Worried About the Wrong Things

by Jacqueline Ryan Vickery

  The complexity of understanding privacy online is due in part to the fact that cues aren’t always neatly distinct or visible. The architecture of platforms means that we are not always aware of who has access to online information in the same way we might be in the physical world. Five characteristics of networked publics that render the spaces distinct from the physical world have been identified by boyd (2010): online information is persistent, searchable, replicable, scalable, and locatable. This makes it particularly difficult to assess expectations of privacy and to express different identities in delineated contexts. These dynamics lead to what boyd describes as “context collapse”—moments and practices in which social, spatial, and temporal boundaries converge, often in invisible, unpredictable, and uncontrollable ways. Again, this is not wholly unprecedented or unique to the virtual world, as boundaries certainly collide in the physical world as well, yet the delineations and effects can be more subtle or even invisible online. Additionally, we often have less control of the contexts and visibility of information when we are online.

  As an example of context collapse in the physical world, consider what happens when a student on a date runs into a teacher or when a mother overhears her daughter having an intimate conversation with friends in her bedroom. In each of these examples, an individual is articulating and expressing a particular identity in a particular context and the unanticipated presence of someone outside of that context disrupts the social setting and expectations. However, in the physical world we often are aware when these distinct contexts overlap and can adjust our sense of identity and behavior accordingly—for example, we see our teacher and adjust our behavior, or we hear our mother and decide to whisper or change the subject. When we experience context collapses online, they can be more complicated because it is be easier for our audiences to remain unknown and invisible. And because our interactions are archivable and searchable, it can be more difficult to imagine all current and future audiences. When we share something online, we have an intended or imagined audience in mind (Marwick and boyd 2010), yet the information persists and can be replicated and accessed in contexts other than the one we intended and imagined. The complications of identity performance and privacy expectations are disrupted and amplified to some degree when our interactions and disclosures are tracked, saved, and replicated in different contexts.

  Degrees of Visibility

  If privacy is not a binary and the concepts are blurry, how then do we manage and negotiate privacy norms online? One way to conceptualize online privacy is to shift the conversation away from private and public and instead to think about degrees of visibility. Research indicates that while young people (and adults) understand that social network sites are not entirely private, they still often interact on the sites as though they are private (Barnes 2006). This is in part because they have reasonable expectations that their quotidian interactions and conversations are not of interest to most people. Likewise, they assume that what they are sharing is contained and interpreted within a particular context. When online, individuals have a sense of whom they expect their audience to be on any given site, and act according to the norms of that particular audience and space.

  Problems arise, however, when contextual expectations are breached. The legal and communication scholar Helen Nissenbaum (2004) refers to this understanding of privacy as “contextual integrity.” She argues that within any context or sphere there are always two complementary and normative expectations at play: norms of appropriateness (what is appropriate to share in a specific context) and norms of flow, or distribution (what is appropriate to transfer or share from one party to another). Together these normative frameworks structure expectations and practices of contextual integrity. Nissenbaum argues that complaints of privacy violations are justifiable when norms of appropriate sharing or appropriate distribution are breached. The idea of contextual integrity is particularly productive when analyzing teens’ privacy strategies because teens and adult society often have different understandings and expectations of the social contexts in which young people disclose information, perform identities, and communicate with one another. Social norms are not only powerful strategies for influencing behaviors, but they are also contextually bound and constantly contested; for that reason, they are often open to multiple interpretations and understandings. If we are to understand how young people conceptualize online identity and negotiate expectations of privacy, we must first and foremost understand their social expectations and norms. As I’ll explain, norms vary between different social groups and are significantly influenced by identification, subject position, and localized practices. It is not as simple as “young people think of privacy as X”; rather, we must consider how different social groups create and police norms differently. Only then can we have a more nuanced conversation about the consequences of visible online identities and privacy expectations.

  Nissenbaum’s concept of contextual integrity is incredibly useful, though I think we can go a step further and consider expectations of visibility that also speak to the integrity of contextual specificity. In addition to competing understandings of context and norms between teens and adults, young people’s expectations of visibility are also increasingly at odds with the architecture of particular social media platforms. Sites such as Facebook frequently change how information is aggregated, publicized, and accessed within and beyond the site. Much like boyd’s (2008) analysis of the introduction of the news feed to Facebook, changes in the platform itself can breach users’ expectations of the context and visibility in which information is shared. As will be further addressed later in this chapter, this was the case when Facebook made Likes more visible. Participants didn’t expect their Likes to be private; however, they became upset when the context and visibility of their actions were altered and heightened.

