; the claim is at most that the user can gain genuine psychological insight or relief from engaging in the fiction.

The idea that SL and similar virtual worlds can be useful vehicles for expressing and exploring aspects of oneself in a fictional context is a plausible one, and undoubtedly many subscribers do use these worlds in just this way. The claims we saw earlier, however, suggest that for some the sense in which these worlds are more than games and avatars more than fictional characters is more properly ontological, and it is worth trying to get a handle on how this stronger claim might be understood. The question that remains is thus the question of whether there is any sense in which the actions and experiences of avatars might be those of the user in a more literal sense, and this is the question I will address in what remains. The question as formulated is still somewhat vague, but the discussion so far provides a criterion for a positive answer. Any analysis that claims that an SL avatar is somehow literally the same person as the user must distinguish the relevant avatar-user relation from the kind of relations of expression we have been discussing that can hold between an author or actor and the characters she creates or portrays and which do not invite claims of literal identity.

I believe that such a distinction can be made, at least in the case of some users, and in a way that makes more literal identity claims plausible. To see how, we will need a working account of personal identity.

2 The Narrative View of Personal Identity

There are, of course, many different accounts of personal identity on offer, and no lack of controversy about the merits and demerits of different views. Here, I will outline one particular approach—the narrative approach—which I believe not only to be a highly promising way of thinking about the relevant questions of personal identity but also especially illuminating in exploring issues of personal identity online. According to the narrative approach the identity of a person is narrative in structure, and the individuation of persons is indexed to the unity of single ongoing narratives. There are a great many different versions of the narrative approach and these diverge in deeply important respects. All are complex, and describing any in detail is beyond the scope of this discussion. In this section, I will thus do no more than provide the salient points of my own version of the narrative approach, the Narrative Self-Constitution View (NSCV), which I have developed and defended elsewhere.¹²

NSCV is in a broad tradition of psychological accounts of personal identity that stems most directly from John Locke. Locke famously argues that it is sameness of consciousness rather than sameness of substance that constitutes personal identity. If the psychological life is transferred from the body of a prince to the body of a cobbler, Locke argues, the resulting person will be the prince and not the cobbler. He would be responsible for the prince’s actions and not the cobbler’s; those who were close to the prince could continue their relationships with him but those who had relationships with the cobbler could not, and so on.¹³ The basic Lockean intuition has proved to be deeply (although of course not universally) compelling. Its pull is seen not only in the fact that for many years views based on these insights were by far the most predominant philosophical accounts of personal identity in the literature,¹⁴ but also in the multiple sources in popular culture in which similar cases of consciousness transfer are treated as cases in which the person moves from one body to another.¹⁵ Despite the popular and philosophical appeal many have found in this claim, it has proved remarkably difficult to turn it into a stable and defensible account of personal identity over time. It is not at all obvious how to come up with a respectable account of continuity of consciousness that continues to support the original intuition about its relation to personal identity.

Neo-Lockeans define continuity of consciousness in terms of “psychological continuity.” This is a complex relation, defined somewhat differently by different philosophers. We can think of it very roughly, however, as trying to capture a certain stability in the contents of consciousness over time. The general sentiment is expressed by David Lewis who tells us,

I find that what I mostly want in wanting survival is that my mental life should flow on. My present experiences, thoughts, beliefs, desires, and traits of character should have appropriate future successors. My total present mental state should be but one momentary stage in a continuing succession of mental states. These successive states should be interconnected in two ways. First, by bonds of similarity. Change should be gradual rather than sudden, and (at least in some respects) there should not be too much change overall. Second, by bonds of lawful causal dependence.¹⁶

Neo-Lockeans argue that persons, as fundamentally psychological entities, can remain the same through gradual replacement of psychological “parts” just as any complex object can persist through change if that change is gradual enough and properly caused.¹⁷

For a variety of reasons this kind of account does not seem to capture the desired notion and psychological continuity theories are vulnerable to well-known and serious objections.¹⁸ I have suggested that the identity-defining relation should be defined not in terms of psychological continuity but instead in terms of narrative unity. We constitute ourselves as persons, I argue, by coming to understand our lives as narratives with the form of the story of a person’s life. The notion of personhood at work here is the Lockean notion which views a person as a self-conscious being with the capacity for moral responsibility and prudential self-interest.¹⁹ Living the life of a person is living a life that involves what Locke terms “forensic” activities. Roughly speaking this means employing normative judgments and recognizing ourselves as subject to them. The basic claim of NSCV is that in order to live a life of this kind we need to have a conception of ourselves as beings who live such lives²⁰ We need to think of ourselves as governed by norms in order to be capable of being so-governed and this, the argument goes, entails having an autobiographical narrative.²¹

