In the late 1990s, subsidized computers and Internet connections in public places were often proffered as an answer to the “digital divide,” the concern that in an information society those without the economic resources to access information technology are disenfranchised from participation in public life. A variety of projects and people worked within this rationale to provide the public free access to computers and Internet connections at schools, libraries, and community centers. Yet, despite large public investments and a small amount of public debate about these centers, little is known about exactly how they have been used. In place of this knowledge there is a vague overarching narrative of technological progress and the preconception that “public Internet access remains in the realm of the educationally and economically disadvantaged” (Lee, 1999, p. 346), at least where free public computers are concerned. More specifically, the typical rationale for free computer centers in public places presumes that the Internet is essentially a transmitter of important information that is to be learned and that this justifies its subsidy. This is sometimes called the “digital library metaphor” for the Internet (Stefik, 1996, p. 6) or the “library model” of public Internet provision (Dobscha, 2004). This vision of Internet use underpins the policy mechanisms that produce and subsidize publicly-funded computer centers.
This article describes a study that employed ethnographic methods to examine a computer center reserved explicitly for children in a socioeconomically depressed area of San Francisco, California. As this article will show, the details of public Internet use by young children stand in contrast to the public rationales and cultural justifications that are usually given for Internet use. Children, as any parent could guess, play with the Internet. Carey’s (1989) useful distinction between transmission vs. ritual communication suggests that the idealized vision of the Internet as a transmitter of information (the “library metaphor”) weighs heavily on the minds of the users and providers at these centers. Children playing with the Internet do not look up facts. Their play does not resemble the transmission of extrinsically useful information and the closing of the “digital divide.” This leads to surprise, anxiety, guilt, and even despair among adults, despite the expectation common to other circumstances that young children like to play.
There is a well-developed interdisciplinary literature that would predict and explain this symbolic and “ritual” (using Carey’s term) Internet use: interdisciplinary theories of play (Gadamer, 1989; Huizinga, 1950). However, due to the ideological baggage accompanying “play,” a play-theoretic approach to subsidized public Internet access is likely to remain unpalatable among policymakers and some researchers. While it is acceptable to use public funding to build children’s playgrounds, playing with public computers is not yet as respectable. This article draws on the literature on play to explain children’s use of public computing in San Francisco and to show how play inverts much of what is known and expected of computer use. It also investigates the way in which the idealized notions found in public policy and law about the “digital divide” and the information economy become fodder for everyday arguments among parents, children, teachers, and librarians about what Internet use ought to be like.
The dominant approach: public internet access as transmission
Some technologies readily suggest their uses, the way that a glove looks like it will fit your hand. Computing is not so simple since it consists of radically multipurpose devices that could be put to a variety of ends. Stefik (1996) explains that metaphors and stories used to describe the Internet are as important to understanding it as the capabilities of the technology itself. Metaphors about the Internet such as “electronic library,”“virtual reality,” and “electronic marketplace” contain lessons: Some uses are always valued and others are devalued. For instance, historically the computer has been readily acceptable as a device of production, and the appropriation of public funds to support access to it has been appropriate when couched in terms of creating productive members of society or moving society as a whole toward an information economy (Castells, 1996). Across the developed world, major policy initiatives in the 1990s sought to subsidize computer-mediated communication for these and other reasons. The United States passed the Telecommunications Act of 1996 and expanded the concept of publicly-subsidized universal telephone service (“universal service,” see Mueller, 1997) to include new technologies like the Internet (Aufderheide, 1999; Hammond, 1997). Unlike prior telephone subsidies, the 1996 Act proposed an institutional model in which schools and libraries are the chief access point for otherwise disenfranchised Internet users, and they receive substantial public funding to provide free Internet service. In the U.S., this program cost from $2.25 billion to $4.6 billion per year from 1998 to 2004 (for estimates, see Ellig, 2006; Hausman, 1997; Jaeger, McClure, & Bertot, 2005).
The concept of universal service is uniquely American in that rationales for it are unlikely to be based on equity or welfare (Rapp, 1996). Instead, U.S. policies employ economic rationales and focus on system benefits. In this manner American arguments for universal service are very comparable to those for universal education (Sawhney, 1994). These policy goals map to the “transmission model” in Carey’s understanding of communication as ritual vs. communication as transmission (Carey, 1974, 1977, 1989). The notion of communication as transmission is premised on the archaic use of the word “communication” to mean physical movement, and in this sense communication is meant to be the transfer of some discrete thing (such as news, information, or knowledge) between otherwise disconnected entities (such as minds, people, or computers). This sort of communication echoes the high value placed on rationality since the enlightenment (for more detail, see Peters, 2001, ch. 2). Carey argued that transmission is a privileged idea of communication.
