Much has been written about the volume of personal data collected by social media platforms (Werbin, 2011; Fuchs, 2012; Bollmer, 2016) and its use by both governments and private industry (Fuchs, 2011, 2012). The collection of data on a medium’s audience, and its implication in capitalist production have been discussed for decades (Fuchs, 2012; Meehan, 2002, 2005; Smythe, 1977, 1981). But the sheer volume and particularity of data collected, and the way they are used automatically, instantaneously, and invisibly by algorithmic processes is something new (Bollmer, 2016; Fuchs, 2012; Terranova, 2015). This volume of data remains after the death of a user and continues to be monetized both directly, in the case of a deceased person’s digital image being used to spur further engagement; and indirectly, as the data continue to be accessed by the algorithms that govern the platform(s) the deceased had used.
The following chapter will discuss the ways that the data of the living is collected and stored by social media platforms, how that information continues to be used to profit the platform after a user dies, and the implications of these developments. It will focus largely on Facebook, as the most populous and active of the major social networks (Elliot, 2015; Pew Internet Research, 2017), but my intention is to suggest qualities that exist in this type of communicative technology, not the particular flavor or brand. Using Smythe’s concept of the “audience commodity” as a guide, this inquiry will also engage with modern work that has built on that approach. The latter part of the chapter will use the concept of “immaterial labor” (Lazzarato, 1997) as another approach to the “work” of social media users and as a way to imagine the continued labor of users after death. The chapter will engage with autonomous Marxist theory about labor. Finally, this chapter will close by briefly discussing a related phenomenon, that of the production and viral sharing of videos of Black Americans being killed by the police. These videos are shared by activists to uncover police violence but, paradoxically, the videos are caught up in the same processes that monetize any other user generated content online.
The Audience Commodity in the Age of Social Media
The work of Dallas Smythe on the “audience commodity” (1981) put forward the concept of the audience of a medium (Smythe was largely concerned with broadcast television, but touched on radio and newspapers as well) giving up their time in the service of conglomerate profits. The audience’s time becomes the commodity, their “audience power” the product for entertainment companies to sell to advertisers. “Because audience power is produced, sold, purchased and consumed, it commands a price and is a commodity” (Smythe, 1981/2012, p. 187). This time is then packaged into demographic segments using the increasingly complex TV ratings and viewership data-collection apparatus employed by ratings agencies such as A.C. Nielsen (Smythe, 2012/1981). As Smythe puts it: “Scientific sampling yields results as reliable for audiences as it does for grain, sugar” (Smythe, 2012/1981, p 189). As technologies and methods for sorting and quantifying audiences have advanced, new products have come to the market, purporting to offer more specific, granular audience classification for advertisers targeting television viewers (Meehan, 2005). This process of quantifying and sorting audiences is the precursor to the data collection done on social media users today.
Smythe has been criticized for accepting the accuracy of this type of ratings data. Meehan, for example, has critiqued Nielsen’s flawed methodology and Smythe’s apparent acceptance of it as “scientific” (2002, 2005). This critique has little to do with Smythe’s major theoretical move, however, which brings communication into the productive base by focusing on the economic relations between the audience and the media industries, instead of the ideological content being communicated through those media. Smythe saw this as the “blind spot” in critical approaches to the media (Smythe, 1977), which had been much more concerned with the “‘messages,’ ‘information,’ ‘images,’” etc. in mediated communication that he saw as “superficial,” and “divorced from real life processes” (Smythe, 2012/1981 p. 187).
Smythe’s characterization of the work of the audience as labor in the production of a commodity has been returned to by scholars who study broadcast media, and more recently by those researching digital platforms and their users (Caraway, 2011; Dolber, 2016; Fuchs, 2012). This commodity, audience power, can be sold to advertisers and the state, and the information gathered is used to spur those same audience members into further labor, audience or otherwise, and into the consumption of certain goods over others (Smythe, 2012/1981). Decades after Smythe advanced the concept, the collection of a medium’s user data and its sale to advertisers operates somewhat similarly, but in an exponentially more complex and automated fashion (Dolber, 2016; Fuchs, 2012). The great difference is that it is no longer the users’ time that is valuable, it’s all the rich data made available about that user, and about user demographics in aggregate, that advertisers seek. The web 2.0, and later social media revolutions have allowed for a much deeper and more invasive collection of user data (Fuchs, 2012; Terranova, 2015). And those users, as human beings, tend to die.