  Youth desire, seek, and expect some control over the visibility of these interactions. What is concerning, particularly for low-income youth, is the extent to which their practices are scrutinized, pathologized, trivialized, and criminalized within popular imagination and the law (as discussed in chapter 2). The Internet opens up new opportunities for young people to participate in networked publics—that is, “to see and be seen, to socialize, and to feel as if they have the freedoms to explore a world beyond the heavily constrained one shaped by parents and school” (boyd 2014, pp. 201–202). However, harm-driven expectations that are perpetuated via increasingly negative media and adult attention will serve to further silence and marginalize teens’ voices and practices. Rather than risk scrutiny, they may choose to speak and engage in more insular and private spaces that are disconnected from the more public spaces and platforms in society. As was further discussed in chapter 5, low-income young people in particular are likely to miss out on opportunities for networking, promoting their creative work, voicing their experiences, and participating in networked publics. The reasons are complex, but we need to pay attention to the ways adult expectations, as well as the architecture and policies of social media platforms, monitor and regulate young people’s expectations of visibility and of what consequences.

  Networked Visibility

  I often hear both adults and teens repeat some iteration of “If you’re careful about what you share online, then you have nothing to worry about.” As has been noted, this attitude falsely constructs privacy as an individual decision and strategy. It neglects to take into consideration how our behaviors and interactions are networked and how the visibility of our social networks reveal a lot about us that we cannot control. Further, it reduces a complex issue—identity expression and speech—to an issue of mere individual responsibility, and thus dismisses the questions of ethics and boundaries altogether. What is absent from these discourses and narratives is the extent to which our online networks—that is, to whom we are connected—inadvertently reveal a lot of seemingly private information about us. Alarmingly, users of social media platforms have little knowledge of or control over what their
networks make visible and thus reveal about their identities, associations, and interactions, yet networks themselves generate, augment, and manage visibilities.

  To illustrate how networked identities mitigate visibility, I conduct an exercise each semester with undergraduate students enrolled in my Digital Media and Society course. I choose a friend from my own Facebook network, then view his or her friends. I take a screen grab of their friend list, capturing twelve random friends in their list (my friend’s image and name are not revealed). I then show the screen grab to my students and ask them what they can infer about my friend from twelve of their friends.3 At first students are a bit uncomfortable or reluctant, but with a little prodding they start articulating their assumptions. “Well, most everyone is black, so maybe the person is black.” Others will use names to make assumptions: “Well, it seems a lot of people have Spanish-sounding names, so maybe the person is from Mexico or an immigrant.” Others will take note of their age: “There’s a lot of pictures of kids and families, so maybe the person is in their thirties or forties.” I showed screenshots of twelve friends of one of my gay friends, and it did not take long for a student to accurately guess the friend was a white gay male in his thirties. Students have been able to accurately identify which friends are Muslim and which are Jewish. Others will identify where people work or where they received their education and take guesses about the person’s occupation and education level. One student noted that one of my friends had friends from all over the country and guessed that the friend might have moved around a lot. Sure enough, the friend was a professor who had lived in four states in seven years. I’ve had students accurately guess a friend’s industry—social work, news reporter, and teacher—on the basis of the friend’s connections. Interestingly, students even make assumptions about political affiliations and ideologies, even in the absence of overt clues. One student accurately commented: “All their friends are white, and kinda look alike. A lot of them have kids or pictures with their partner or spouse. I don’t know, it just makes me think she’s a conservative white woman.” And this assumption was eerily accurate.

  Students are amazed they are able to so accurately identify so much information about someone on the basis of only twelve photos from their public Facebook connections. Occasionally they will guess something inaccurate, for example, assuming a person is a college student when in fact the person is a 50-something professor. And occasionally they are not able to make many assumptions at all. However, the wrong assumptions are just as significant, maybe for some even more significant, than accurate guesses. Regardless of the confirmed accuracy, our list of visible connections serves to reveal aspects of our identities and networks. This is a matter of privacy that extends beyond the kinds of information we deliberately choose to reveal about ourselves online. These assumptions can be used to make decisions about us in a variety of contexts that are beyond our control and without our consent or even knowledge. But this is the nature of networked privacy: we can exert some control over the information we intentionally choose to share online, however, we have limited control over what our social networks reveal about us. Networked visibility makes evident that we cannot reduce privacy to that of individual responsibility, but rather privacy itself is networked.