Public policy and culture generally reward what Carey might term the “secular” use of a computer to retrieve information over a distance. This is the “library model” of the Internet mentioned earlier. Public Internet access as transmission is an instrumental use of a computer to seek and deliver information. It is consistent with rationales both to realize economic and efficiency gains as compared to other forms of communication and to develop a more productive workforce. Consistent with current policy’s emphasis on the school and library (both well-established sites for the transmission of knowledge via books), subsidized access to the Internet in public places can be rationalized as granting the ability to transmit skills and information to the poor and uneducated.
Previous scholarship and the problems with transmission
To date, there are only a few other studies of free public Internet access points, and they have been largely exploratory. Only one comparable study was observationally based; it considered a free library setting and discussed children (Balka & Peterson, 2002, 2004). Research in the community informatics tradition has considered free community technology centers, but has not considered children (for a review, see O’Neil, 2002). The conclusions of these studies are not yet coherent, although Balka and Peterson (2004) suggest that free computer users in Vancouver, Canada were often children and some already have access to the Internet via other means. Additionally, a research brief about libraries generally noted that public Internet access points often manage to reproduce oppressive limits on the groups they are designed to help, specifically women and the elderly (Dobscha, 2004).
Communication in Internet cafés has received more attention, but this research is less relevant in that what motivates users to pay for Internet access has been a primary concern. Studies of cybercafés have found that each café can evolve a particularistic local culture (Miller & Slater, 2000; Wakeford, 2003). One early study rejected the notion that “public” use in a café involved the positive features usually associated with publicness, or that public use was that different from solitary private use (Lee, 1999).1
Some of the previous studies of Internet access in public places are vulnerable to Carey’s critique, introduced earlier, that researchers privilege transmission-related rationales for communication. For instance, using a similar research design to the present study, Balka and Peterson (2004) found that “current use of the Internet in libraries is inconsistent with the goals articulated in current public policies” (p. 152; see also Balka & Peterson, 2002), and Dobscha (2004) found that “[p]ublic internet spaces…do not seem to be serving the needs of those they were initially supposed to be serving” (p. 161). This sort of conclusion in policy analysis research depends upon a false sense of surprise. In fact, one should expect an examination of any law or public policy’s enactment in everyday life to be substantially different from the ideal representation of behavior made by a statute or a speech. Whether jaywalking or Internet access is at issue, the law and policy about a topic is a special kind of value-laden language often used to specify ideals and norms; it is never a description of what actually happens in the street (Moore, 1978). A number of researchers (including the author) have made the mistake of expressing surprise that subsidized Internet connections do not actually result in a disenfranchised child ready to sit down in front of the computer for an afternoon of job training to better herself. As Lee (1999) notes, an unnecessary focus on “governmental rhetoric characterizes much of the work in this area” (p. 346).
The study described here began as an exploratory study with a theoretical framework that emerged inductively. Early in this research, the relevance of play theory became clear, and the question of what users “ought” to do with the Internet emerged from a disproportionately large number of the conversations observed in San Francisco. As this study will explain, users largely did not conceive of the Internet as the transmission of economically valuable skills and information. However, children expend considerable effort to demarcate and defend their non-instrumental use, while adults worry and feel guilty about the fact that the use of these centers does not fit the ideals specified by policy to address the “digital divide.”
Why this should be so surprising is one of the puzzles this study addresses. Other communication media are readily acceptable as sites for play. In the context of communication research on children co-viewing television with siblings and adults, television has been found to be a stimulus to imaginative play (Singer & Singer, 1976) and an object around which play is actively organized (Reid & Frazer, 1980). Earlier research which focused exclusively on online discourse (Danet, 2001; Ruedenberg, Danet, & Rosenbaum-Tamari, 1995) also explained the Internet in terms of play. Even newspapers, a communication medium often associated with the transmission of information (news), have been studied as objects of play (Glasser, 1982, 2000).
Play and a relational technology
In contrast to the idea of transmission, one might expect that children not yet familiar with the connotations of computing and these debates in public policy would be free to justify the computer as not merely useful for transmission, but also for what Carey terms “ritual” communication, a relational use of computers as a way to maintain society across time rather than exchanging messages across space (Carey, 1989). In ritual communication, it is not the exchange of information that is relevant, but the cultural understandings developed through an interaction. As early as 1974, Carey criticized communication scholarship for its lack of interest in symbolic and ritual communication. However, a wide variety of interdisciplinary scholarship outside of communication has considered ritual communication within theories of play (for a comprehensive review, see Sutton-Smith, 1997). As Huizinga stated flatly in 1950, “ritual is play” (p. 5). In the years since Carey’s critique, communication scholarship has begun to address this issue in research on a variety of media and is beginning to incorporate the link between ritual and play. Understanding communication as play invites us to “…think beyond the rational, utilitarian, extrinsic, and almost always instrumental reasons individuals give for their attention to, and interest in, the programs and publications—and now Web sites—of their choice” (Glasser, 2000).
