Early and persistent scholarly concerns with online identity emphasized the ways that computer–mediated communications have allowed new, inventive, and creative presentations of self, and the lack of connection between online identity and the facts of off–line life. After the ascendency and following ubiquity of Facebook, we find our online lives transformed. We have not only seen online identity reconnected to off–line life, but we have seen, through the particular structures of social networking sites, our online lives subjected to newfound pressures to unify self–presentations from various constitutive communities; pressures different from and in some ways greater than those of off–line life. After describing identity in computer–mediated communications prior to Facebook, and investigating the kinds of changed conditions brought about in social networking sites, I put forth a dramauthentic model of post–Facebook online identity. This model is comprised of three methods of exposure through multiply anchored self–presentation (mixed, agonistic, and lowest–common–denominator) and four strategies of interaction (spectacular, untidy, distributed, and minimized), each of which are employed non–exclusively and at different moments by most social networking site users.
2. Friendship online and mutual self–interpretation
3. Self–presentation before the Facebook era
4. Self–presentation in the Facebook era
5. Identity in the Facebook era: Strategies of self–exposure
6. Identity in the Facebook era: Strategies of identity–construction
7. Concluding comments and directions for future research
In the earlier days of Internet studies, the question of self and online identity online centered on how digital media allowed for the construction of a self identity untethered from the unchosen circumstances of off–line life. The earliest and most persistent questions of selfhood in digital media have been motivated by and concerned with the optionality of carrying over elements of the “RL” self. This problem, and the set of questions having to do with identity, community, and personal relationships, can be seen clearly even in the very titles of what have been paradigmatic studies on this era of online life: e.g. Sherry Turkle’s (1984) The second self.
In the era of social networking sites (SNS), digital media present us with a very different and in some ways opposite set of concerns. While it is certainly the case that a great deal of pre–SNS Internet use took place using identities intended to reflect the off–line self rather than as a form of creative identity–play and reimagining of the self  — and certainly the case that a good deal of identity–play still takes place today  — the valences of both code and norms have become reversed.
The chief concern used to be whether the online self was too untethered, and how the online self was or ought to be related to the “real” off–line self. Today, instead, the problem is that the online self is too tethered, and tethered in ways that reduce user autonomy and identity performance. This shift is well–recognized in the prominent contemporary concerns with privacy, surveillance, and data mining, but is not fully reflected in current scholarship on identity performance online. In this article, I intend to bring together work on this new, strongly tethered milieu of online identity performance in order to present a unified model of identity performance on SNS. My aim is to present a model which is as general as possible while maintaining concrete and detailed grounding in the specific processes and interfaces of SNS self–performance.
To highlight the broad changes in online identity performance from a space generally characterized by disconnection and play to a space generally characterized by non–optional constraints, we will focus on emblematic kinds of online identity performance, leaving aside the admittedly many exceptions. This results in what seems to me an unavoidable openness to the charge of overgeneralization, and I can but request charity in understanding that I mean to make general but certainly not universal claims regarding the large–scale shift in the context and form of online identity performance which I intend to identify and theorize.
Before beginning the process of articulating the precise nature and origin of this general shift, we can attempt to articulate what this shift consists of in its largest–scale view. This attempt should be prefaced by the proviso that this preliminary articulation will be, at this point, merely suggestive. It does, however, represent the overall view that I hope to establish in a more grounded, responsible, and defensible manner by the close of this paper.
As Lee Rainie and Barry Wellman (2012) have expressed this shift,
At first, the Internet and Mobile Revolutions aided this segmentation [of context– and community–relative self–performances]: Email, texting, and mobile calling are usually one–to–one media. But the rise of social media has brought people back into one network — happily or not. The most popular social media such as Facebook have offered limited ability — so far — to deal with the subtleties of how people really function in different segments of their networks. Rather, the sites tend to treat each person’s network as a monolithic entity that functions in a let–it–all–hang–out milieu. 
As a broad picture, this basic view of how SNS have brought about a return to consolidated identity is right, but does not capture the way that a consolidated network actually produces greater limits on contextual self–performance than do off–line social networks and groups. It is also important to emphasize, as Rainie and Wellman well–recognize in the wider discussions of their text, that while the SNS architecture is largely based on a “let–it–all–hang–out milieu,” user strategies of managing identity in that milieu are complicated, and certainly not limited to merely giving in to the “let–it–all–hang–out” ethos of authenticity built into the SNS architecture. Still, their characterization that “Facebook is mostly about groups rather than networks”  accurately portrays the performance of self on SNS as a return to a former form of self–performance rather than as a distinctive new architecture of networked yet fully — and ever more fully — tethered self–performance.
How then have the new consolidated networks of SNS not merely returned us to tethered identities, but altered and changed these structures of identity performance, whether or not we agree with or even acquiesce to the idea that we should be the same “self” in each context? To put it in a self–consciously circuitous way, the problematic relationship between the online and off–line selves has been deproblematicized in a problematic manner. The strikingly voluntary construction of self–narrative emblematic of earlier online identity has been re–attached to less–voluntary and involuntary aspects of off–line life, collapsing the wild and limitless freedom of identity–construction partially back into the familiar and everyday dramaturgical self–construction of multiple self–presentations to various constituent communities — but now with an architecturally–imposed need to reconcile those selves with one another in a way not required by the architectures of off–line life, due to the promiscuous intermixing of communities in the information feeds of our Facebooked sociality.
Obviously, this characterization is, at this point, abstract, and dependent on terms and phrases which are in need of definition and explanation. In the following, we will engage in a sociologically informed phenomenology of the structuring of identity performance on SNS — structuring by both norms and code  — that will allow us, in concrete terms, to explain and justify this characterization by constructing a model of self–performance showing these changes in architectures, the changing context of identity performance that they create, and the strategies which users adopt in reconciling interpersonal autonomy and communal connection in this altered online environment.
2. Friendship online and mutual self–interpretation
To get a concrete start on the question of self–identity online in the current technological moment, we will begin by looking at a debate from the previous moment. By seeing in which ways this debate has become defunct, we can throw the contours of the current moment into relief and delineate its outline. To this purpose, we will turn to Adam Briggle’s defense of friendship online in his 2008 article “Real friends: How the Internet can foster friendship.”
Briggle is responding to Cocking and Matthews (2000), whose position he summarizes in part as that “exclusively online close friendships” are “simply psychologically impossible” (Briggle, 2008) because online interactions
[diminish or eliminate] acts of non–voluntary self–disclosure, which are necessary for the mutual shaping characteristic of close friendships. The Internet is a context dominated by voluntary self–disclosure, which enables one to choose and construct a highly controlled self–identity. It creates a distorting filter on aspects of ourselves that are normally disclosed to friends in offline contexts, thereby short–circuiting the mutual interpretation and shaping of identity that contribute to the depth and character of close friendships. 
Briggle agrees with Cocking and Matthews’ view that a fundamental aspect of the process of friendship is reciprocal interpretation, allowing for personal growth and self–discovery, in line with the description of a friend as a mirror in the Magna Moralia (Aristotle, 1984a). His reply to their argument is that computer mediated communication (CMC) presents no structural barrier to this process of mutual and reciprocal interpretation, and that therefore there is no reason to think that a purely mediated (i.e., online first, online only) relationship cannot be a close friendship. Using a compelling example, that of a Union soldier on the front of the U.S. Civil War who is a pen pal with a school teacher in Boston, he makes a strong case that the distance created by asynchronous communication untethered from face–to–face reality actually opens new possibilities for the kind of close friendship which leads to authentic self–understanding and mutual disclosure. Briggle argues that when we are tied to others, whether they are our fellow soldiers, our co–workers, or our poetry club, we must present a certain version of ourselves limited by the requirements of maintaining that connection, maintaining that community, and serving the goals and shared values which may be central to that connection or community. Freed from any involuntary aspects of association and any limits on self–presentation, we are able to explore parts of the self that we would otherwise need to hide from view — and, further, in the written and asynchronous communication typical of most online interaction, we have great opportunities for introspection and careful, deep self–disclosure.
In reading Briggle’s argument today, we see three different images of the self in our involvement with different moments of in the development of digital media.
In the argument from Cocking and Matthews, published 14 years before the time of this writing, we see a view which today seems reactionary and essentialist, responding to new technologies of self–performance as suspect and inauthentic and requiring that the self be defined in terms of off–line interactions. Today, we may wish to push further than Briggle’s criticism and ask why the off–line self rather than the online self is the exclusive proper subject of interpretation and relationship–building.
In the response from Briggle, published six years before the time of this writing, we see a view which today seems quite intuitive and clear: fictionalization and narrative–constructions are technologies of the self , and even though they can be misused, they are powerful techniques which can certainly support building meaningful personal relationships. We might note that Briggle’s view here is well–supported by Turkle’s (1995) discussion in Life on the screen of “taking things at interface value,” in which we treat the self established in writing as a legitimate Other with which to build a relationship, even if it may not be representative of the entirety of the person performing that self. We may even think of the eighteenth–century practice described by Habermas, “that social game in which after dinner everyone withdrew to write letters to one another” , which places special value in the depth created by the narrativization of less–mediated interaction.
