523This is a draft of a manuscript forthcoming in: S. Walz and S. Deterding (Eds.), The GamefulWorld. Boston MA: MIT Press.13Playful Heterotopias or Technologies of Control? Foucault,Governance, and GamificationJennifer R. WhitsonIn this chapter, I draw from Michel Foucault to frame self-tracking and gamification in terms ofthe governance of modern liberal nation-states where subjects willingly govern, regulate, andoptimize themselves. I introduce the quantification of the self, showing how it is used ingamification movements and how it is leveraged to promote a care of the self, as well as furtherenrolling individuals in normalization projects. I argue that current gamification projects are notinfluenced by playful design (and much less a focus on fostering creativity and exploration), buttake something entirely different from games: the feedback mechanisms such as leaderboards,damage meters, and point systems that allow users to manage risk as well as pinpoint “approved”routes toward mastery and self-improvement. I conclude with some cautionary thoughts aboutthe difficulty uniting play with non-game governance projects, given that play inherentlyencourages players to push against, reshape, and find movement between rules, sometimesbreaking these rules altogether.

524IntroductionLuka is five years old. He has been “quantifying” since he was two and a half. First, he startedcollecting his daily “numbers” on a bathroom scale that records his weight, calculates his bodyfat, and uploads these data to a website charting his growth. Then, when Luka was four, he wasgiven a Fitbit, a wireless activity and sleep tracker that counts each step he takes during the day.While Fitbit itself promises to make fitness fun by awarding badges for distance traveled,calories burned, and stairs climbed, Luka and his dad have invented many other games to “getmore steps,” including—if Luka notices he hasn’t beaten yesterday’s total—racing around thehouse like an Olympic sprinter before bedtime. One of Luka’s favorite games is competing withBruce from his dad’s work, who is also connected to Fitbit’s cloud service. On most days Bruce,a sedentary soul, shows up on the family’s leaderboard losing to Luka. Winning against a realgrownup, let alone one from his dad’s work, is pure joy (Carmichael 2012).Luka, or more accurately Luka’s dad, is a member of the quantified-self (QS) movement(Schuller 2012). As I argue in this chapter, QS is closely tied with gamification. By gamificationI mean “the use of game design elements in non-game contexts” (Deterding et al. 2011). Bothemphasize self-improvement through tracking, both are tools of self-governance, and bothcharacterize a new care of the self based on data. Not surprisingly, tools for quantifying the selfare also examples of gamification, including Nike , Mint, Runkeeper, Health Month, fitocracy,Daily Burn,, and, of course, Fitbit. While gamified apps often fail to deliver theplayful and game-like spaces they promise in their marketing rhetoric (for example, see JaakkoStenros’s chapter in this volume on the missing playfulness in gamified applications, and seeRalph Borland’s chapter in this volume on how play is evoked in marketing the PlayPump), what

525they excel at is providing meaningful feedback to users that is then enrolled in motivatingbehavioral change (see both C. Scott Rigby and Conor Linehan and colleagues’ chapters in thisvolume for more on this). Mastering the self through the application of gamified data is animportant method of governance. It is this relationship between gamification, quantification, andgovernance that I explore in this chapter.Part 1: Governance, Surveillance, and the Care of the SelfIn part 1 of this chapter, I draw from the work of Michel Foucault to first define what I mean bygovernance. I then argue that governance—and knowing the desires of those who aregoverned—is reliant upon surveillance. I briefly trace the historical relationship betweensurveillance and governance, from discipline in the eighteenth century to more modern modes ofcontrol focused on consumption and desire. Here, I begin to make links to commercialgamification products, such as the Fitbit, and how they enroll our desires for self-mastery andimprovement into a new care of the self, a care of the self that is also predicated upongovernance.Defining GovernanceWhen I talk about governance, I am referring to something broader than the voting practices anddemocratic institutions of the state that are the focus of Greg Lastowka and ConstanceSteinkuehler’s chapter in this volume. I am referring to government as the “conduct of conduct.”In other words, government is not just a state domain but includes “[A]ll endeavours to shape,guide, direct the conduct of others, whether these be the crew of a ship, the members of ahousehold, the employees of a boss, the children of a family or the inhabitants of a territory. And