  Similarly, we need to avoid couching privacy within a debate of “hiding” information. As Daniel Solove notes (2007, p. 764), constructing privacy in terms of hiding information “myopically views privacy as a form of concealment or secrecy” and ignores the complexity of privacy as a concept that is also about agency, transparency, control, context, and disclosure. My in-class exercise reveals the extent to which privacy is networked, but also the extent to which privacy far exceeds concerns of hiding “bad” information. Clearly, there is nothing wrong with being a Muslim, or pregnant, or gay, or black. However, these aspects of our identities can lead to blatant or unintentional forms of discrimination. Those who argue that we can and should be the only ones responsible for our own privacy speak from a highly privileged and protected subject position. The need for greater privacy protection is not merely about the ability for us to hide bad information about ourselves. It is imperative that privacy be understood within the context of a society that values, privileges, and protects some identities—or rather devalues, exploits, and discriminates against some identities—more than others. While we do have an individual responsibility to be careful what we share online, understandings of online privacy must evolve to encompass the public and networked nature of visibility, power, and privilege.

  Deliberate Disassociation

  What is important to note for marginalized youth, is the extent to which their localized and racialized identities, interests, and communicative practices—in other words, their cultural capital—is often misinterpreted, policed, and decontextualized within white hegemonic society (Carter 2003; Giroux 2009; Mendoza-Denton 2008). Because teens navigate different social contexts that include differentiations across age, class, and ethnic boundaries, it can be difficult for marginalized teens to simultaneously manage privacy and fit in with the “right” group. One way this is played out among marginalized young people is via a practice of disassociation; that is, some participants employ deliberate strategies to disassociate themselves from peers whose identities could lead to negative (adult) assumptions about their practices and preferences.

  For example, Miguel and his twin brother Marcus (14 years old, undocumented Mexican immigrants) explained that they were anxious about an informal reunion with friends from elementary school because “whenever me and my friend check them on Facebook, they’re all, like, ghetto and stuff, and I’m nowhere near that.” Countless times throughout the study students explained the ways they attempted to distinguish themselves from peers they perceived of as “ghetto,” a word used by virtually all participants—regardless of ethnicity—to primarily describe black peers and Mexican-Americans who were “too loud,” were “trashy,” or tried to “act hard” and “start fights.” These perceptions are embedded within the broader racialized and geographical context of Freeway High and go beyond their use of social media and are outside the scope of this chapter. Nonetheless, their anxieties about race, class, and reputation were heightened, articulated, and made visible in online spaces.

  Jack (17 years old, white), an upwardly mobile senior, explained that he blocked people who used “made-up names” on Facebook. While browsing his Facebook one afternoon, he pointed out a Facebook user with a made-up nickname. “See, there we go, Jamie Suckafree (notwitdabullshit)—that’s a ghetto name. People are starting to do it on Facebook. I don’t even know who that is. See, that’s why I blocked that person.” There is even an entire Tumblr page dedicated to “ghetto Facebook names.” As you can imagine, the page mocks not just the names but the users as well, and includes many derogatory assumptions based on the users’ ethnicities. As the trend of creating “ghetto” Facebook names became more popular during the time of our study, it was not uncommon for participants to explain they had blocked or unfriended such people. They often explained it was because the profiles were “annoying” or because they “didn’t know that person anymore.” However, as Marcus and Miguel also explained, they were anxious about being associated with such people. In a networked society, our relationships are made visible and reveal information about us that can be taken out of context and thus used against us. Some participants blocked and unfriended people with whom they did not want to be publicly associated. Their anxiety was rooted in a concern that adults and other peers would jump to conclusions or make assumptions about them on the basis of their peers. The legitimate concern served to police their relationships, as well as expectations and performances of (often racialized) identities. Although Miguel, Marcus, and Jack were concerned about how others might perceive such connections, and understandably so, there is nothing inherently wrong or risky about connecting with peers who express identities in these ways. The use of false names can also be read through a lens
of visibility strategies that made it more difficult to search and identify those teens’ profiles. But Miguel, Marcus, and Jack were aware of the extent to which others might make assumptions about them on the basis of these visible connections—assumptions they did not want others to make. At a time when more than three fourths of employers use digital media to screen and monitor employees (Webb 2014), strategies of deliberate disassociation and self-censorship might prove beneficial; the visibility of their connections and practices open up opportunities for discrimination based on misunderstandings of cultural and social capital.