And yet, even though we recognize ourselves in Briggle’s view, we know that it is a photograph, not a mirror: we know how to think of ourselves through these features, and they are familiar and comfortable to us, but our features have since changed in ways we are still having trouble seeing. We recognize ourselves in his view, and yet know that we have changed, and this gives us a good opportunity to see who we are today, and how we have changed. Briggle’s snapshot is, I think, ideal for this because it captured our air in Barthes’ (1981) sense — his view captured well who we were; I agree fully with his claims, it’s only that we have changed since then. Five years’ time is an age in technological matters at today’s pace, and we have entered a new era: the era of Facebook.
3. Self–presentation before the Facebook era
While Briggle’s argument is concerned with friendship, since the theory of friendship employed is focused on reciprocal interpretation, it has clear relevance to questions of personal identity. As indicated by his references to Goffman’s (1959) Presentation of self in everyday life, a dramaturgical theory of the self is a basic perspective and touchstone, and can serve as a cue for understanding the change in personal identity depicted in the digitally–mediated and digitally–unmediated selves. To get a view of the self in line with Briggle’s argument, we will first consider interactions apart from social networking sites (SNS), allowing us to then look separately at the changes brought by SNS.
Within digitally–unmediated self–presentation, we might imagine digitally–unmediated interactions taking place (1) within more accidental associations, such as family or hometown peers; (2) within communities of common values, such as faith– or politically–based groups; (3) within communities of common activity, such as work or volunteer groups; and, (4) within more fully voluntary associations, including friendships and social groups. In each of these interactions, we present the self differently, and find it inappropriate to designate any particular presentation of self as exclusively authentic. Each separate contextual self–presentation within these communities is limited by the facts of our existence; our particular attributes, characteristics, and capabilities. There is also certainly a limit to how divergent these selves can be from one another without the threat of being called to account for the disparity, should someone from one community encounter us in the context of a different community.
This threat is highlighted within constitutive communities, which Sandel (1982) defines as people’s associations that are “not a relationship they choose (as in voluntary association) but an attachment they discover, not merely an attribute but a constituent of their identity” . While here we will avoid contentious claims about any “true” underlying identity, this definition is useful in pointing out why those communities that users themselves consider to be constitutive are a greater source of identity threat as compared to communities which are in Sandel’s terms instrumental or sentimental — based in utility and self interest, like many professional associations for many members, or based in emotional connection independent of core values, like the connection people often feel to others from one’s hometown, even when they find themselves wanting to distance themselves from the values, beliefs, and way of life endemic to their birthplace.
Within digitally–mediated self–presentation — again, excluding social networking sites — these aspects of the self can be easily separated. Numerous selves may be presented to numerous communities — from the political discussion board, to the parenting Web forum, to comments from a YouTube account , to the pseudonymous blog — with few limits to the presentation of self, and little threat that the various selves should be connected unless we ourselves choose to connect them. Without these limits and in the absence of a threat that we may be called to account for disparities between these selves, a remarkable capacity for free–play in self–presentation emerges. When we look at this in commonplace examples, such as the ability to engage in political debates which would be viewed as questionable by hometown friends, or to be active in a community of faith without disclosing those associations to an unsympathetic employer, we see a flourishing of multiple selves which strikes us as a boon for self–realization. In these cases, we are likely to see the compartmentalizability of aspects of the self as supportive of autonomous identity construction: we are freer to be who we are, in some cases, when we can express who we are through fragments rather than facets .
4. Self–presentation in the Facebook era
With the ascendency and following ubiquity of Facebook, the circumstances and architectures of digitally–mediated self–representation have changed radically. There are numerous fairly obvious changes here that deserve enumeration, but the extent of the transformation also requires some more in–depth discussion. It is also worth saying from the outset that I will be emphasizing Facebook in particular, since it is Facebook that has really brought about these changes, at least within the U.S., and in much of the rest of the world as well. I do hope, though, that the relevance of much of the following to other SNSs should be clear, and I intend to avoid any discussion of issues that are truly particular to Facebook, such as its troubling policies centered around harvesting user information to generate saleable market analysis for businesses.
We’ll begin with the more obvious changes.
4.1. Real names
In digitally–mediated self–presentation prior to Facebook, pseudonymous communication was common in many, perhaps most contexts, and use of one’s real and full name was a matter of choice. On Facebook, users almost entirely present themselves under real and full names, due to Facebook user policies and to the dominant social norm on the site.
4.2. You are your own avatar
The process of building networks through individually request–initiated connections requires mutual recognition prior to establishing direct communication or interaction. While some context to aid recognition may be supplied through “friends in common” and through primary listed network (e.g., employer, school, or location), users have a strong reason to supplement name– with face–recognition through use of a recognizable image, and, while deviations are frequent and very accepted — as for example, mothers’ relatively frequent use of their child’s face as their profile picture — a norm of using a photograph featuring the user’s face is well–established. By contrast, in other online spaces and in online life prior to SNS, the choice to present oneself as oneself, and to use either real name or recognizable image or both as one’s avatar, was dictated far more by personal preference and by one’s own aims in engaging with a given online community, space, or forum.
4.3. Unitary profile
Only one profile can be created per user–account, a significant departure from our habit of presenting ourselves differently in different contexts. Differential access to that self–presentation may be granted by revealing some profile information to only some groups or specified individuals, but it is not possible to have two or more entries in a given category (e.g., “Religious Views”) provided to different groups or individuals. Maintaining these divisions, however, is cumbersome, as is maintaining multiple user–accounts, another work–around.
It is certainly true, as several scholars have noted in different ways (Donath and boyd, 2004; Papacharissi, 2002a, 2002b; Papacharissi, 2010) that “information communication technologies provide individuals with additional tools for the production of more detailed and controlled performance of the self” , but even though, in Goffman’s terms, the face that is presented, the performances given off, and the setting, props, and front of the self are largely within the realm of user control — with the exception of opt–out tagging and identity cues given by implication from friends’ posts and comments on a user’s wall — the unitary profile places this control within a context where the only options available to choose from represent compromises, and sometimes difficult or costly compromises. The unitary profile, when combined with the unitary wall, the promiscuous intermixing of audiences (4.6 below) and the difficulty of managing segregated communicative contexts (4.7 and 4.8 below), results in a choice of props and settings for identity performance which are largely voluntarily chosen, from what we fill in for the “about” section, how we fill it in, and what we leave blank; to what kind of image we project through the profile picture; to what setting and associations we give off through our cover photo; to what set of concerns and values we project through the topics of our posts and contents of our comments. And yet this choice of props and settings presents unprecedented threats to variously performing divergent identities by the fact that these props and settings must serve as the frontstage for self–performance before multiple constitutive communities, as well as, often, the setting for the backstage of those performances as well, as will be further described below (4.8).
4.4. Opt–out tagging
The ability of users’ friends to tag them in photos, videos, and notes — these items then appearing on the user’s wall until and unless they are specifically deleted — creates a constant threat that communities for whom a tailored self–presentation has been created will be presented with alternate images of the user, unintentionally, by differing communities. These communications, furthermore, may be entirely in line with the self–presentation given to the particular communities from which they emerge, resulting in a situation where a friend may tag a user in an activity entirely normal within their shared community, but which is transgressive to another community of which that user is also a part — and this may be done without the knowledge of the user, and without the tagging friend having any intention to communicate with the other community, any awareness that that community exists, or any idea that there is a disparity in the self–presentations of the user to these communities. What appears to one friend as a quotidian and unremarkable documentation of shared activity — holding hands at the beach, or sitting together at prayer — may appear to another friend of the tagged user as a distressing revelation of a questionable secret life.
4.5. Obligation to sign up and sign in
Social sacrifices required by non–participation have become significant for many of many generations and for many different reasons. In the case of students, Facebook has become a substantial center of social life — indeed, it seems a not uncommon practice among high school users to friend a large portion of their schoolmates, including those who they may view as not just frenemies, but in some cases actual enemies, resulting in an environment where self–presentation may be particularly fraught with anxiety and risk. In college students and emerging adults, Facebook may be an important tool in social organization, and significant social activity may take place on the site itself, leaving the infrequent user or non–user increasingly disconnected from her own close friends. In older adults, Facebook may present an irreplaceable value in providing easy, frequent, and multimedia contact with distant friends, distant children, and distant grandchildren. I make no claim that these are the primary generational social obligations which strongly encourage participation, and they are certainly not the only significant such obligations, but highlight these examples only as paradigmatic social forces driving users to sign up and sign in.
4.6. Promiscuous intermixing of audiences
While users may choose to target their own communications and limit access to content linked, written, or uploaded, this must be done on a case–by–case basis, and is inhibited by code (Lessig, 2006), by SNS user norms, and by norms of social interaction. SNS user norms are also inhibiting insofar as it seems to be either common knowledge, or at least a common assumption, that most SNS users do not target postings most of the time — certainly it can be said at least that such targeting is not transparent to either those included or those excluded, and any lacunae in postings received caused by such targeting, even if discovered, may be at least as easily explained by Facebook’s inscrutable algorithm determining which posts are displayed to which users within the poster’s network.