When approaching the political economy of death on social media in a mode inspired by Smythe, the focus must be on the processes that create value from user activity, and how it may continue to be used for corporate profit. What kind of user data is collected, what is stored after death, and how does it continue to be used to profit the platforms that collect it? The content of all the communicative acts that occur during and after any user’s life are much less important, akin to the “free lunch” of television programming (Smythe, 1981, p. 194). This modern free lunch may not be the carefully produced content that Smythe spoke of, made to “whet the prospective audience members’ appetites and thus (1) attract and keep them attending to the program, newspaper, or magazine; (2) cultivate a mood conducive to favorable reaction to the advertisers’ explicit and implicit messages” (Smythe, 1981, p. 194). Under modern communicative capitalism, the “free lunch” that keeps users returning is the constant communication and connection allowed by social media platforms. Jodi Dean describes the draw of networked digital platforms as the “form and vehicle for the individualized consumption, participation, and creative needs expression of subjects compelled to be personally satisfied” (2010, p. 135). After a user dies, their content and its continued presence in these platforms can serve as a free lunch as well, drawing engagement from living users (Bollmer, 2013; Klastrup, 2015; Rossetto, Lannutti, & Strauman, 2015).
It doesn’t require a lot of digging to see how much economic power these major social media platforms wield. The companies’ publicity materials are quick to point out just how large and lucrative these platforms have become. Facebook’s most recent investor report showed the company’s net income surpassing $10 billion in 2016, and provided this list that underlines the platform’s staggering user engagement:
- Worldwide, there are over 1.86 billion monthly active Facebook users (Facebook MAUs) which is a 17 percent increase year over year.
- There are 1.15 billion mobile daily active users (Mobile DAU) for December 2016, an increase of 23 percent year-over-year.
- 23 billion people log onto Facebook daily active users (Facebook DAU) for. September 2016, which represents a 18% increase year over year.
- There are 1.74 billion mobile active users (Mobile Facebook MAU) for December 2016 which is an increase of 21% year-over-year
- On average, the Like and Share Buttons are viewed across almost 10 million websites daily. (Facebook, 2017).
Table 2. Facebook User Statistics.
Facebook’s marketing materials for advertisers provides a fairly deep description of the kind of analytics the company offers to advertisers (Facebook, n.d.), and a description of its data collection processes is used to lure investors in its newsletter (Facebook, 2017). To understand just how deeply social media is insinuated into human life, the description of the types of data that these platforms collect and store can be illuminating. Facebook’s many different APIs (Application Programming Interface) provide software that advertisers may use to track user activity and target ads (Facebook, n.d.). These APIs offer powerful big data analytics that can be simply accessed using a command line interface. For example, the targeting feature allows advertisers to, “Specify targeting options for your ad set in four areas: interests, demographics, behaviors and locations” (Facebook, n.d., para. 2). But within those categories are data points as fine as neighborhoods-frequented, type of device used, and how much time is spent on Facebook. (Facebook for Developers, 2017). The platform also offers what it calls the “Audience Network”, a service that “lets you extend your ad campaigns beyond Facebook to reach your audiences on mobile apps, mobile websites and videos,” and its marketing notes that “We use the same Facebook targeting, measurement and delivery to make sure each ad on Audience Network helps you reach your campaign goals” (Facebook Business, 2017, para. 3). The service allows Facebook’s collected data to be leveraged by companies to target users even when they aren’t on the platform.
A deeper analysis of these data collection processes is certainly called for. But it’s hardly controversial to say that the users of social media platforms are the raw material that the industry uses to produce value. These statistics clearly show a data collection and advertising infrastructure that’s significantly more sophisticated than the audience sampling of Smythe’s time. These Facebook stats also show an incredible level of user engagement with that platform, and a massive and constant sharing of personal data.
These same users will inevitably die, and all of the intricate data that was collected on each of them will continue to produce for the massive corporations that own their chosen social media platforms. The body releases its contents as it decomposes, in one final contribution to the life processes of the world. Likewise, deceased users slough off a final volume of information into waiting data centers, and the instance of a death creates a new category of data to be collected from the friends and family who use the same platform to mourn them. (Bollmer, 2013; Klastrup, 2015; Rossetto, Lannutti, & Strauman, 2015).