Although I cannot come near a full exposition of what NSCV involves, it is important to say a bit more about some features of the view. First, although it requires that persons have ongoing autobiographical narratives, it does not demand that they consciously or explicitly articulate their lifestories continuously—or ever. Having a narrative of this sort is instead a way in which persons implicitly organize their experiences and undertake their deliberations. A person experiences what happens in the present in light of what has come before and what is expected or planned for the future, and in this way temporally remote times are brought into present experience. The basic idea is fairly straightforward. Walking up to the door of a house is a different experience if it is walking up to the door of the home where one’s loving family waits, walking up to the door of an empty house after a bitter divorce, walking up to the door of one’s childhood home, or walking up to the door of one’s first house after years of struggle and poverty overcome. The claim here is not just that the event has different significance in these different contexts—although of course it does—but the stronger claim that in each case the person walking up to the door will actually have a different experience, and this whether the person is explicitly thinking about the context or not. The significance that history and future confer on an event is actually incorporated into the individual’s experience of it in a way that has important implications for overall quality of life and for decision making. The background context in which the momentary event is placed is the person’s autobiographical narrative. In this way past and future events can actually be a part of present consciousness.²²

According to NSCV, the limits of a person are determined by the limits of a narrative, and the integrity of a single person consists in the unity of a narrative. The actions and experiences appropriated into a person’s ongoing narrative self-conception are, for that reason, rightly attributable to her. Of course, if this view is to be plausible at all there must be constraints on an identity-constituting narrative. If the view implies that it is possible for someone to weave just any life story and become whatever she describes it is a nonstarter. The constraints flow from the fact that the narrative must be an intelligible story of a person’s life, that allows the narrator to achieve the structure of conscious experience and agential capacities that allow her to interact in the relevant way with other persons. The two main constraints are the “reality constraint”, which says that someone’s narrative must conform to fundamental and largely uncontroversial everyday facts about the nature of the world we live in (e.g., humans do not usually live more than 300 years; they cannot get from Chicago to Paris in less than 4 s; they cannot be in two places at one time), and the “articulation constraint” which says that someone must be able to articulate parts of her narrative locally when appropriate (e.g., to have something to say in response to questions like “How old are you?” “Where do you come from?” “Are you employed?” “Are you married?” “Do you have any children?” “Where were you yesterday?” etc.)

The justification for these constraints is that they ensure that a narrator has applied the concept of personhood to her own life so as to be able to engage in the forensic relations that constitute the form of life that is personhood. It is not required that a narrative be entirely free of inaccuracies, or that a person is never at a loss to explain how she got to where she is or where she is going from there. Instead, it requires that when a person is faced with what almost anyone would see as clear evidence of an anomaly in her narrative she either correct it or explain why it is not really anomalous; and that when she is unable to articulate basic facts about the trajectory of her life even to herself she at least recognize this as problematic. There is, of course, a great deal more that would need to be said to give a complete picture of this view, let alone to defend it. This is not, however, the place to provide these details, and this brief sketch will be enough to allow us to make some headway on the topic at hand—personal identity online.

3 Narratives and Second Life

Our question is whether there is some meaningful metaphysical sense in which the claims that for some residents of SL their avatars are themselves, and the actions of their avatars their actions, might be true. The view outlined in the last section gives us a more specific way into this question: we can look at the narratives of the user and avatar and ask whether they are in any sense a single narrative.

There is one trivial sense in which we might think that they are. The narrative of the user, if it is to meet the reality and articulation constraints, will have to include the fact that a certain number of hours a day are spent at the computer animating an avatar in SL, and the details of what the avatar does will also need to be part of the user’s narrative as things that, through her keystrokes, she caused as actions of the avatar is SL. But this kind of unified narrative is unquestionably only the narrative of the flesh-and-blood user into which the narrative of the avatar is subsumed as a description of a game. A narrative of this sort makes the relation of the user to the avatar roughly like the relation of someone playing Monopoly to the iron or Scottie dog she moves around the board—or at best like the expressive relation of an author or actor to a character described in section one. The actions and experiences of the avatar are incorporated into the user’s narrative as a fiction, and are thus the user’s actions and experiences only in the sense that the actions of the Monopoly piece are those of the player or the actions of the character those of the actor who portrays her. Undoubtedly, many users have just this relation to their avatars, but these are not the users who are likely to claim that they are identical to their avatars or that the actions of their avatars are their actions. If we are looking to understand the stronger sense in which for some residents SL might genuinely be a place that they go, and their avatars genuinely themselves, we will need a stronger form of narrative unity between SL and RL.