Our long–established habits of social interaction also do not include regular and purposeful targeting of communications independent of contexts which aid in identifying and thinking through our likely audience. Hull, et al. (2011) helpfully quote Strater and Lipford’s (2008) claim that
while managing identity and privacy is a continuously negotiated process in face–to–face interaction, online interactions make the case–by–case decision–making process difficult. Users rarely interact with each other synchronously. Instead, decisions of privacy, what to disclose and how, must be made a priori and explicitly. 
In addition to this shift from “ad hoc decisions” to “a set of ex ante rules for determining how information should flow in and between contexts” , Hull, et al. also point out that “the abstraction involved in asynchronous, online social networking encourages a gap between a user’s perceived audience and actual audience;” that “although Facebook theoretically has a highly granular set of privacy settings, users do not appear to be taking advantage of them;” and that the “binarization of social relations into ‘friend’ and ‘not friend’ ... flatten[s] out all the nuances of face–to–face interactions”  — all of which, when combined with the unitary profile and wall, tends to result in a context of simultaneous performance before multiple audiences, significantly different from both online and off–line non–SNS contexts of identity performance.
We are unaccustomed to deciding in each interaction who, precisely, we are speaking to largely because our physical and social contexts of communication are typically community–relative — as we’ll see further in section 4.7 below — whether in the location and social purpose of face–to–face encounters or in the more specific online locations of blogs, forums, e–mail threads, and so forth. Managing these different audiences through use of code — e.g., through selecting from different friend networks or circles — requires a constant and explicit thinking–through of who we choose to be relative to those groups. And to think through managing these self–performances on sensitive matters requires not only the constant thought of those groups, but also recalling who precisely is contained within those groups and imagining communicating with each of them individually. The process quickly becomes burdensome, difficult, and constantly subject to missteps, in addition to requiring a certain unfriendliness and circumspect mistrust toward one’s network. It should not be surprising if few users indeed choose to regularly target communications. For these reasons, many of which will be discussed at further length below, for the most part we can treat SNS interaction as a broadcast to a variety of different audiences who are promiscuously intermixed.
4.7. Loss of unintentional architectural supports for identity management
Taken together, these aspects of Facebook involve ever more people in circumstances of online self–presentation which are not merely a reversal of the previous distinction between computer–mediated and computer–unmediated communication, but which actually swing the pendulum farther than its point of origin, in some aspects at least. We are not merely no longer typically communicating on a variety of different sites, on which we create a variety of self–presentations, through which we are represented pseudonymously, and which are disconnected from and difficult to connect to one another — that is, the speciation in this vast and fecund ecosystem of personal identity has not merely been halted — but rather, in Facebook, the forces previously existing offline which inhibited self–presentations of a greater divergence from material circumstances or from one another have closed in upon us online even more than they ever had off–line. This has happened not only in the more clear ways delineated above, but also, as we’ll see in the following, through the basic but subtle way in which digitization can change the meaning and importance of the information it represents by both replicating some aspects of pre–digital architectures which are viewed as functional, and also failing to replicate or consider the functionality of other pre–digital architectures.
Facebook is designed to re–create our off–line social networks in its social graph — but while Facebook has been quite successful in motivating users to digitally reconstruct those connections, Facebook has not digitally reconstructed in code the off–line unintentional architectural barriers (Lessig, 2006) which we use to navigate between our various communities and maintain multiple self–presentations, tethered to those contexts and not identical to one another. This includes the creation of what Papacharissi (2010) has well–described as ‘architecture[s] of publicly private and privately public spaces’ from ‘convergence of technologies ... convergence of spaces ... [and] convergence of practices“  — all of which is intensified and made all the more effective by the other identity–tethering factors outlined in the above and following sections. As a result, the integration of various communities in the SNS does not re–create the dynamics of off–line self–presentation, but instead introduces a demand for a set of self–presentations that can be reconciled with one another that is actually more radical than in digitally–unmediated communication, wherein the call to account for disparities between self–presentations is no longer a mere possibility and a threat, but is instead a task at hand in most interactions.
To demonstrate this kind of emergent outcome, and the need to consider carefully which elements of pre–digital architecture we ought to re-create in code, consider the following example used in my course in Computer ethics. In the United States, those convicted of certain sex crimes must register their address with law enforcement, and they may be subject to certain requirements, for example, to live a certain distance from schools. Further, the registry of sex offenders is a matter of public record, and is searchable by name or by postal code. I ask my students to consider the difference between (a.) allowing the information to be public, but requiring citizens to personally contact local government in order to access records; (b.) allowing anyone to search the database anonymously from their own home, including international searches by foreign nationals; and, (c.) allowing third–party Web sites to provide a portal to the database which displays results to anonymous geographical searches, and locates registered sex offender’s current addresses using Google Maps . Regardless of my students’ beliefs about what kind of information should be accessible and to whom, they find it abundantly clear that the mere digitizing of information produces a substantial qualitative change in the meaning and effect of public availability of that information. I ask them to imagine what it would be like to go to a town or county building, look someone in the eye, and request the information. Perhaps no question would be asked regarding their reason for accessing the information, perhaps the question would not even be implied in a look given by the government employee, but they find even being subject to the gaze of even a quite uninterested other to be a significant barrier to any casual access of these data. The facts of time and space are also a source of ontological friction, in Floridi’s (2005) terminology — even a few minutes’ drive to town hall would be sufficient to ensure that those who access the information are very likely to have some sort of meaningful motivation to do so. Off–line activities are limited by their localization in time and space, and involve the need to communicate synchronously and in proximity with others, and these basic facts shape our choices, actions, and interactions significantly and in ways we often fail to take note of, and for these very reasons — the significance and non–obviousness of the influence of these facts — we are likely to find that transferring data of personal interest and concern into digital formats may transform presentations of self and other in fundamental and unexpected ways.
In digitally–unmediated life, the mere fact of activities and interactions being located in space and time, and, except in unusual circumstances, the mutual visibility and recognizability of persons in synchronous spatial proximity, form architectural limitations to access of these potentially sensitive personal data. These barriers, furthermore, do not for the most part exist due to a purposeful attempt to contextually filter information access, but arise naturally through the exigencies of physical reality. You may gain, for example, knowledge of the private affairs of a male acquaintance by seeing him in the waiting room of the OB/GYN, but this is not because, like an abortion clinic, the building was designed to allow entrants to protect their privacy — it is instead that there are few reasons for men to be present in the building, and the building, by its nature as a building, limits knowledge of who is there by virtue of its being in a particular location, other than that of e.g., the workplace or gym, and by virtue of its having walls and doors.
By contrast, on Facebook, there exist no barriers of this sort; no architectural limits that contextualize data access in a way that supports maintaining separation between self–performances intended for a specific or limited audience. Digitalization, as Floridi (2005) has argued, has not enhanced or augmented capacities to gain, retain and transmit information so much as it has re–ontologized the infosphere itself. We have options in code to recreate these boundaries by sharing with specific groups (or circles in Google+), but, at the time of writing at least, these options in code are cumbersome — and, even if they were not cumbersome, they would still represent specific choices that must be made at the time of sharing information in each and every instance of sharing, whereas in off–line life, for example we do not need to choose to wait to see our doctor only in her waiting room and not simultaneously also in our office, our college roommate’s apartment, and our mother–in–law’s living room.
4.8. Regional ambivalence
As Papacharissi (2010) notes,
In environments that are both privately public and publicly private, the sequential arrangement of backstage and front stage is upset. The backstage no longer signals privacy and the front stage does not guarantee publicity. SNSs potentially collapse front and backstage into a single space, by allowing privately intended information to be broadcast to multiple public audiences, and delivering publicly produced information to private and intimately known audiences. Moreover, the individual must assess not one situation, but potentially an infinite number, in which the same self–performance must maintain authenticity, coherence, and relevance. 
While much of this situation has already been addressed in the above, as for example in the promiscuous intermixing of constitutive communities, the co–location of back and front regions, noted preliminarily above in consideration of the unitary profile (4.3), calls for detailed and separate articulation.
The backstage region of self–performance — defined by Goffman (1959) as “a place, relative to a given performance, where the impression fostered by the performance is knowingly contradicted as a matter of course”  — is often defined through physical architectures in our off–line lives. These back regions may be places where props for frontstage self–performance are stored — as in Goffman’s examples of the back–office or dressing room — or where activities of solitude are engaged in: as he says, “the same women may leave The Saturday Evening Post on their living room end table but keep a copy of True Romance (“It’s something the cleaning woman must have left around”) concealed in their bedroom.”  But backstages can certainly also be social spaces as well, as in the hotel kitchen or other staff–only work spaces , or in hunting lodges and locker rooms .
Backstages, though they are spaces from which a frontstage performance is understood as a performance, are certainly not spaces free of their own self–performances. The executive’s private office  is a place where professional decorum may be dropped for more informal dealings, and yet the informality performed there may still be a performance; even a self–conscious one intended to project relaxation rather than being a felt expression of it . It’s not even the case that the backstage performance should be regarded as more authentic, whatever we may mean by authenticity — “In a sense,” Goffman says of more or less consciously ‘playing a role’, “and in so far as this mask represents the conception we have formed of ourselves — the role we are striving to live up to — this mask is our truer self, the self we would like to be.” 