In other words, while computers, televisions, and newspapers may transmit skills and information, people also play with them with little expectation of acquiring anything from them directly (cf. Glasser, 1982; Stephenson, 1964, 1967). While this approach has also been used to understand computer-mediated communication (see, e.g., Danet, 1995, 2001), the most widespread work on computer-mediated communication is largely from a functional psychological perspective (e.g., Turkle, 1984, 1995) or focuses on play within the topic of maintaining ties within virtual communities (e.g., Jones, 1995, 1997, 1998; Smith & Kollock, 1999).
When considering play, a large, diffuse literature from a number of fields proposes functional understandings of specific kinds of play at the level of the individual. While play may be non-instrumental from the perspective of the player (play may seem useless), a great deal of scholarly effort has been expended to demonstrate that play is adaptive and useful. This approach is a straightforward evolution from the earliest writing about play (see Rubin, 1982). In the 18th century, Friedrich von Schiller envisioned play as the necessary expenditure of surplus energy. Karl Groos in 1901 found play to be the elaboration and perfection of learned behaviors. Genetic psychologist G. Stanley Hall envisioned play as useful catharsis in his recapitulation theory. Since it is in fact quite difficult to demonstrate that most play is functionally useful and adaptive, other literature instead takes the approach that the word “play” is “a linguistic waste-paper basket” for activities that are mistakenly grouped together in common speech, but should be theoretically distinct (Millar, 1968, p. 11). Individual approaches most common in social psychology have the positive effect of then being able to functionally explain some play forms as need satisfaction, social learning, socialization, or personality development, but this leaves a wide swath of what is usually termed “play” embarrassingly unconsidered.
A distinctly different approach that will be employed here developed in the literature of sociology: an approach that “proceeded without much regard for either the general conception of play used by the [earlier] philosophers and psychologists, or for their conclusions” (Giddens, 1964, p. 80). This view begins with the idea that the assignment of the signifier “play” to diverse activities in diverse contexts is no accident. The most frequent contemporary starting point in this tradition is Johan Huizinga (1950), who articulated play as an activity that is freely chosen, bounded, outside “ordinary” life, and totally absorbing for the player (p. 13). Huizinga argued that some play is present in all elements of life across history. Indeed, cross-cultural examinations find remarkably parallel evolution of concepts for play (Huizinga, 1950) and games (Csikszentmihalyi & Bennett, 1971) in different societies. Huizinga defined play in opposition to seriousness, a limiting formulation that seems true only for contemporary society, and perhaps not even then (Anchor, 1978). He also defined play as “an activity connected with no material interest” (Huizinga, 1950, p. 13). A somewhat modified approach recalls Veblen (1899) and understands play as leisure and class signification (Hearn, 1976/1977), or as a way for society to “[leave] hope in the dispossessed that free competition is still possible in the lowly stations of life” (Callois, 1961, p. 115).
The present study, following Glasser (2000), follows this broader approach and employs Gadamer’s relatively recent philosophical analysis of play (1989) with the theoretical framework introduced by Huizinga (1950). In this thinking, “every game presents the [person] who plays it with a task…the purpose of the game is not really solving the task, but ordering and shaping the movement of the game itself” (Gadamer, 1989, p. 107). Gadamer argues that it is an error to consider play at the level of the individual (p. 104). Instead, play is fundamentally relational; it is a “mode of being” subsuming both the object of play and the consciousness of players. This mode requires, “not necessarily literally another player, but something else with which the player plays” (pp. 105-106). Play is then “experienced subjectively as relaxation” or pleasure, even though it may require considerable effort, because it “absorbs the player” and in this way frees her from the burden of initiative (p. 105). The player makes decisions, but their consequences are limited to the game. Play requires a demarcated space to play (a playing field, explicit or implicit) and may be limited in time. Play is “ordered” and has rules, although the rules may not be written down (Huizinga, 1950). While play may involve a task, it has no external goal as such, and is often self-renewing or repetitive (Gadamer, 1989, p. 103). Finally, games are voluntary, and playing a game can be a way to express freedom (Huizinga, 1950).