526it also embraces the ways in which one might be urged and educated to bridle one's ownpassions, to control one's own instincts, to govern oneself” (Rose 1999, 3).I am interested in how gamification is tied to the forms of neoliberal governance that arefocused on the privatization and deregulation of the state, while simultaneously inducing citizensand corporations to regulate and govern themselves. Specifically, in this chapter I want toexamine how gamification is used to encourage citizens to govern themselves, in terms of takingincreased responsibility for their health care, education, and workplace productivity, as well ashow they are encouraged to become more loyal consumers and clients. For example, how doesgamification ensure that a five-year-old boy like Luka is aware his body-fat index and itsrelationship to his long-term health (and thus his potential burden on the health care system)? Inthese cases, the governors are not only state agencies, but also educators, employers,corporations, and even individuals such as Luka’s dad. The governed are people like Luka—aswell as you and I—the individual users of gamified applications.Governance in the Foucauldian sense is productive. It is opposed to domination, whereinsubjects have no other option but to obey. As put by Nikolas Rose, knowledge of those to begoverned is key to this productivity:[T]o govern is to recognize that capacity for action and to adjust oneself to it. Togovern is to act upon action. This entails trying to understand what mobilizes thedomains or entities to be governed: to govern one must act upon these forces,instrumentalize them in order to shape actions, processes and outcomes in desireddirections. Hence, when it comes to governing human beings, to govern is topresuppose the freedom of the governed. To govern humans is not to crush their

527capacity to act, but to acknowledge it and to utilize it for one's own objectives.(Rose 1999, 4)In other words, governance is about knowing subjects and their motivations and desires wellenough to determine how to get them freely and willingly to enroll in the governor’s projects,and thus govern more effectively. This entails a much different conceptualization of power thandomination through force—the governors are not focused on punishing the governed, butrecruiting them as willing participants. In this sense, power is not a thing, but a relationshipbetween people in which one affects another’s actions. It is productive, rather than violent orrepressive. It involves making a free subject do something he or she would not have doneotherwise. Power is not just localized in the state and other authorities, but is present in allrelationships.Governance is a thus a process of translation, forging alignments between the objectivesof authorities wishing to govern and the personal projects of those organizations, groups, andindividuals who are subjects of government. Not surprisingly, there is a lot of interest in usinggamification as a technology of government that shapes users’ conduct in the hope of producingcertain desired effects (such as using gamification to increase productivity in call centers) andaverting certain undesired events (such as using gamification to reduce employee churn andabsenteeism).Surveillance and the PanopticonDeveloping the knowledge required to discern the desires of subjects and govern themaccordingly depends upon surveillance, making visible the space over which government isexercised: “defining boundaries, rendering that within them visible, assembling information

528about that which is included and devising techniques to mobilize the forces and entities thusrevealed” (Rose 1999, 33). The role of surveillance in governance is a central theme ofFoucault’s earlier work (Foucault 1977). Discipline and Punish traces the history of governmentfrom pastoral to feudal to near modern times, asking: How do we go from an unruly,undifferentiated mass of people, to the orderly, productive, collection that we see today?Foucault argues that in order for society to thus organize itself, the key is render eachindividual visible, to separate them out, to closely observe, and then compare them to each other.The way this is achieved differs according to the point in history and the techniques available.For example, Foucault used Jeremy Bentham’s blueprint of the panopticon as a vivid example ofdisciplinary power. A prison oriented around a central guard tower, individual prison cells in thepanopticon create a ring around this central tower. At all times, prisoners are exposed to the gazeof those within the tower, though prisoners cannot see the other inmates, nor tell whether theguard tower is occupied or not. In this case, the simple fact that one may be observed is central toevoking socially accepted behaviors.As put by Bart Simon, the power of the panopticon is twofold:On the one hand, there is a concern with processes of subjection andnormalization that arise through the internalization of the gaze, while on the otherthere is a concern with processes of administration, social sorting and simulationthat occur independently of embodied subjects. (Simon 2005, 1)This administrative power of the panopticon quickly diffused to other spaces, such as schools,hospitals, factory floors, and military service, and initiated the birth of the record system (IDsystems, police file systems, medical records, and academic grade systems), as well as written