Jurgenson and Rey (2012) write that “most commentators who cite [Goffman] today,” at least within the context of social media research, “[overlook] that the front stage and back stage, the visible and invisible, are dialectically linked” , and go on to lament claims that “our lives have become all front stage,”  or a “middle region” between front and back stages . One example might be Pearson’s (2009) discussion of the “glass bedroom,” the image itself seems to represent a loss of truly backstage spaces, but Pearson’s view is more complicated: “Inside the bedroom,” Pearson (2009) writes,
private conversations and intimate exchanges occur, each with varying awareness of distant friends and strangers moving past transparent walls that separate groups from more deliberate and constructed ‘outside’ displays. The glass bedroom itself is not an entirely private space, nor a true backstage space as Goffman articulated, though it takes on elements of both over the course of its use. It is a bridge that is partially private and public, constructed online through signs and language.
While this view is nuanced and valuable, it does not capture the dialectical relation between front and back regions, whereby the front stage exists as a front stage by virtue of its relation to a back stage, and the back stage, for its part, exists as a back stage by virtue of its relation to a front stage.
Jurgenson and Rey’s alternate view of the dialectical linkages between front and back stages calls upon the image of a burlesque “fan dance” that variously reveals and conceals different aspects of the self/selves, and well captures the complexity of SNS users’ approach to privacy and publicity: we do not simply expose or conceal some self or other, but perform aspects of self momentarily in a way which is public and yet also establishes withdrawn, backroom, and intimate connections, either through performing in different ways at different moments, or through “social steganography” (boyd and Marwick, 2011) or other means of projecting different messages and identities simultaneously to different audiences .
But the ambiguity and fluidity of front– and backstages is not new to SNS. Where architectures separating front from back regions are not present, we still move quickly into and out of backstage moments, as in the sotto voce utterance in a crowded room, and may find that, if we walk up to the wrong conversation at a party, that we have entered into a backroom conversation in which we feel deeply uncomfortable — or we may be spared this awkward circumstance by being confronted instead by a sudden (and also awkward) halting of the conversation. In a more general sense, every conversation that ventures the simplest step beyond “small talk” becomes a fan dance of this sort, where personal anecdotes and statements of views and values occupy ambiguous positions as some mixture of fronting, backstage admission, uncomfortable oversharing, purposefully over–intimate pseudo–backstage posturing intended to strategically generate letting–down of fronts on the part of the other, etc., etc. We also certainly see these dynamics in non–SNS forms of online communication as well. Webforums, for example, often establish architectures for front– and backstage communication — most frequently in the form of “off–topic” boards as distinct from the more frontstage “on topic” boards — but discussions in explicitly on–topic spaces often enough get sidetracked and hijacked, and even business–like threads “get personal” in sudden and unexpected ways. These ambiguous hybrids and sudden shifts in the regions of self–performance are not unique to SNS, although we do certainly see them there as well.
What is, however, distinctive and novel in SNS is the regional ambivalence of particular communications. Due to both the invisibility and variability of the architectures used by other users, we do not know when interacting with others whether we are in front or back regions (or both simultaneously) in their self–performance, and regularly produce accidental identity threats to others. For the most part, we interact with others in a state of unawareness whether a given post is visible to their entire network or only parts of it. (It is true that we can gain some knowledge of this by looking to see whether a post is public or limited in some way, but are not likely to check this for any but the most obviously sensitive communications.) Worse yet, numerous posts over time may present to us a view of our friend’s preferred performed identity, through accretion and sedimentation of information flows emanating from the user — and naturally it is that accumulated identity which we interact with in public replies, even though it may be that the identity we see is made up in part of posts directed to a more limited group of which we are unknowingly “insiders.” We’ll address each of these possibilities in more detail.
A user may manage their identity by choosing to only friend those whose communications may be trusted to be in keeping with those representations which will be comfortable for the rest of the user’s network — whether this is because all those friended can be assumed to keep everything on the frontstage, or because all those friended are those which can be comfortably invited backstage. In either case, it is difficult for their friends to know that this has been done. To illustrate a different user–created architecture of identity management, consider the following: I have a friend who frequently posts comments and links that clearly identify political and (non–) religious orientations likely to be unpopular or even dangerous within his line of work. It did not occur to me that these many posts were visible only to a selected group of his friends — a fact I was informed of by private message after he removed a relevant link I had posted on his wall. I had assumed that his friend list had been curated to allow the backstage communication he was engaging me and others in, where in fact differential friend lists were being employed to make some of these communications frontstage and others backstage. In posting this communication on his wall, I had suddenly wandered onstage while continuing a backstage–only dialogue — or, to use a different analogy, it is as if we had been speaking candidly of a sensitive matter on the stage as the audience filed in, and I suddenly turned on his microphone. The difference, of course, is that I didn’t know there was a stage or audience — and how was I to have known in this circumstance? A further complication: while the architecture allows him to make some utterances in a frontstage region and others backstage, it does not allow me access to the same architecture, even now that I know that it is being used. an’t post to his “in–group” — only he can — and so I can only respond in a thread, guessing whether it is front– or backstage, or post on his single and unitary wall, ambivalently both front– and backstage.
Users change these modes of interaction over time as well, producing further identity–threatening uncertainties. Many users start accounts for personal reasons, and expand them over time to ever–more sectors of their lives, resulting in an older network of friends who may continue to interact in a backstage manner even as the user network comes to include increasing numbers and an increasing proportion of professional contacts. Things may work out the other way as well. Like many academics who joined in the earlier days of Facebook, I initially created my account to interact professionally with students and colleagues, expanding over time in the other direction. As I’ve gotten married and had children, my profile and posts have become increasingly personal and backstage. In another kind of example, I have one contact — a graduate student — who systematically defriended her professional–only ‘friends” during a time of great personal stress, in order to convert her hybrid public–private/front–backstage network into a purely personal and backstage network where she could speak freely about her stress and turmoil, and reach out for emotional support in very private and vulnerable ways. I know about this because she mentioned having defriended me for this reason in a post which, however, showed up directly in my newsfeed. The source of this accidental oversharing is Facebook’s current structure of “followers”: if you allow followers, someone, when defriended, is removed from your newsfeed — but you are not removed from hers. From my perspective, her posts had simply become much more personal as she entered into a time of great difficulty, and I believed I had been implicitly invited to become a source of support and care for her, when in fact she had explicitly attempted to remove me from her SNS space as part and parcel of its conversion to a backstage region. After I informed her of this oversharing, having seen her mention having defriended professors and professional contacts, she did not go on to remove me from her followers, and so I now occupy an ambivalent audience to her communications. I am a known party to them, and yet have not fully and clearly been invited backstage; perhaps she would ideally not rather have me there, but is not concerned enough to do anything about it. Perhaps she would be glad to hear words of support and understanding beyond those I spoke initially — or perhaps she prefers civil inattention as a response to these backstage communications leaking into a frontstage context .
Many users, instead of using the architectural remedies mentioned above, accept this ambivalence knowingly, and simply communicate alternately or simultaneously in front– and backstage regions before an undifferentiated audience, counting on the prudence of their friends to self–select whether they are insiders or outsiders to a given topic, exhibiting care and informality in the one case, and civil inattention in the other. This regional ambivalence is characteristic of SNS sites, for, even when there is an attempt to separate out regions, this separation is unclear to others.
These, then, are the factors active in Facebook which have brought our navigation of identity online back from an open ecosystem of identities to the constraints of the dramaturgical self characteristic of off–line interaction, and beyond it as well: (4.1) Real names, (4.2) You are your own avatar, (4.3) Unitary profile, (4.4) Opt–out tagging, (4.5) Obligation to sign up and sign in, (4.6) Promiscuous intermixing of audiences, (4.7) Loss of unintentional architectural supports for identity management, and (4.8) Regional ambivalence. Taken together, these form a distinctive SNS version of context collapse, which Michael Wesch (2009) defined with regard instead to YouTube, as
an infinite number of contexts collapsing upon one another into that single moment of recording. The images, actions, and words captured by the lens at any moment can be transported to anywhere on the planet and preserved (the performer must assume) for all time. The little glass lens becomes the gateway to a black hole sucking all of time and space — virtually all possible contexts — in on itself. 
In SNS, however, the ‘sucking black hole’ is not the collapse of the context of our action into (potentially) all others, but the collapse of the context of our action into numerous but largely known or knowable others — all those contexts of our constitutive and non-constitutive communities, including our friends and friends–of–friends. In this way, the context collapse distinctive of SNS and resultant from the eight factors outlined above has more in common with the “drain hole” in the universe that Jean–Paul Sartre (1993) identifies with the subjectivity of the Other before us, where “[t]he appearance of the Other in the world corresponds ... to a fixed sliding of the whole universe, to a decentralization of the world which undermines the centralization which I am simultaneously effecting” . In Sartre’s more fundamental and existential example, we discover that we have a secret and ultimately unknowable aspect, trapped within the subjectivity of the Other — who we are cannot be understood in full by ourselves alone, for we are also that aspect of ourselves which disappears into the perception of the Other. In SNS, we suffer a context collapse which presents to us our simultaneous co–presence among multiple contexts which are, however, all our communities; groups of persons with which we more–or–less identify and through which we more–or–less conceive of ourselves. This, then, is not the “black hole” of innumerable perceptions to which we have no particular connection, but a more specific and threatening “sliding” wherein we know ourselves to be simultaneous present in different contexts each of which we identify with and have invested in. The distinctive context collapse of SNS is Sartrean in that each of the selves created in the subjectivities present within our network are meaningfully and pressingly felt as relevant to who we “really” are; a version of ourselves which has some bearing on our “true” selves insofar as we have made personal investments and valuations in the perspectives of those persons and communities whose unknowable perception and reception our performances disappear into. “Haters gonna hate” does not free us from identity threat on Facebook as it does on YouTube, and negative perceptions, stated or unstated, have a real bearing on our lives and self–conception, rather than being limited to judgments of unconnected others to a self–performance broadcasted in disconnection from our constitutive communities and personal relationships.