To find this type of unity, it will be more productive to start by thinking of the narrative of the avatar. There is, of course, a question about whether an avatar can have a narrative at all. If avatars are no more than complex game pieces or fictions then to say they could operate with an implicit self-narrative is just a mistake. Attributing a self-constituting narrative to an avatar thus begs the question about the identity of user and avatar in important ways. Still, assuming such a narrative is an efficient way to get at the insights that will allow us ultimately to see the sense in which an avatar might really have a narrative. It is at least true that within the world of SL avatars present as individuals with narratives that are, in many respects, like the narratives of RL persons. They have names and histories; they meet other avatars, make friends, fall in love, get married, buy property, go shopping, have hobbies, and so on. For the sake of argument, then, let’s provisionally take the narrative of the avatar within SL at face value as a narrative of an “avatar-person” and consider the relation of the avatar’s narrative to that of the user.

On the surface, this move doesn’t seem to help very much with establishing unity between avatars and users. If we do take the avatar to have a person-constituting narrative it is clear that that narrative will be very different from the narrative of the user. The avatar may be a different sex or species than the user, have different friends, a different profession, and so on. Even the parameters of the reality constraint are going to be different within SL. Avatars can fly, cover vast distances almost instantaneously, change sex without medical intervention, and do many other things that the user cannot do in RL. Even those who insist that SL is a “real place” draw a distinction between it and RL.²³ Given all of these facts it seems safe to say that there is no coherent way to include SL events in one’s narrative as if they were indistinguishable from RL events.

If we look more closely, however, we will see that although the narratives of the avatar and the user are distinct, episodes within them nevertheless interact with one another in ways much more like the ways in which different episodes within a single person-constituting narrative interact with one another than the ways in which the actions and experiences of a fictional character interact with those in the life of the author who creates them or the actor who portrays them. We can see this from two directions: The impact of the avatar narrative on the user narrative and the impact of the user narrative on the avatar narrative.

The elements of SL that get the most press with respect to blurring the lines between RL and SL are those in which what happens in SL impacts a user’s RL narrative in some very direct way. There are many examples of the phenomena I have in mind here. Often they are connected to SL’s vibrant economy, and the easy way in which L$, the currency of SL, can be exchanged for RL currency. Many RL users have earned a great deal of money from the actions of their avatars in SL—selling real estate, designing clothes and shoes and skins, building houses, and so on.²⁴ There is all kind of work that can happen in SL; the avatars of ethnographers can conduct interviews, avatars can get training for jobs that will get carried out in RL, and political campaigns have offices there.²⁵ Musicians whose avatars perform in SL can get RL recording contracts, and their users can get RL gigs, on the basis of their avatar’s SL following. Art created in RL can be displayed and sold in SL (if it is electronically based), and vice versa.²⁶ The blurring of lines between RL and SL can also be seen in the relationships that flourish there. Not infrequently, the users of avatars who are romantically involved in SL will try to meet in RL and RL relationships and marriages have started in this way. One couple whose avatars met in SL, and who then married also in RL, nevertheless felt that they needed to have a wedding in SL as well. Less happily, RL divorces have been initiated on the grounds of SL relationships. ²⁷

All of these phenomena show ways in which the narratives of avatars interact with the narratives of users. When avatars make significant money within SL their earnings are part of the RL narratives of their users, and the RL narrative takes up the implications of this fact; the financial security and success that come from the activities of the avatar contribute to the general sense of financial wellbeing that casts a light on the user’s interpretation of all that happens in RL. It is not uncommon for SL residents to say that their SL romantic commitments are just as morally and emotionally binding as their RL commitments, and in various important ways impact the possibility and nature of RL relationships. ²⁸

More generally, experiences and events in SL can have a profound impact on the overall character of the RL narrative. Successes, failures, friendships and traumas experienced online feed back into self-perception in RL. Residents who are shy in RL, for instance, often report that their outgoing avatars help them to be more assertive in RL. Jeremy Bailenson of Stanford found “that the qualities you acquire online—whether it's confidence or insecurity—can spill over and change your conduct in the real world, often without your awareness … even 90 seconds spent chatting it up with avatars is enough to elicit behavioral changes offline—at least in the short term.”²⁹

It may seem as if these phenomena do not yet reveal a profound difference between the way an avatar’s narrative impacts a user’s on the one hand and the way in which a character’s narrative can impact an actor’s or author’s on the other. After all, many authors and actors make money through characters they create or portray. Miley Cyrus can become rich as Hannah Montana, and can perform live concerts in that character. Stars of television series also often make appearances and interact with fans in character. As for the romantic phenomena, the tendency of onscreen romances to turn into offscreen ones is legendary and, more generally, the outgoingness or confidence of a character portrayed or created may bleed somehow into the actor’s or author’s life, giving them confidence they lacked before in propria persona.