529systems for identifying individuals and—most importantly—tracking them over time. The abilityto differentiate subjects and to track their performance over time further incites a desire on thepart of the subjects to normalize, to fit in. By normalize, I mean conforming to the idealizednorm of conduct. For example, Foucault uses the example of military drills, where each soldier istaught precisely how to stand, march, present arms, and so forth. Soldiers are then rewarded orpunished for conforming to or deviating from this ideal. Today, normalization can be seen inbody-fat index ratios, as well as scores, ranks, and grades.This two-part disciplinary power—part focused on administering populations, and partfocused on self-governance—fought the chaos of previous, more violent, forms of governmentby using surveillance to order individuals and, by doing so, impose efficiency and productivity.In the popular imaginary, however, the panopticon is equated to domination in the form ofGeorge Orwell’s 1984 and the oversight of Big Brother.Beyond the Panopticon: Games as GovernmentIn the face of neoliberalism, the panopticon crumbles. At heart, disciplinary panopticism relieson individuals who want to become ideal citizens, part of a civilized polity who would governthemselves “through introspection, foresight, calculation, judgement and according to certainethical norms . . . the social objective of the good citizen would be fused with the personalaspiration for civilized life” (Rose 1999, 78). The overarching impetus to become an idealcitizen, however, dissolves as people focus more and more on individualized goals andaspirations. Meanwhile, the bastions of this disciplinary governance—the church, the factory, thestate—disappear in the face of deregulation, replaced by new domains, more agile models ofproduction, and a de-emphasis of the collective social body. As put by Nikolas Rose, “Today,

530perhaps, the problem is not so much the governability of society as the governability of thepassions of self-identified individual and collectivities: individuals and pluralities shaped not bythe citizen-forming devices of church, school and public broadcasting, but by commercialconsumption regimes and the politics of lifestyle, the individual identified by allegiance with oneof a plurality of cultural communities” (Rose 1999, 46). Accordingly, theorists such as GillesDeleuze have extended Foucault’s work to consider how governance is enacted in spacespremised on automation, dividuation,1 and consumption. In these spaces, control operates on amore free-floating, adaptive basis that is rooted in desire rather than social conformity. In short,consumer society has replaced civil society, and thus modes of governance adapt.In a consumer society, surveillance shifts from tracking individuals to monitoringbehavior and consumption patterns. Populations are constituted as consumers to be seduced intothe market economy. This monitoring is predicated upon limiting access to places andinformation and developing ever more intimate consumer profiles. Power is enacted throughreconstructions of consumers’ behavior, habits, and actions, knowledge that enables moreeffective governance. Accordingly, in the society of control there is a movement away fromhuman watchers and their associated value judgments, and a movement toward seeingindividuals only as bits and bytes in vast ebbs and flows of information. By tracing theaggregated desires of shoppers and system users, finding patterns in their flocking behaviors likea school of fish, and then channeling these behaviors, organizations thus enact governance,knowing subjects and their motivations and desires well enough to determine how to get them tofreely and willingly enroll in their projects.Whereas the panopticon depends on the individuals, first individuating them in order toform them into a more productive social body—like cogs in a vast machine—Deleuze’s (1992)