This modified and intensified return, in our online lives, to the difficulties of managing contextual selves native to dramaturgical identity, combined with additional pressures to present oneself in a unitary fashion, as if each presentation were clearly and immediately a presentation of a singular true self, gives rise to a distinctive variety of selfhood required in the era of Facebook. We can call this dramauthentic identity.
5. Identity in the Facebook era: Strategies of self–exposure
In dramauthentic identity, we must perform selves to constitutive communities not variously — as in much of off–line life, where within one context one aspect of the self may be presented, and within another context, another aspect — but simultaneously. In dramaturgical self–presentation, identity is variably anchored, whereas in dramauthentic self–presentation, identity is multiply anchored.
I have chosen the term “dramauthentic” not because identity on Facebook must be more authentic than the off–line, dramaturgical self, but rather because authenticity is an issue for self–presentation on SNS in a new and prominent way, due to the factors outlined above, which collectively result in a constant need to perform within multiple contexts as if one has only a singular, unitary self, constant through all constitutive communities — which need, however, may be met either through an attempt at reconciliation of self–presentations into a more–or–less authentic singular performance of self, or through a newly more self–conscious attempt at contextual and managed performances of self which have a newly more self–conscious intention to categorize and select from relationships for differential access to an identity beyond and in some way inclusive of those performances .
It should be emphasized that I do not mean to presume that a unitary and singular performance of self is more authentic than variably anchored and contextual self–performances — but, instead, that I mean to draw out how the perspective placed in code in Facebook and to a significant extent in other SNS as well implies this privileging of unitary self–performances. It should not be surprising that this is the judgment implied by Facebook’s architecture: Mark Zuckerberg is quoted in an interview with David Kirkpatrick (2010) saying, “You have one identity ... The days of you having a different image for your work friends or co–workers and for the other people you know are probably coming to an end pretty quickly ... Having two identities for yourself is an example of a lack of integrity” . Personally, I am inclined to say instead that multiple self–presentations allow for more authentic exploration and expression of the self, as this allows, in Heidegger’s (1962) terms, for us to navigate our way through our entanglement in the “They” while maintaining a separate interiority with a critical distance from various social selves, allowing for more authentic self–development. Those readers who wish to deny the meaningfulness of any “authentic” self at all will also disagree with Zuckerberg, and should also see no benefit for authenticity in requiring unitary self–presentation. In any case, the use of the term “authentic” in describing what will be called here “dramauthentic identity” should not be taken as a claim, implicit or explicit, that an “authentic self” exists, or is either well– or poorly–served by requiring a unitary self–presentation. The point is instead merely that authenticity as a social norm and as a constant task at hand in communicative practice, structured in new ways in code, is a pressing issue for us in the age of SNS more so than in not only our previous online lives, but more so than in our previous off–line lives as well.
In multiply–anchored self–presentation, multiple aspects of the self must be either (5.1) implied to one community without clear presentation to another, (5.2) reconciled by presenting some aspects of identity to a community from which we might otherwise prefer to hide them, or (5.3) reduced in scope and depth, limiting exposure of aspects of identity which do not match with preferred self–presentations within various contexts.
5.1. Mixed exposure
A mixed strategy of exposure is difficult to maintain and takes place under constant threat of context collapse, as some members of some of a user’s communities are unlikely to know the proper boundaries of self–performance relative to other communities of the user which may be present. For example, you might be friends with several public school teachers, but have no way of knowing without careful observation of their posting habits (and perhaps their deleting of posts) which are more or less prepared for communications which they would not wish to have seen by students or students’ parents — whether their wall is public, limited, or private; whether they are friends with students, with former students, with students’ parents, with former students’ parents, or none of these; or whether they are friends with administrators or school board members. This danger, of course, extends far beyond tagging someone dancing on the table last Saturday, but can include something as simple as sending along a news story that might be of interest, but which is critical of certain educational practices or policies, or a video implying an even mildly risqué sense of humor. Some users circumvent the lack of control of self–presentation created by the ability of others to post and tag by simply refusing to friend anyone who is not likely to understand the acceptable boundaries of such actions, or by refusing to friend anyone who they do not wish to be audience to the majority of anticipated third–party sharing.
5.2 Agonistic exposure
The strategy of exposure of unifying self–presentation by a challenging sharing of aspects of identity we might usually hide from some communities, we could call an agonistic exposure. Paradigmatically, we can consider choosing, under the constraints of loss of control through third–party sharing, to “come out” to one’s parents — although numerous other forms of self–disclosure (faith, politics, etc.) to numerous other communities (work, church, etc.) could be used as equally clear and dramatic examples. These paradigmatic examples are representative of numerous smaller disclosures which, though perhaps less dramatic, represent thoroughgoing ways in which small, contextually–relative interactions targeted to a particular audience become directed indiscriminately towards a promiscuously–intermixed assemblage of audiences. These disclosures are typically of innocuous information, such as hobbies, interests, one’s sense of humor, or daily minutia of home or family life, but for that are not without significant social meaning. Each of these disclosures can be a challenge to others; they say this is who I am, accept me; or this is the way my life is lived, how does yours compare?; or I am a full person beyond your contextual knowledge of me, will you choose to involve yourself with me further? When a coworker with whom we have only a formal business relationship posts funny cat videos, we must ask whether we choose to make ourselves the audience by responding, and in so doing, bring a new element into that relationship. When an acquaintance reaches out for emotional support, we must decide whether we think of ourselves as, or are willing to become, among those in his network who can be depended upon in difficult times — and whether our “opting in” to that support network will be viewed as a welcome deepening of that relationship or as an unwelcome intrusion. When a family member posts about politics, we must decide whether our relationship may be threatened or deepened by “going there.”
The costs of adopting an agonistic strategy exposure are potential alienation or abandonment, but the potential benefits may also be significant. The potential benefits for the user in finding that some separation between various self–presentations was unnecessary is clear: maintaining such separation requires effort and care, may be psychologically difficult or stressful, and may diminish the value and depth of personal relationships. The potential benefits for others are perhaps less obvious: by choosing an agonistic exposure which embraces the promiscuous intermixing of audiences, recipients of communications are freed from the normal constraints placed on relationships, and may choose to pursue connections forestalled by the contextual communication which takes place offline and online outside of SNS. A professional contact may never have been approached as, for example, a person to turn to for casual conversation or discussion of parenting, but agonistic exposure allows constant opportunities to allow parties to a relationship to reimagine their relationship and explore what it may or may not be able to become.
5.3. Lowest–common–denominator exposure
The third strategy of exposure, reducing the scope and depth of our self–presentation on SNS to limit exposure of information which may be troubling to some members of some communities, we could call lowest–common–denominator exposure. Here, the user may offer a sanitized and expurgated self–presentation, in which the threat of mixed exposure and the danger of agonistic exposure are avoided at the expense of the decreased ability to use SNS to maintain and develop deeper relationships more closely connected to mutual recognition and emotional support — “symmetrical esteem,” as Schlesselman–Tarango  puts it, drawing on Axel Honneth’s (1996) Struggle for recognition. Some users, moving even further, respond minimally to the obligation to sign up and sign in, using SNS only as necessary to maintain non–SNS–based relationships with those who generate obligations to sign up or sign in — for example, accepting friend requests in order to avoid perceived impoliteness, but never making friend requests; choosing not to upload a profile picture; or accessing the account only when notified of a message or posting, but choosing not to post or initiate interactions.
These three strategies of exposure in the context of multiply–anchored self–presentation have focused on ways in which aspects of the self may be revealed or hidden from various communities. They have, in this way, spoken as if the performance is an attempt at a representation of a previously– or elsewhere–formed self, whether that performance is (as in agonistic exposure) a performance before constitutive communities, intended to reveal information, or (as in lowest–common–denominator exposure) a performance intended to conceal information tied to one or more constitutive community, or (as in mixed exposure) a performance intended to reveal information differently to different constitutive communities. All this speaks of self–presentation of a more–or–less underlying identity within communities rightly regarded as constitutive, as implied by the use of the term “exposure” rather than e.g., “construction.” We should not, however, ignore self–construction of identities which are more invented or aspirational than expressive of a more–or–less underlying identity, and self–construction within communities which are not constitutive so much as they are chosen as activities of self–invention and growth or change of self–identification. Surely, though, there are no clear boundaries to be found here, for at least three reasons.