Control of Personal Data
Upon a death, a user’s surviving loved ones do exercise some control, with a few options for the social media content of the deceased. As mentioned in Chapter 3 (p. 39), Facebook instituted two options in 2009 that addressed these concerns: the “memorialization” of a person’s profile effectively making it private, only available to those already connected to it and sealed from further changes (Kern, 2013; Marwick & Ellison, 2012); or the option of an “R.I.P. Page” that functions more like a brand’s page, and is open to the public (Klastrup, 2015). Effectively, then, there are 3 ways that a person can endure on Facebook after they die: as a normal profile page, still able to “interact” with others; as a memorial profile, sealed off from Facebook’s algorithmic processes and only visible to those who were already connected to it; and as an R.I.P. page that is created by others in memory of the deceased, though still likely using many of that dead user’s shared assets such as photos and quotations (Klastrup, 2015).
As of February 2015, Facebook allows individual users an aspect of control over their own profile should they pass away. The “Legacy Content” menu is buried deep in the privacy settings menu, not prominently displayed. Facebook may not want to advertise the settings, but they do give users something akin to a last will and testament with regards to their Facebook presence after they’ve died. The user can choose a “Legacy Contact” to manage their profile after their passing, a person able to “do things like pin a post on your Timeline, respond to new friend requests, and update your profile picture. They won’t post as you or see your messages” (Facebook, n.d.a). This contact will become the keeper of the memorialized page, a responsibility that only becomes greater the more personal data is available on the user’s profile. Facebook will continue to store all the collected data and communications on the dead user. The company withholds the right to change the terms, noting, “we may add additional capabilities for legacy contacts in the future” (Facebook, n.d.a, para. 5). The other option is to have the account deleted automatically “If you don’t want a Facebook account after you pass away, you can request to have your account permanently deleted instead of choosing a legacy contact” (Facebook, n.d.a). Of course, if the user has not made any selections here, their loved ones are left with the same options mentioned above. Research into death on Facebook has shown that many profiles remain active, and continue to interact with other users (Elliot, 2015; Kern, 2013; Marwick & Ellison, 2012). There may not be living users to add updated content to these active accounts, but their profiles remain available for many of the automated processes described above. (Meisler, 2012).
Some other social media platforms have options to give some control over user to the deceased’s family. Twitter will allow “verified” family members to request the deactivation of a user’s account and to archive its tweets, but the family member must provide “information about the deceased, a copy of your ID, and a copy of the deceased’s death certificate” (Twitter, 2015, para. 2). Professional networking site LinkedIn requires similar documents as well as a link to an obituary and an e-signature (LinkedIn, n.d.). However, some platforms still have no clear provisions for dead user’s accounts and content. SnapChat, for example, has no stated policy for dead user’s accounts. The multimedia messaging app is known for the ephemeral nature of its communication, and its privacy policy claims that user content is deleted from its servers (Snap Inc. 2017). But the policy also notes, “the rest of our services may use content for longer periods of time, which means those services may follow different deletion protocols,” and “we may retain the content as long as necessary to offer and improve the services” (Snap Inc. 2017, subheading: “How Long We Keep Your Content”).
In the most recent accessible year, 2014, the CDC reports 2,626,418 deaths in the United States (Kochanek, Xu, Murphy, Minino, & Kung, 2016, p. 1). Of course, many of those people may not have been social media users; unsurprisingly, the major causes of death were those that inflict older people toward the “natural” end of life. The same study lists the top 15 causes of death. Of those listed only one, suicide as the 10th most common, is a major cause of death for young people (Kochanek, et al., 2016, p.1). In 2014, less than 27 percent of adults over 65 had any kind of social media account, as opposed to over 84 percent of those aged 18 to 29 (Pew, 2017, fig. 4). Still, the data suggests that hundreds of thousands of younger Americans, likely social media users, will die every year. And the usage of social media by older Americans who are more likely to die is rising as well, already at 35 percent by 2016 (Pew, 2017, fig. 4). As of 2016, 62 percent of online adults 65 and over are on Facebook (Greenwood, Perrin, & Duggan, 2016, p. 4).