There are subtle yet important differences between the case of SL and the cases of writings and acting, however. In the former case the effect on the user’s narrative comes not only from her pretending to be someone else, but through what the avatar does in SL—activity that is only possible and profitable because SL has a dynamic economy and society of its own in a way that a novel or play does not. And while an actor’s spouse may file for divorce if he begins an offscreen affair with his costar, she is unlikely to file for divorce just because his character has an on-screen affair with her character, nor is an actor likely to actually feel constrained in offscreen relationships by a commitment his character has made even though he might feel so constrained by offscreen commitments.³⁰

The intimate points of interaction between user and avatar narratives are perhaps easier to see, however, when we look at the influence running in the other direction—from the narrative of the user to the narrative of the avatar. It has been observed that while space can be virtualized time cannot, at least not in the same way. This means that the activities and interactions of avatars in SL will necessarily be interrupted by the activities of their users. “AFK” or away from the keyboard is a standard feature of SL, and is used to explain why an avatar is unavailable or behaving strangely (i.e., if the avatar is left dancing at a club automatically but ignores greetings from friends the explanation may be “sorry, I was afk for a few minutes—there was a knock on the door”). An avatar might also explain slow responses by saying “I’m typing with one hand—broke my arm” even though the avatar is perfectly unharmed within SL.

A particularly salient instance of this phenomenon is provided by Tom Boellstorff who tells the story of an incident where his avatar sat down to a tea party with Trishie, a resident living as a young girl in SL. After telling him that “mommy left us lunch” Trishie excused herself—“brb [be right back], gotta check the meatloaf.” Boellstorff continues, reflecting that they “were talking about virtual food in her playhouse when she suddenly said she had to go afk and would ‘be right back’ (brb) because she had to check an actual-world meatloaf in an actual-world oven. Trishie had told me earlier that in the actual world she was a mother with two children; she was preparing food for her actual-world children while her virtual-world parents had left food for her in Second Life.”³¹ This is very different from the relation between a character and author or actor. Only in the most postmodern of works will a character explain why she does what she does by reference to the life of the author or actor.³² We might explain a difference in a character’s demeanor in terms of the fact that there is a new writer or actor creating the character. The character herself will do so within the fictional narrative, however, only in works explicitly aimed at breaking down the boundaries between fiction and reality.

There is a great deal more that needs to be said about the details of interaction between these two narratives, especially involving cases where the relation between avatar and user is not one-one. This overview will be sufficient, however, to draw some important preliminary conclusions.

4 What it all Means

Our question is whether there is any meaningful metaphysical sense in which an SL avatar can truly be the same person as her user. My proposal is that we use NSCV to address this question, but instead of asking whether the narrative of the avatar is the same as the narrative of the user, we ask instead whether these distinguishable but interrelated narratives are both part of a single, broader person-narrative. The answer to this question, I suggest, is that they sometimes are,³³ and that the sense that in some instances SL is more than a mere game and the avatar more than a fictional character is best expressed through the fact that sometimes the RL narrative of the user and the SL narrative of the avatar are, as it were, subplots in the more comprehensive narrative of the resident, a person who lives sometimes in RL and sometimes in SL. Both sets of adventures are part of the same life because, although distinguishable sub-narratives, they impact each other along the most fundamental dimensions of narrative interaction.

This proposal as it stands is quite vague, and there is only so much precision I can give it in the space remaining. It will be possible to get the basic idea, however, through an analogy. For the moment put aside online contexts and think instead of a responsible husband and father who travels every two months to the same Las Vegas hotel to meet his college buddies and let loose, expressing elements of his personality that do not fit easily into his everyday life anymore. For him, Vegas is a special context, distinguishable from his life at home. He sees it as a bubble in which he can temporarily suspend the demands of his everyday life and so suspend his everyday life-narrative insofar as it no longer governs or directs his experiences and actions in the same way it does back home. Different rules apply here; in some sense the reality and articulation constraints within Vegas are different from those in everyday life, and so the “Vegas-with-the-guys” narrative can be distinguished from the “home-life” narrative.