First, there is no clear boundary between online and off–line identity construction. Character formation is not an off–line–only process and, especially as successive generations get online earlier and earlier in life, the self that we expose through online performances cannot be assumed to be pre–formed prior to online performances, but is instead formed through online and off–line performances.
Second, when we join a community or adopt an identity under an aspirational self–performance, this process of self–reinvention does often enough lead to an internalization of self–consciously adopted values, and a transition from aspirational attributes into more permanent and unselfconsciously held elements of character. Sometimes we just fake it, but sometimes we fake it until we make it.
Third, we are not always well aware of when we are performing a truthful exposure of an existing self or facet of ourselves, and when we are performing an aspirational or even fairly fictionalized self. Indeed, we may wish to question whether there is any clear self to be found outside of various self–performances, to which we could even possibly be “truthful.”
For these reasons, we should take care to note that self–performances involve a dialectical relation between self–exposure and character formation — each takes place only alongside and by means of the other, as Aristotle (1984b) noted long ago in the Nicomachean Ethics,
This, then, is the case with the excellences [virtues] also; by doing the acts that we do in our transactions with other men we become just or unjust, and by doing the acts that we do in the presence of danger, and being habituated to feel fear or confidence, we become brave or cowardly ... . Thus, in one word, states [of character] arise out of like activities. This is why the activities we exhibit must be of a certain kind; it is because the states [of character] correspond to the differences between these. It makes no small difference, then, whether we form habits of one kind or of another from our very youth; it makes a very great difference, or rather all the difference. 
It is beyond the scope of this investigation to consider what role dramauthentic identity plays in character formation — the important point to be made here is more simply that we cannot easily separate aspirational or even fictitious self–performances from truthful self–representation, but find instead a complex and dialectical relation to be at work.
The remaining question, then, is how these strategies of self–exposure are used to construct self–identities. While remaining agnostic regarding how these identities are related to authentic selves or true character — indeed, remaining agnostic regarding how meaningfully such things can be said to exist — we can outline four strategies of identity construction that, along with these three strategies of self–exposure, constitute dramauthentic identity.
6. Identity in the Facebook era: Strategies of identity–construction
We may chart different strategies of identity–construction on two axes, resulting in a way of imagining a field of possible modes of interaction falling into four quadrants. On the x–axis, we can chart a continuity from (-1, 0) divergent constructions of identity to (1, 0) unitary constructions of identity, and, on the y–axis, a continuity from (0, -1) reactive constructions of identity to (0, 1) proactive constructions of identity. Charting on these axes results in four quadrants which can be characterized by their extreme points; I will describe them under these terms: I. spectacular identity (1, 1), II. untidy identity (-1, 1), III. distributed identity (-1, -1), and IV. minimized identity (1, -1).
Figure 1: Four forms of dramauthentic identity in quadrants, defined using divergent–unitary as x–axis and reactive–proactive as y–axis.
These are strategies of interaction which most SNS users move among — this is not meant to be a typology of users, but rather a view of different strategies of use that users engage in at different moments, although it is certainly true that users may have a more–or–less strong tendency towards or away from one or more certain kinds of use. These strategies are shown on a field to represent how, while what will be described below are four extremes of behavior, interactions exhibit these different strategies not absolutely but to a greater or lesser degree.
6.1. I. Spectacular identity
This is a strategy of interaction which actively seeks to create a spectacle of identity in a sense approximating that of Debord’s Society of the spectacle . Here, a unitary identity is actively produced as a medium through which the network is intended to interact with the user. Spectacular identity is groomed as a representative of the user through judicious choice of profile identifications, through kinds of postings and comments, and through those included and excluded in friend lists, circles, or on Twitter, follows, influence, and proportion of follows to followers. Baroncelli and Freitas (2011) have also drawn a parallel between certain sorts of SNS user activities and Debord, wherein they claim that in self–spectacularization, “personal life becomes a capital to be shared with other people,” and that, influenced by celebrity culture (Hedges, 2009), “although most individuals do not attempt to become a celebrity, many of them often end up organizing life according to the underlying codes of the celebrities’ culture,” seeking recognition through a repackaging of the self as unique, talented, and situated within a personal screenplay (Baroncelli and Freitas, 2011).
It is this strategy of interaction which Muhr and Pedersen (2010) seem to have in mind when characterizing the function of Facebook identity in terms of Žižek’s (1998) idea of interpassivity, where the SNS representation of value commitments and fulfilling life activities can free the user from actually choosing and committing to values or actually finding activities of life meaningful. Similar to the way tourists may use a photograph in which an important and culturally weighty landmark may be seen behind them as a consumable substitute for the cultural value which might have been obtained by actual engagement with that landmark and its history, the Facebook identity can be viewed as a location wherein the markers of life in accordance with the user’s self–conception may be displayed, displacing the drive for realization of that life in favor of mere acquisition of evidence of its realization. But, while this is an important and troubling possibility, I see no reason to suppose that the spectacular use of SNS identity–construction as an ambassador of the self should preclude also actually striving for or achieving those aspects of identity of which it is a spectacle.
Spectacular identity is not necessarily merely a front; it may be an idealized, aspirational, or fictitious self, but may also be a false identity only insofar as any such spectacle of self necessarily differs from that of which it is a spectacle. The spectacular self, then, is a groomed identity manufactured as a consumable product and as an image preceding and conditioning interactions, whether or not it is constructed of veridical representations.
6.2. II. Untidy identity
This is a strategy of interaction which is proactive in posting, sharing, and demonstrating values, commitments, and identity, but does not seek to present a groomed, unitary self–image, instead allowing divergent markers of character and identity to coexist. This untidy self–presentation — or, perhaps, refusal to present a unitary self — occurs primarily through two common sorts of interaction: unedited third–party construction of the self through tagging and posting, and personal over–sharing through broadcast of communications emerging from or proper to a narrower audience.
In third–party self–construction, users are tagged or mentioned in notes, photos, etc. and in this way a depiction of the self, sometimes even a false or outdated depiction, is projected onto the user. This also occurs through the direct posting of material to the user’s wall, e.g., a news story accompanied by “We were just talking about this!” or “Thought you’d be interested.” Through these third–party postings, an involuntary construction of self occurs, which may be divergent from one or more voluntary self–presentations either online or off–line.
In addition to this involuntary self–construction, users will not infrequently post generally something proper to a more specific audience. This over–sharing may be done agonistically, where there is a challenge to other audiences to choose to ignore, participate, or tolerate the communication; accidentally, where the user would prefer to have removed several friends from access but did not have the potential for oversharing in mind; apathetically, where the user neither intends to challenge nor forgets to target, but simply can’t be bothered to maintain and groom self–presentation to differing audiences; invitingly, where the communication is of direct relevance to a few, but the user is open to the possibility that others may have an unforeseen interest or concern in the topic or information; or self–importantly, where the user overestimates, sometimes markedly, the number of recipients who will find interest in the communication, for example, of where they are currently waiting in line, or how they feel about that week’s celebrity scandal.
Through involuntary third–party projection of identity and various forms of over–sharing, an untidy identity may emerge, in which identity on SNS appears as an overlapping and kaleidoscopic pastiche of different concerns, interests, contexts, communities, and moods. A unitary identity is not constructed, as in spectacular identity, but the self is instead projected backwards as some center point or line of contiguity connecting this variety of appearances.
6.3 III. Distributed identity
This strategy of interaction is reactive rather than proactive, choosing to respond to the content of others rather than creating or posting content. This allows the user to maintain divergent self–presentations while minimizing untidiness, due to the lack of connections between friends to whom the user might present herself differentially.
As noted previously, in off–line communications, the mundane conditions of space and time provide architectures that determine audiences contextually, limiting the threat that the actor will be called to account for disparities between self–presentations. By posting in response to the posts of others, or by posting on the walls of others — rather than initiating interactions through posts of one’s own — it is possible to perform various identities before audiences sharing similar contexts of concern and interpretation while minimizing the visibility of those disparate self–performances through distribution into various networks constructed by and around others.
This is no real solution to privacy concerns — if the user’s privacy settings are open to the public or to friends of friends, discussions on friends’ walls will be accessible to other friends, and we are, furthermore, quite unlikely to know the privacy settings of our friends — but our intention here is to investigate strategies of self–presentation, not methods of locking down informational access. As a strategy for self–presentation and identity–construction, distributed identity allows for contextual communication, avoiding both untidiness and the construction of a unified identity by at least attempting to decline to participate in wider and perhaps more contentious forms of social interaction.
6.4. IV. Minimized identity
This is a strategy of interaction, also reactive rather than proactive, but unitary in presentation, in which the user makes no active attempt at unitary self–performance, but instead maintains a unitary self–presentation through filtering of viewable content, creating a locked–down representation of self. This minimization may be accomplished by selective untagging and choosing not to post or upload pictures, sometimes refusing to post even a profile picture. By choosing to avoid self–presentation altogether, speaking only when spoken to, so to speak, a user may avoid a positive construction of self on SNS, maintaining only a placeholder onto which their friends may project the image of the user presented elsewhere online or off–line.