Worldwide, it is estimated that 8,000 Facebook users die every day (Elliot, 2015, p. 381). One marketing company’s publicity stunt claims to have used mathematics to predict 2065 as the year there will be more dead than living profiles on Facebook (Hiscock, 2014). The same “study” also estimated 428 Facebook users die every day and 312,500 every month (Hiscock, 2014). Dubious methodology and the implicit assumption that Facebook will endure aside, this marketing company wouldn’t have pursued the answers to these questions were there not an interest in how to approach the phenomenon coming from the businesses that use its marketing services.
My point with all these statistics is this: the number of social media users dying will likely continue to increase, and many of the social media profiles of those dead users will continue to “remain in perpetuity” (Kern, Forman, & Gil-Egui, 2013), neither deleted nor sequestered from the algorithmic processes that profit their respective social media platforms. The profiles and data of these users will continue to labor after the user’s death. They will drive “engagement,” as virtual spaces for other users to spend their time, and thus continue to offer their usage data to the platform, and be served ads. This seems to me a very compelling “free lunch” (Smythe, 1981, p. 194) to draw in living users whose time on the platform is valuable.
The data collected on the dead users also will continue to be used in creating new targeting analytics for advertising as well, still “rich venues for mining and analyzing social media data” (Leskovec, 2011, p. 277). Digital technology has allowed the work of the audience to continue producing its commodity from beyond the grave. Even after the productive capacity of a dead user declines, when other users stop interacting with the dead profile and the dead user’s unrefreshed data no longer triggers algorithmic advertising, that user’s aggregate data will forever be held by the platform. Each new dataset and product that a social media platform creates to sell to advertisers will contain the still-valuable data of the dead alongside that of the living.
The Immaterial Labor of the Dead
As the section above has made clear, the users of social media have taken on a much more active and productive role than Smythe’s commodity audience did in their time. The nearly 2 billion worldwide users of Facebook (Facebook, 2017) actively engage with the platform, commenting and sharing, creating memes perhaps, or taking polls; each action creating more data to be used to better target them for advertising, to further their engagement, or to further the engagement of those connected to them. These users are engaging in the production of wealth through their “immaterial labor” (Lazzarato, 1997) for these social media platforms. The second section of this chapter discusses labor in what many have described as a new configuration or type of capitalism brought on by information technology (Hardt, 1999; Lazzarato, 1997; Moulier-Boutang, 2012; Terranova, 2000). To simplify: the labor of many does not happen in physical work, but in work that involves communication and/or creativity, and produces intangible goods or the affective conditions that allow for commerce (Hardt, 1999; Lazzarato, 1997). This kind of immaterial labor includes the “kinds of activities involved in defining and fixing cultural and artistic standards, fashions, tastes, consumer norms, and, more strategically, public opinion” (Lazzarato, 1997, p. 132). Much of the very broad literature that touches on these concepts is concerned with creative professionals, information workers, customer service providers, and others for whom their vocation necessitates communication over material production. Countering the rampant tech utopianism of the early web era, these critiques sought to point out that the “relative abundance of cultural/technical/affective production on the Net, then, does not exist as a free-floating postindustrial utopia but in full, mutually constituting interaction with late capitalism” (Terranova, 2000, 43).
With the advent of the world-wide web, and in particular the “web 2.0” explosion of platforms that allowed for users to easily generate their own content (Andrevic, 2013; Dean, 2005; Terranova, 2000, 2004), many scholars began to increasingly focus this approach to production on the “voluntary” participation that web users were devoting to their unpaid labor online. To Terranova, “the Internet is animated by cultural and technical labor through and through, a continuous production of value that is completely immanent to the flows of the network society at large” (2000, p. 34). This “free labor” is “simultaneously voluntarily given and unwaged, enjoyed and exploited, (Terranova, 2000, p. 33). Considering the above-mentioned value of our engagement with social media, and the amount of time many put into it, a consideration of the immaterial production of social media users is timely. Much recent work has begun to examine this phenomenon using the concept of immaterial labor, such as Pybus’ discussion of the immaterial labor of “tweens” online (2011), or Jin and Feenberg’s work on the monetization of social media content (2015).