531society of control is automated by technology. Rather than relying on prison guards and drillsergeants to discipline individuals, control operates according to machinic demands and is relianton codes and passwords, on data and databases. The body is transformed into pure information—the data double—so it can be rendered more mobile and comparable. Information, such asshopping habits, user preferences, bank account numbers, voting preference, location, and soforth, is separated from individuals and recombined in new ways outside of their control. Theserecombinations, such as user and customer profiles, are based on the criteria deemed salient bythose with access to the information, be they government officials or corporate marketers.Instead of individuals—irreducible and with an autonomous sense of agency—the newsubject of governance is instead the dividual, an artifact of data mining searches and computerprofiles. People, each as individual wholes, are unimportant. What is important are how massesof people can be broken down into more manageable parts by collecting the data streams theytrail behind them, filtering out the parts deemed important, and ignoring the rest. Dividuals arethen governed automatically through databases and levels of access and exclusion. For example,in banking transactions, your name and identity are entirely unimportant. What matters iswhether you hold the appropriate account card and provide the correct PIN password to accessyour account. In financial institutions, you are abstracted into streams of numbers andtransactions that are aggregated with the transactional streams of other clients, which are thenused to streamline operations and predict future economic patterns.On the part of the user, governance is short term and rapidly shifting, but at the same timecontinual, unbounded, and ruled by pleasure and desire. We are not confined. We are freeconsumers. We monitor ourselves or submit to monitoring willingly in order to maintain oraugment social perks. This pattern is characteristic for today’s Web services, such as those that

532Google provides; from efficient Internet searchers, to maps and real-time traffic reports, to cloudstorage, to e-mail and social networking services, to YouTube. The data we willingly divulge areused to “serve us better.” It is here that we see the link between gamification and governance.Gamification enables a form of governance much closer in alignment to what Deleuze proposes.2We broadcast our personal data as the price of participation.With the Fitbit, Luka’s name and identity are rendered unimportant, as is the fact thathe’s using the system as a way to interact and bond with his father. What is important is whetherthe system is registered to a valid online account, and that a steady stream of data—in this case,steps, acceleration data, location, and data on when the system is turned on and off—is beingsent back to Fitbit. These data are combined with the usage patterns of other Fitbit dividuals, aswell as amalgamated with the demographic data culled from these users’ online profiles. Itprovides clues as to what traits, demographics, and usage patterns may correlate to the mostprofitable or loyal users, as well as insight about who to target marketing at, and how to improvethe system’s algorithms and tracking capabilities to attract more lucrative clients. However,gamification as governance promises something more than just tailored services. It promises totell us more about ourselves.The Care of the SelfWhile links may be drawn from discipline, normalization, and panoptic surveillance togamification, in comparison to the governance exhibited in the factories and prisons of theeighteenth, nineteenth, and twentieth centuries that Foucault described, it is somewhat difficult toimagine Luka’s competing with Bruce “from dad's work” as akin to the activity of a prisoninmate. Luka is playing with surveillance in a much more self-directed way. The impetus to track

533and monitor his daily numbers comes from within, and from his dad, rather than the formalauthorities such as the school or state. In this sense, Luka is using gamification in terms of atraditional care of the self. The fact that this care of the self is bundled with other technologies ofgovernance (i.e., Fitbit monitoring Luka and influencing his health and consumption patterns) islargely irrelevant. Accordingly, this playful quantification presents powerful new opportunitiesfor governance.The term care of the self refers to the later work of Michel Foucault (1988). Foucaultargues that the care of the self was a foundational principle of all moral rationality up until theCartesian moment and the Enlightenment. Foucault draws heavily from the Socratic dictates thatone must care for oneself and know oneself, arguing that through this self-reflection and care,individuals come to see themselves as responsible for constituting themselves as moral subjects.This care of the self was achieved in three ways: (1) knowing how to live without luxury,through abstinence, (2) regularly subjecting oneself to a thorough examination of one’sconscience, and (3) be in constant control. Already, we can draw parallels to many gamifiedapplications, such as Health Month or SuperBetter, which prioritize similar forms of selfreflection as a route toward self-improvement.Elsewhere (Whitson and Haggerty 2008), I have used the moral panic surroundingidentity theft to show how this care of the self takes a different modality in the digital age: wenow care for our virtual selves, curating and maintaining the accuracy of our “data doubles,” theinformational profiles (market profiles, credit histories, social networking accounts, and even theavatars and account settings for online games) that have become the lifeblood of our interactionswith others and the real objects of governance.