It may be that the user makes us of SNS privately and only when and as necessary — responding to the obligation to participate by setting up a minimal profile and visiting only when notified of friend requests, messages, or wall postings — but this may not be the case: this minimized strategy of identity construction is actually compatible with a fair amount of regular SNS use. It may be that the user feels no impulse to perform the self, or to share with or respond to friends, but still chooses to be a silent audience to their communications. It may be that the user, out of discomfort with the other compromises above, chooses to respond to SNS communications off–line, or through chat or private message, rather than constructing an SNS identity. Regardless, a minimized strategy of identity construction does not necessarily imply a lessened amount of time spent on SNS, but only a lessened engagement with others through the SNS itself.
In some sense, the minimized strategy of identity construction is a form of dramauthentic identity that attempts to opt out of performing a dramauthentic identity at all. SNS are viewed as a source of information rather than interaction, and self–performance is engaged in only in the more familiar or less dangerous spaces of off–line communication, through one–on–one electronic communications, or in online spaces, like Web forums, that grant users greater architectures of context control.
7. Concluding comments and directions for future research
Through these three kinds of SNS exposure (mixed, agonistic, and lowest–common–denominator) and these four kinds of identity construction (spectacular, untidy, distributed, and minimized), we see a great variety of ways of performing dramauthentic identity, suitable to the great variety of kinds of social circumstances we find ourselves within, and the great variety of psychological, emotional, and social dispositions which influence identity management in circumstances, like those on SNS, where managing personal identity represents an unending succession of uncertain and imperfect compromises between our various roles, relationships, and responsibilities. This detailed investigation of conditions, limits, and strategies of post–Facebook navigation of identity in community has hopefully made clear why I began by claiming that the problematic relationship between the online and off–line selves has been settled in a troublesome manner: the question of whether the online self is a true representation of the off–line self has been resolved — but it has been resolved by tying the online self to the variety of off–line selves in a way that requires a stability and unity of the online self which no longer represents, but instead constricts the off–line self, and, furthermore, produces ever more blurring between off–line and online self–presentations. I hope, also, that another of my initial claims will now be no longer suggestive and obscure, but quite clear and concrete: in the current moment in technologically–enmeshed identity, the strikingly voluntary construction of self–narrative of earlier online self–identity has been re–attached to less–voluntary and involuntary aspects of off–line life, collapsing the wild and limitless freedom of identity–construction partially back into the familiar and everyday dramaturgical self–construction of multiple self–presentations to various constituent communities — but now with an in some ways far greater architecturally imposed need to reconcile those selves with one another, due to the promiscuous intermixing of communities in the information feeds of our Facebooked sociality. The online self is no longer a reflection of or departure from the off–line self, or at least not merely so, but is instead a space in which off–line and online selves are called to account for their diversity, sublating these public, private, and contextual online and off–line performances in a way requiring a new kind of self–performance within the collapsed contexts of our constitutive communities.
While this investigation has been quite lengthy, it should be noted that it is nevertheless still quite incomplete. I hope that it shall prove of use to those seeking a model and a theoretical framework to support more specific, applied, and concrete research on particular aspects and dynamics of identity performance on SNS. I also hope that it will provoke theoretical critique and expansion. A specifically feminist or queer–theoretic perspective on the topic would, I am confident, identify important and quite general aspects of identity performance on SNS that I have either failed to give proper weight and consideration to, or have failed to even notice and outline. Similarly, Marxist or post–Marxist perspectives will discover important constraints and dynamics which do not appear in this account, but which are also necessary for this theory to be synoptic and well–founded. Consideration of kinds of social capital construction could helpfully expand this model by considering strategies directed towards either bonding or bridging social capital (Aldridge, et al., 2002) and either cognitive or structural social capital (Uphoff and Wijayaratna, 2000) — or simply by looking into whether different user strategies are more, less, or in some way variously or simultaneously directed towards pursuing and maintaining either strong or weak ties (Granovetter, 1973). Considerations of regional or national cultures of SNS use would certainly be revealing and valuable enrichments of the model. We might note, for example, that Gil de Zúñiga and Valenzuela (2010) found that in the United States “women, African Americans, and Hispanics use social networks more heavily than men and non–Hispanic whites” , which may well be accompanied by different cultures and strategies of use among these various demographics. I am also aware of some scholarship currently underway which could add valuable considerations of generational dynamics in strategies of exposure and identity construction. For example, provisional findings from Vittadini (2013) show that the “privacy culture” of older generations tends to begin from a perception of SNSs as public spaces, leading to a tendency towards what I have called a “lowest common denominator” strategy of exposure, while younger generations have a “privacy culture” more in keeping with a variable “contextual integrity” (Nissenbaum, 2004), tending towards what I have called a “mixed” strategy of exposure, and perhaps even what I’ve called a strategy of “distributed identity construction.” These generational considerations promise to offer a great deal toward a fuller, more substantial, and more robust realization of a theory of identity performance on SNS.
I hope that, in this way, this articulation of a basic model of “dramauthentic identity” can serve as perhaps a skeleton, or at least a spur, for the formation of a full, well–rounded, and strongly– and multiply–grounded theory.
About the author
D.E. Wittkower is Assistant Professor in the Department of Philosophy and Religious Studies at Old Dominion University.
E–mail: dwittkow [at] odu [dot] edu
1. As Paasonen (2002) pointed out, when we look at actual behavior, even long prior to the era of SNS — with the MOOs and MUDs that Turkle well–addressed as clear exceptions — a great many common and everyday uses of the Internet do and did not push online identity too far afield from off–line identity, for example, “[h]ome page practices are less about fantasies of transgressing one’s corporeality and social location, than about manifesting one’s presence online“ (p. 30).
2. As Hongladarom (2011) notes, “[c]ontrary to the received view that Facebook and Twitter tend to reflect the true identity and personality of the users (since according to the view users tend to be more revealing of their personal information), in the Thai context at least the characteristics of the earlier anonymous online discussion tends to carry over onto social networking sites, though in a visibly different form” (p. 537). The common use of personally unidentifiable images as profile pictures and of pseudonyms rather than real names, in the Thai context, is due in part to limits on freedom of speech — but, certainly, this behavior, though more widespread in the Thai context, is seen elsewhere as well.
3. Rainie and Wellman, 2012, p. 16.
4. Rainie and Wellman, 2012, p. 141.
5. Although only insofar as those norms and code enter into the phenomenology of the user. There is a different story to be told from the back end, and an interesting one: for a nice discussion of “algorithmic friendship,” the socio–technical hybrid social environment fostered by Facebook’s hidden EdgeRank and GraphRank algorithms, see Bucher (2013).
6. Briggle, 2008, p. 73.
7. Martin, et al., 1988; Dean, 2010, pp. 49–51.
8. Habermas, 1989, p. 50.
9. Sandel, 1982, p. 150.
10. Although this is becoming more difficult, as Google makes it increasingly difficult to refuse to tie YouTube accounts from real–name Google+ profiles.
11. In less commonplace examples, such as the much–maligned furry community, we may be skeptical that self–presentation is the proper term for interacting with others as e.g., an anthropomorphized squirrel, or that the “true self” of a reclusive thirty–something woman might be a homosexual male were–tiger. Even were we to insist upon the term “self–construction” rather than “self–presentation” in these cases, it is clear that those who engage in these interactions find them to be meaningful, and the impossibility of integrating these constructed selves with the other selves presented in other contexts more clearly limited by the facts of physical reality does not present any immediate reason why these fantastical self–identities should be discarded out of hand as illusory.
12. Papacharissi, 2010, p. 69.
13. Strater and Lipford, 2008, p. 111.
14. Hull, et al., 2011, p. 294.
16. Lessig, 2006, p. 61.
17. Searches by name or zip code can in fact be done at the U.S. national sex offender registry (http://www.nsopw.gov/); another governmental site (http://www.familywatchdog.us/) also places results on a Google map, with a color code and further information available by clicking on each address.
18. Papacharissi, 2010, p. 142.
19. Goffman, 1959, p. 112.
20. Goffman, 1959, p. 42.
21. Goffman, 1959, pp. 114–122.
22. Goffman, 1959, p. 125.
23. Goffman, 1959, p. 126.
24. Goffman, 1959, p. 134.
25. Goffman, 1959, p. 19.
26. Jurgenson and Rey, 2012, p. 65.
28. Jurgenson and Rey, 2012, p. 66.
29. Jurgenson and Rey, 2012, pp. 66–67.
30. Goffman, 2010, pp. 4–5fn.
31. Wesch, 2009, p. 23.
32. Sartre, 1993, p. 343.
33. Jodi Dean (2010) has noted another rather different connection with authenticity endemic to blogs but applicable to SNS as well:
More significant is the performance of authenticity enabled by Twitter — along with Facebook’s newsfeed and the mood updates on MySpace: the short glimpses into someone’s life as it is being lived seem real. The seem real in part because they are only glimpses, fragments, and indications rather than fully formed and composed reflections and in part because we witness them being seen by others. (p. 36)
34. Kirkpatrick, 2010, p. 199.
35. Schlesselman–Tarango, 2013, p. 7.
36. Aristotle, 1984b, p. 1,743: 1103b12–26.
37. Although the way that this connection with Debord is being used here is somewhat different from that in Vejby and Wittkower (2010).
38. Gil de Zúñiga and Valenzuela, 2010, p. xxxiv.
Stephen Aldridge, David Halpern, and Sarah Fitzpatrick, 2002. Social capital: A discussion paper. London: Performance and Innovation Unit.