Terranova’s free labor and Lazzarato’s immaterial labor both draw from the worker-centric theory tradition known as autonomism which has its roots in the Italian operaismo (literally, “workerism”) movement, and a resulting move in European Marxist thought to focus on workers and their lives over hierarchies and structures in capitalism (Terranova, 2000, Hardt, 1999, Cuninghame, 2010). A key component of this move was the concept of the “social factory” insinuated into the everyday lives of the working class, where “work processes have shifted from the factory to society, thereby setting in motion a truly complex machine.” (Negri, 1992, in Terranova, 2000. p. 133). In the social factory, the creation of value is intermixed with social processes and much of it is unpaid. Fuchs ventures into this territory in his modern revisiting of Smythe: “the factory is not limited to the space of wage labour, but extends into everyday life” (2012, p. 716). He connects Smythe’s audience commodity to immaterial labor by returning to one of the founders of operaismo.
At the highest level of capitalist development social relations become moments of the relations of production, and the whole society becomes an articulation of production. In short, all of society lives as a function of the factory and the factory extends its exclusive domination over all of society. (Mario Tronti, 1962, in Fuchs, 2012, p. 719)
This broad body of theory, seeks to describe modern capitalism and capitalist production in Marxian terms, but to also note a change in the system that humanity labors under. Like Smythe, these theorists return to Marx, reinterpreting his centuries-old writings and applying them to the present. But Smythe sought to append modern processes to Marxist theory, while many theorists coming from the Autonomist Marxist tradition see a new mode of capitalism following the industrial, “Fordist” factory age. Beyond simply post-Fordism, this “cognitive capitalism” is a new configuration of capitalist production with new configurations of labor and value, as well as new possibilities for workers’ resistance. (Moulier-Boutang, 2012). In Cognitive Capitalism (2012), Moulier-Boutang attempts to describe this new mode of capitalism and the role of workers within it.
The mode of production of cognitive capitalism, if we want to give a description that is concrete but sufficiently general to cover all of all of its various situations (the production of material goods, services, signs, and symbols), is based on the cooperative labour of human brains joined together in networks by means of computers. The very rapid development of organisational forms such as project management, arrangement of small units articulated into networks and operating under outsourced relations of subcontracting, partnerships and logically based relationships is the public manifestation of this transformation. (Moulier-Boutang, 2012, p. 57).
Moulier-Boutang returns to Marx to theorize this new capitalist mode, seeking new insights from Marx’s compiled notebooks, The Grundrisse, “the autonomists’ favorite text,” according to Terranova (2000, p. 44). Like autonomists before him, he draws from a passage known as “The fragment on machines” (1971), in which Marx describes a changing process of laboring brought on by the “fixed capital” of machines. The passage describes processes whereby machine aided production allows for “general social knowledge” to become a “direct force of production” bringing labor into the “real life process” (Marx, 1971, 706). Moulier-Boutang describes this epoch of capitalism, pointing out “this important fact of a living activity that co-produces labour as living activity” (2012, p. 54). This productive configuration of the living labor of human workers and the technological configuration of machines, or “fixed capital,” is what Marx referred to as the “general intellect.”
Nature builds no machines, no locomotives, railways, electric telegraphs, self-acting mules etc. These are products of human industry; natural material transformed into organs of the human will over nature, or of human participation in nature. They are organs of the human brain, created by the human hand; the power of knowledge, objectified. The development of fixed capital indicates to what degree general social knowledge has become a direct force of production, and to what degree, hence, the conditions of the process of social life itself have come under the control of the general intellect and been transformed in accordance with it; to what degree the powers of social production have been produced, not only in the form of knowledge, but also as immediate organs of social practice, of the real life process. (Marx, Grundrisse, 1971, p. 706, emphasis in original)
This body of theory, then, is concerned with life processes as labor under this new stage of capitalism. It is “the development of the social individual which appears as the great foundation-stone of production and of wealth” (Marx, Grundrisse, 1971, p. 705). Though Marx, and even many modern theorists couldn’t have predicted just how intertwined individuals’ social actions have become with the production of wealth in this era of mediated connection. Much of this work as well as studies of social media use (Pybus, 2011; Waite & Bourke, 2013) would suggest that those life processes are playing themselves out in a virtual, mediated space as much as they are in the “meatspace” that living bodies physically exist in. Though a casual observation could probably reveal as much. The sheer number of Facebook daily users mentioned above speaks to the powerful wealth creation that these processes entail. In a more recent paper, Terranova referred to this modern communicative paradigm as “[t]his direct integration of the social relation into a market-oriented economy” (Terranova, 2015, p. 114). It’s important to point out, however, that this production continues whether or not the living labor of human beings is actively engaged with it. These platforms continue to create value from the data collection mentioned above. In his discussion of living labor on Facebook, Horning describes the process: “Life experience (self-production, self-expression, relationship forming, etc.) is being transformed through social media into something abstract and redeployable behind the scenes by capitalist firms: data.”