534In the informational era, our physical bodies seem to fall away. Yet, QS and gamificationmovements are now bringing the body back in. I argue that this characterizes a new modality ofgovernance that leverages a new set of desires—exploration, curiosity, self-mastery—thatcharacterizes both QS and gamification. Gamification, unlike QS, is also imbricated withdiscourses of play that effectively shape how it operates.Part 2: Gamifying the Quantified SelfWhile part 1 of this chapter provided the theoretical background of Foucauldian governance andthe historical evolution of governance techniques, part 2 shows how gamification and the QSmovement come into play. In this section, I continue with my discussion of the care of the self,explaining how the quantified-self movement parallels gamification movements. I then make thenovel argument that fun is irrelevant in gamification. “Fun” in gamification is, for the most part,empty marketing rhetoric. What matters in gamification is giving users actionable feedback onhow to improve. This feedback is what games do well and is where the real link between gamesand gamification lies.The Allure of the Quantified SelfBefore moving onward to gamification, it is first useful to provide some background on QS. TheQS movement was started in 2007 by Gary Wolf and Kevin Kelly, both editors of Wiredmagazine. Interested in using ubiquitous technology to track the self and thus develop selfknowledge, quantified self is also known as “self-tracking,” “body data,” “living by numbers,”“self-surveillance,” “life-hacking,” “personal analytics,” and “personal informatics.” Thesesystems collect information about the user and present it back to them, treating people as both

535the object and the subject of its function (Li 2011, 9–10). Users enroll in QS programs out ofcuriosity, and continue with them because the data provided are so compelling: “they continuebecause they believe their numbers hold secrets that they can’t afford to ignore, includinganswers to questions they have not yet thought to ask” (Wolf 2010). The quantification of theself—by compiling the intricate details of our lives and then rebroadcasting them to us in newways—promises to tell us something about ourselves that we did not already know.While the quantification of the self has commonalities with the time-honored tradition ofjournaling and the care of the self as an ethical practice of reflection detailed by Foucault (1988)and the list-making more recently described by Umberto Eco (2010), what is different is theprecision, complexity, and the amount of the data collected, as well as the way it is ultimatelypresented back to the chronicler. Instead of leaving it up to us to decide what is worthchronicling, and then delegating our spotty memories to provide the details, the journalingprocess in the era of QS is automated, enabling incredibly precise details.The quantification of the self is not new, but automation greatly expands its scale andscope, as well as its effectiveness at telling us our secrets. As stated by Gary Wolf, we trackourselves all the time:We step on a scale and record our weight. We balance a checkbook. We countcalories. But when the familiar pen-and-paper methods of self-analysis areenhanced by sensors that monitor our behaviour automatically, the process of selftracking becomes both more alluring and more meaningful. Automated sensors domore than give us facts; they also remind us that our ordinary behaviour containsobscure quantitative signals that can be used to inform our behaviour, once welearn to read them. (Wolf 2010)

536The unifying methodology of QS is data collection, followed by visualization of these data andcross-referencing, in order to discover correlations and modify behavior.We are used to measuring and quantifying many things in our lives—from optimizingassembly line production, to measuring how fast our computers operate, to grading ourintelligence, to using software to clock how many hours, minutes, and seconds we work eachday. This disciplinary monitoring is commonplace in public spaces (work, school, hospitals).What is new with QS is that individuals are now willingly monitoring themselves innondisciplinary spaces and making these details public. For example, whereas measuring foodintake or mood was previously an activity restricted to health institutions and revealed only toexperts such as nurses and doctors, we now use tools such as diet and health tracking apps toshare and broadcast this information to an unspecified public.Technologies such as the Fitbit or SuperBetter enable us to measure, chart, and quantifywhat was previously unquantifiable and also allow us to transmit and share what was previouslyprivate. It is now relatively simple to measure and analyze patterns in our sleep, exercise, sexlife, food intake, mood, location, alertness, productivity, and even our mental health and spiritualwell-being. We effortlessly track and measure, display and share all of this heretofore unknowndata using our computers, smartphones, and gaming consoles.While the QS may represent the extreme pole of “self-knowledge through numbers”(Wolf 2009), most of us, in one form or another, have quantified our lives in one way or another,from tracking our fuel consumption on smartphone apps, to monitoring our infant’s diaperchanges, feeding times, and sleep schedules, to subscribing to Mint to help us track our spendinghabits. Most of the time we are using gamification to do so. This is not a coincidence. Games andgamified apps are excellent tracking devices.