Aristotle, 1984a. “Magna Moralia,” In: J. Barnes (editor). Complete works of Aristotle: The revised Oxford translation. Princeton, N.J.: Princeton University Press, pp. 1,868–1,922.
Aristotle, 1984b. “Nicomachean Ethics,” In: J. Barnes (editor). Complete works of Aristotle: The revised Oxford translation. Princeton, N.J.: Princeton University Press, pp. 1,729–1,867.
Lauane Baroncelli and André Freitas, 2011. “The visibility of the self on the Web: A struggle for recognition,” Proceedings of the Third International Conference on Web Science, at http://andrefreitas.org/papers/WebScieence_Baroncelli_Freitas_2011%20FV.pdf, accessed 31 March 2014.
Roland Barthes, 1981. Camera lucida: Reflections on photography. Translated by Richard Howard. New York: Hill and Wang.
Taina Bucher, 2013. “The friendship assemblage: Investigation programmed sociality on Facebook,” Television & New Media, volume 14, number 6, pp. 479–493.
doi: http://dx.doi.org/10.1177/1527476412452800, accessed 31 March 2014.
danah boyd and Alice Marwick, 2011. “Social privacy in networked publics: Teens’ attitudes, practices, and strategies,” paper presented at the Oxford Internet Institute Decade in Internet Time Symposium; draft at http://www.danah.org/papers/2011/SocialPrivacyPLSC-Draft.pdf, accessed 31 March 2014.
Adam Briggle, 2008. “Real friends: How the Internet can foster friendship,” Ethics and Information Technology, volume 10, number 1, pp. 71–79.
doi: http://dx.doi.org/10.1007/s10676-008-9160-z, accessed 31 March 2014.
Dean Cocking and Steve Matthews, 2000. “Unreal friends,” Ethics and Information Technology, volume 2, number 4, pp. 223–231.
doi: http://dx.doi.org/10.1023/A:1011414704851, accessed 31 March 2014.
Jodi Dean, 2010. Blog theory: Feedback and capture in the circuits of drive. Malden, Mass.: Polity.
Judith Donath and danah boyd, 2004. “Public displays of connection,” BT Technology Journal, volume 22, number 4, pp. 71–82.
doi: http://dx.doi.org/10.1023/B:BTTJ.0000047585.06264.cc, accessed 31 March 2014.
Luciano Floridi, 2005. “The ontological interpretation of informational privacy,” Ethics and Information Technology, volume 7, number 4, pp. 185–200.
doi: http://dx.doi.org/10.1007/s10676-006-0001-7, accessed 31 March 2014.
Homero Gil de Zúñiga and Sebastián Valenzuela, 2010. “Who uses Facebook and why?” In: D.E. Wittkower (editor). Facebook and philosophy: What’s on your mind? Chicago: Open Court, pp. xxxi–xxxvii.
Erving Goffman, 2010. Relations in public: Microstudies of the public order. New Brunswick, N.J.: Transaction Publishers.
Erving Goffman, 1959. The presentation of self in everyday life. Garden City, N.Y.: Doubleday.
Mark Granovetter, 1973. “The strength of weak ties,” American Journal of Sociology, volume 78, number 6, pp. 1,360–1,380.
Jürgen Habermas, 1989. The structural transformation of the public sphere: An inquiry into a category of bourgeois society. Translated by Thomas Burger with the assistance of Frederick Lawrence. Cambridge, Mass.: MIT Press.
Chris Hedges, 2009. Empire of illusion: The end of literacy and the triumph of spectacle. New York: Nation Books.
Martin Heidegger, 1962. Being and time. Translated by John Macquarrie and Edward Robinson. Oxford: Blackwell.
Soraj Hongladarom, 2011. “Personal identity and the self in the online and offline world,” Minds and Machines, volume 21, number 4, pp. 533–548.
doi: http://dx.doi.org/10.1007/s11023-011-9255-x, accessed 31 March 2014.
Axel Honneth, 1996. The struggle for recognition: The moral grammar of social conflicts. Translated by Joel Anderson. Cambridge, Mass.: MIT Press.
Gordon Hull, Heather Lipford, and Celine Latulipe, 2011. “Contextual gaps: Privacy issues on Facebook,” Ethics and Information Technology, volume 13, number 4, pp. 289–302.
doi: http://dx.doi.org/10.1007/s10676-010-9224-8, accessed 31 March 2014.
David Kirkpatrick, 2010. The Facebook effect: The inside story of the company that is connecting the world. New York: Simon & Schuster.
Lawrence Lessig, 2006. Code: Version 2.0. Second edition. New York: Basic Books.
Luther H. Martin, Huck Gutman, and Patrick H. Hutton (editors), 1988. Technologies of the self: A seminar with Michel Foucault. Amherst: University of Massachusetts Press.
Sara Louise Muhr and Michael Pedersen, 2010. “Faking it on Facebook,” In: D.E. Wittkower (editor). Facebook and philosophy: What’s on your mind? Chicago: Open Court, pp. 265–276.
Helen Nissenbaum, 2004. “Privacy as contextual integrity,” Washington Law Review, volume 79, number 1, pp. 119–157; version at https://www.nyu.edu/projects/nissenbaum/papers/washingtonlawreview.pdf, accessed 31 March 2014.
Susanna Paasonen, 2002. “Gender, identity, and (the limits of) play on the Internet,” In: Mia Consalvo and Susanna Paasonen (editors). Women & everyday uses of the Internet: Agency & identity. New York: Peter Lang, pp. 21–43.
Zizi Papacharissi, 2010. A private sphere: Democracy in a digital age. Malden, Mass.: Polity.
Zizi Papacharissi, 2002a. “The self online: The utility of personal home pages,” Journal of Broadcasting & Electronic Media, volume 46, number 3, pp. 346–368.
doi: http://dx.doi.org/10.1207/s15506878jobem4603_3, accessed 31 March 2014.
Zizi Papacharissi, 2002b. “The presentation of self in virtual life: Characteristics of personal home pages,” Journalism & Mass Communication Quarterly, volume 79, number 3, pp. 643–660.
doi: http://dx.doi.org/10.1177/107769900207900307, accessed 31 March 2014.
Erika Pearson, 2009. “All the World Wide Web’s a stage: The performance of identity in online social networks,” First Monday, volume 14, number 3, at http://firstmonday.org/article/view/2162/2127, accessed 31 March 2014.
Lee Rainie and Barry Wellman, 2012. Networked: The new social operating system. Cambridge, Mass.: MIT Press.
Michael Sandel, 1982. Liberalism and the limits of justice. New York: Cambridge University Press.
Jean–Paul Sartre, 1993. Being and nothingness: A phenomenological essay on ontology. Translated and with an introduction by Hazel E. Barnes. New York: Washington Square Press.
Gina Schlesselman–Tarango, 2013. “Searchable signatures: Context and the struggle for recognition,” Information Technology and Libraries, volume 33, number 3, pp. 5–19.
doi: http://dx.doi.org/10.6017/ital.v32i3.3093, accessed 31 March 2014.
Katherine Strater and Heather Lipford, 2008. “Strategies and struggles with privacy in an online social networking community,” BCS–HCI ’08: Proceedings of the 22nd British HCI Group Annual Conference on People and Computers: Culture, Creativity, Interaction, — volume 1, pp. 111–119.
Sherry Turkle, 1995. Life on the screen: Identity in the age of the Internet. New York: Simon & Schuster.
Sherry Turkle, 1984. The second self: Computers and the human spirit. New York: Simon and Schuster.
Norman Uphoff and C.M. Wijayaratna, 2000. “Demonstrated benefits from social capital: The productivity of farmer organizations in Gal Oya, Sri Lanka,” World Development, volume 28, number 11, pp. 1,875–1,890.
doi: http://dx.doi.org/10.1016/S0305-750X(00)00063-2, accessed 31 March 2014.
Rune Vejby and D.E. Wittkower, 2010. “Spectacle 2.0?” In: D.E. Wittkower (editor). Facebook and philosophy: What’s on your mind?. Chicago: Open Court, pp. 97–108.
Nicoletta Vittadini, 2013. “Privacy and SNS: Generational difference in managing privacy and disclosure,” paper presented at Social Media: The Fourth International Transforming Audiences Conference; see http://snsitalia.wordpress.com/2013/09/12/privacy-and-sns-generational-differences-in-managing-privacy-and-disclosure-nicoletta-vittadini/, accessed 31 March 2014.
Michael Wesch, 2009. “YouTube and you: Experiences of self-awareness in the context collapse of the recording Webcam,” Explorations in Media Ecology, volume 8, number 2, pp. 19–34.
Slavoj Žižek, 1998. “The interpassive subject,” at http://www.lacan.com/zizek-pompidou.htm, accessed 23 July 2011.
Received 15 September 2013; revised 25 December 2013; accepted 2 February 2014.
This paper is licensed under a Creative Commons Attribution–NonCommercial–NoDerivs 3.0 Unported License.
Facebook and dramauthentic identity: A post–Goffmanian model of identity performance on SNS
by D.E. Wittkower.
First Monday, Volume 19, Number 4 - 7 April 2014