The immaterial labor (Lazzarato, 1997) poured into social media—the crafting of digital artifacts, writing, sharing, time, meme-making etc.—largely remains after the death of a user, marking an important shift. Before this modern media technology existed the intangible products of the common worker would be exhausted, much like the physical labor value, at death. With the archiving and the algorithmic sharing that social media platforms allow this is no longer the case. The toil of the living in the social factory becomes “fixed capital” in a way, as their data is stored for later retrieval. That data can be used in the social processes playing out over the web, interfacing with living people and machines without being tied to a living person. So, what, if anything does it mean? As the flows of capital become increasingly untethered from anything material or concrete (Terranova, 2000; Moulier-Boutang, 2012), is there a productive unit left behind after death that can continue contributing to this new era of capitalism? If so, does it ever expire? What are the implications for living workers who may be working in a digital graveyard, sharing memes with the data corpses of their friends?
Over time, the value that these profiles can generate necessarily diminishes without new data being added by a living users. That “symbolic immortality… is likely to fade over time” (Sherlock, 2013), leaving a decreasingly effective volume of data for driving engagement. But these technologies will continue to advance, and the collection of personal data will only become deeper and more particular. The corporations behind the social platforms we use will continue to think of strategies and technologies toward the mobilization of that data for profit, including the remaining data of the dead. A continued critical engagement with these media is necessary, at the level of the communicated symbols traveling across them, but also at the level of the economic relations between the corporations that own these media and their users (as I’ve attempted to do here). The new phenomenon of users’ deaths on social media offers an entry point to question just how much users’ lives are freely given to these platforms.
Final Thoughts: Viral Videos of Black Death
The tragic phenomenon of the viral videos that spawned Black Lives Matter (Roos, 2014) has created a unique category to be viewed through the labor lens of this chapter. Thus, I’d close with a suggestion of an area ripe for further research: a discussion of the labor of the dead in the terms of more concrete consumables, the much-shared videos of the death of Black people in police custody. As shareable content, these videos are buttressed by ads and enmeshed in social media analytics. Each iteration of death, then, is a tiny transaction that was never consented to by the deceased. The producers of the videos, who are often the grieving loved ones of those killed, are never compensated for the videos they’ve uploaded, and they’ve relinquished control over the image of their loved one’s final moments as raw material for the communicative processes of social media.
While having many of the same implications inherent to user-generated-content that I’ve mentioned above, the fact remains that the production of the images of these deaths—not to mention the consumption—occurs in and reinforces a racialized distinction between “who may live and who must die” (Mbembe, 2003, p. 11). Mbembe’s concept of necropolitics (2003) and its inverted relationship to Foucauldian biopolitics, describes the control of death as a way to control the lives of people in marginalized populations. These videos of Black Americans deaths are shared to push back against that control, and the “necropower” of the police. But they are, paradoxically, creating a growing body of archived Black death online. The repetition of these deaths online may further contribute to a normalizing of such images and an association of Black bodies with police violence, and black death as inevitable (Harriot, 2017).
This archive will also continue to profit YouTube and the platforms where the videos are shared. The video of Eric Garner’s death at the hands of the NYPD, for example, has been viewed over 300,000 times (New York Daily News, 2014). That’s not considering the myriad reposts, re-edits, and re-broadcasts of the moment of Garner’s death. To suggest that a purely economic analysis of this phenomenon is needed would be ridiculous. But there are openings to consider these videos alongside a structural conversation around mediated death and labor. An analysis of the amount of ad revenue made by YouTube (thus, Google/Alphabet) for example, could highlight the way that even the struggle for justice is subsumed under cognitive capitalism and the racism of the state.
As the data of the dead continues to labor, something of the identity of the dead endures as well. This mediated persistence is only one way that these still-very-new digital technologies are changing the way that life and death are experienced and understood. The conclusion will discuss the way these changes may subtly push against the way humanity conceives of itself.
Wade Keye 