Vegas is part of the world, however, and advertising campaigns notwithstanding it is by no means possible to guarantee that what happens in Vegas stays in Vegas. The Vegas experience and narrative will not be completely cut off from the everyday narrative, but will necessarily interact with it in a variety of ways. For one thing although some rules of daily life will be suspended in Vegas, they probably will not all be, and the broader narrative will set some constraints on the Vegas narrative. The traveler will, for instance, know when he needs to return to work. If he receives a phone call telling him that there is an emergency at home he will cut his trip short. His home narrative will also set bright lines that cannot be crossed. Perhaps he can flirt in Vegas in a way he does not at home, but he should not have an affair; he can spend money in a more irresponsible way than he usually does, but not his daughter’s college fund or the mortgage money; he can drink more than he usually would, but not enough to end up in jail or in the hospital. The space within which he is able to explore his alternate self is thus bounded by elements of his everyday narrative.

The Vegas and everyday narratives also interact in that what happens in Vegas will undoubtedly affect the everyday narrative, and always has the potential to affect it dramatically. The carefree time spent in Vegas may make someone more relaxed or happy in his return to the everyday; or more dissatisfied. Real money can be won or lost in Vegas and that can affect material wellbeing. If it is enough money, it can change the everyday narrative radically. Crossing the line and having an affair in Vegas may break up the traveler’s real life marriage if the infidelity is discovered, or just contribute to a guilty everyday narrative that reinterprets and colors all interactions with his family. Alternatively, on a trip to Vegas he may meet the love of his life and decide to break up the marriage himself, changing the direction of the everyday narrative entirely.

The Vegas narrative is thus not entirely insulated from the everyday narrative—such insulation is not possible. The actions and experiences in Vegas interact with the actions and experiences of everyday life, but they do not do so quite as completely as do other actions and experiences of everyday life. The significance of the Vegas events is indexed to the fact that they occurred in Vegas, but not in a way that makes them fictional. We can thus see an overall narrative that contains both everyday and Vegas events, but integrates them and their overall significance differently. It does not make them cohere with one another in the strong way that events internal to either must, but the ways in which they interact set broad narrative constraints with which both, taken together, must cohere. This is the narrative of a man who lives sometimes at home and sometimes in Vegas.

This analysis reveals something important that NSCV must take onboard, namely that self-narration is a complicated affair. It is not just a matter of thinking of various events (implicitly) as “mine” or “not mine.” Context also matters. The narrative of a life can be one with multiple subplots, digressions, and deviations from the main narrative stream; it need not be linear in a simple-minded way. The coherence an identity-constituting narrative calls for thus need not and should not be a complete consistency or a thematic unity of all desires, traits, and activities, but rather a recognition of the myriad ways in which pursuing each of our multiplicitous facets impacts the pursuit of the others. My claim is that the same kind of narrative structure that applies to Vegas and everyday life-narratives can apply to SL and RL narratives as well, at least in those cases where we are tempted to say that SL is more than a game. SL is, like Vegas, a different context in which different rules apply and a sub-narrative unfolds, but it is a sub-narrative that interacts with the RL narrative in much the same ways that the Vegas narrative interacts with the everyday one.³⁴

This is, of course, only a sketch of how a narrative account of identity in SL and other virtual worlds would go. The Vegas analogy is only an analogy. There are many differences between vacations and online existence, not the least of which is shared embodiment in the former case but not the latter. These would need to be unpacked and their significance investigated in working out a full version of this account of online identity. Much more detail is also required about just what kind of multiplicity a single narrative can tolerate, and the exact nature of the fundamental principles of interaction that weave different subplots into a single narrative must be more precisely articulated. Perhaps most important, even if all of these sub-narratives are part of a single life, they need not all have the same status within this life. Indeed, they probably cannot. This means that they will have to play different roles in constituting identities and the different kinds of roles they might play need to be better understood. There is, therefore, a great deal of work that needs to be done to develop these ideas, and the picture given here is more of a research program than a final account.

NSCV does, however, give us the basic contours of a view of identity that takes seriously the claims that virtual worlds can be a genuine part of a person’s life, and her avatar a genuine part of herself, while still appreciating the very obvious and important differences between SL and RL. At the same time, application of this view to the context of virtual worlds like SL provides a great many insights into what a satisfying narrative view of identity must look like. Not only virtual worlds, but the many types of compartmentalization that occur in RL show that we cannot operate with a simple or straightforward view of narrative, but must develop one that allows for a basic unified narrative that sits as an umbrella over a variety of different sub-narratives which might include quite different desires, traits, and activities from one another. As usual, then, philosophical analysis of these matters leaves us with more questions than answers. But they are more focused questions, and they are questions whose pursuit promises fruitful new approaches to our understanding of the theory of personal identity, of ourselves, and of our doubles and alter-egos.³⁵