537Gamification as Feedback Loops, Not FunPromises of fun and play populate the advertisements of gamification companies such asBunchball, BigDoor, Badgeville, Lithium, SCVNGR, Greengoose, and Seriosity. However,critics such as Ian Bogost (this volume) argue that these are empty promises. Behind the emptybadges and meaningless leaderboards, there is often no “game” in gamification. Accordingly, thefailure of gamified products to sustain users and maintain the breathless hype that preceded theirdeployment makes sense (see Evans 2010). They simply do not deliver. Yet I argue here that thefailure of gamification to provide “fun” (whatever that nebulous word means) does not mean thatgamification as a whole is a failure. What gamification successfully borrows from games are themethods to provide clear feedback and reinforcement to users. This, and not playful or gamefuldesign, is what characterizes current examples of gamification.The ways that games render space visible, from points systems to pathfinding, are what isleveraged in gamification. Feedback methods borrowed from games are key to caring for thequantified self. As Stanford psychologist Byron Reeves and his business partner J. Leighton Reiddiscovered in their research on games and gamification, data visualization techniques fromgames are essential tools in shaping users’ behavior: “Game interfaces set a new bar forfeedback. At any one time, Helen sees progress bars, zooming numbers, and status gauges, all ina well-organized dashboard that lets players know how things are going, good or bad. Numbersindicate the health of players, the time left before an attack, the amount of gold accumulated sofar” (Reeves and Read 2009, 71). Games excel at providing precise real-time feedback to helpplayers chart their current progress and determine how to advance. Feedback thus governsbehavior; steps toward a goal are encouraged in multiple ways and channels, while steps in the

538wrong direction are penalized. Feedback can be immediate, for example, providing a World ofWarcraft player with real-time per second data on how he or she is faring in an attack. Butfeedback also takes mid- and long-range forms, providing information on how a player isprogressing with goals that take weeks, months, or even years to accomplish.In games, performance metrics and feedback are overwhelmingly positive and focused onimprovement, reward, and engagement rather than highlighting deficiencies. Of course, failurestill exists, but the risks and punishments for attempting something and then failing are not assevere. Negative feedback works to highlight areas that require improvement and suggestchanges in tactics that may help in achieving success, rather than punishment. Thus, players canclearly decipher what they need to do in order to progress. For example, each failure of a Worldof Warcraft raiding party provides valuable statistics on what techniques are successful (i.e.,attacks that inflict maximum damage, team formations that provide an optimum balance betweentanks and healers, etc.), and what actions to avoid. Each consecutive failure comes with anincremental improvement in strategy until finally, the raiding party is victorious. Failure providesvaluable information on how to become better: “Quick feedback creates immediacy andcontingency in the interactions. When you make a new move, you know quickly whether theaction was right or wrong. The close connection between behaviour and feedback (it's usuallyobvious which reinforcement applies to which behaviours) increases the likelihood that thereinforcement will be effective” (Reeves and Read 2009, 72).Porting the feedback methods used in games to non-game activities thus makes sense. Weturn to gamification to respond to a gap in our day-to-day lives, where feedback on one’sprogress, cues for future directions, and a place for experimentation and even failure is lacking.

539For the most part, feedback in the real world is much more infrequent and difficult toaccomplish, largely because the automated cycle of data collection, compilation, analysis, andfeedbac

grownup, let alone one from his dad’s work, is pure joy (Carmichael 2012). Luka, or more accurately Luka’s dad, is a member of the quantified-self (QS) movement (Schuller 2012). As I argue in this chapter, QS is closely tied with gamification. By gamification I mean “the use of game d