Yale University Press
New Haven and London
Copyright © 2006 by Yochai Benkler.
This online version is available under a Creative Commons Noncommercial Sharealike license - see http://www.benkler.org/ - and has been reformatted and designated as recommended reading for the NGO Committee on Education of the Conference Of Non-Governmental Organizations in Consultative Relationship with the United Nations in conjunction with the Committee's commitment to the United Nations Decade of Education for Sustainable Development, the International Decade for a Culture of Peace and Non-violence for the Children of the World, the Millennium Development Goals and related international agreements, conventions and treaties.
"Human nature is not a machine to be built after a model, and set to do exactly the work prescribed for it, but a tree, which requires to grow and develop itself on all sides, according to the tendency of the inward forces which make it a living thing."
"Such are the differences among human beings in their sources of pleasure, their susceptibilities of pain, and the operation on them of different physical and moral agencies, that unless there is a corresponding diversity in their modes of life, they neither obtain their fair share of happiness, nor grow up to the mental, moral, and aesthetic stature of which their nature is capable."
John Stuart Mill, On Liberty (1859)
Modern democracies and mass media have coevolved throughout the twentieth century. The first modern national republics-the early American Republic, the French Republic from the Revolution to the Terror, the Dutch Republic, and the early British parliamentary monarchy-preexisted mass media. They provide us with some model of the shape of the public sphere in a republic without mass media, what Jurgen Habermas called the bourgeois public sphere. However, the expansion of democracies in complex modern societies has largely been a phenomenon of the late nineteenth and twentieth centuries-in particular, the post-World War II years. During this period, the platform of the public sphere was dominated by mass media-print, radio, and television. In authoritarian regimes, these means of mass communication were controlled by the state. In democracies, they operated either under state ownership, with varying degrees of independence from the sitting government, or under private ownership financially dependent on advertising markets. We do not, therefore, have examples of complex modern democracies whose public sphere is built on a platform that is widely distributed and independent of both government control and market demands. The Internet as a technology, and the networked information economy as an organizational and social model of information and cultural production, promise the emergence of a substantial alternative platform for the public sphere. The networked public sphere, as it is currently developing, suggests that it will have no obvious points of control or exertion of influence-either by fiat or by purchase. It seems to invert the mass-media model in that it is driven heavily by what dense clusters of users find intensely interesting and engaging, rather than by what large swathes of them find mildly interesting on average. And it promises to offer a platform for engaged citizens to cooperate and provide observations and opinions, and to serve as a watchdog over society on a peer-production model.
The claim that the Internet democratizes is hardly new. "Everyone a pamphleteer" has been an iconic claim about the Net since the early 1990s. It is a claim that has been subjected to significant critique. What I offer, therefore, in this chapter and the next is not a restatement of the basic case, but a detailed analysis of how the Internet and the emerging networked information economy provide us with distinct improvements in the structure of the public sphere over the mass media. I will also explain and discuss the solutions that have emerged within the networked environment itself to some of the persistent concerns raised about democracy and the Internet: the problems of information overload, fragmentation of discourse, and the erosion of the watchdog function of the media.
For purposes of considering political freedom, I adopt a very limited definition of "public sphere." The term is used in reference to the set of practices that members of a society use to communicate about matters they understand to be of public concern and that potentially require collective action or recognition. Moreover, not even all communications about matters of potential public concern can be said to be part of the public sphere. Communications within self-contained relationships whose boundaries are defined independently of the political processes for collective action are "private," if those communications remain purely internal. Dinner-table conversations, grumblings at a bridge club, or private letters have that characteristic, if they occur in a context where they are not later transmitted across the associational boundaries to others who are not part of the family or the bridge club. Whether these conversations are, or are not, part of the public sphere depends on the actual communications practices in a given society. The same practices can become an initial step in generating public opinion in the public sphere if they are nodes in a network of communications that do cross associational boundaries. A society with a repressive regime that controls the society-wide communications facilities nonetheless may have an active public sphere if social networks and individual mobility are sufficient to allow opinions expressed within discrete associational settings to spread throughout a substantial portion of the society and to take on political meaning for those who discuss them. The public sphere is, then, a sociologically descriptive category. It is a term for signifying how, if at all, people in a given society speak to each other in their relationship as constituents about what their condition is and what they ought or ought not to do as a political unit. This is a purposefully narrow conception of the public sphere. It is intended to focus on the effects of the networked environment on what has traditionally been understood to be political participation in a republic. I postpone consideration of a broader conception of the public sphere, and of the political nature of who gets to decide meaning and how cultural interpretations of the conditions of life and the alternatives open to a society are created and negotiated in a society until chapter 8.
The practices that define the public sphere are structured by an interaction of culture, organization, institutions, economics, and technical communications infrastructure. The technical platforms of ink and rag paper, handpresses, and the idea of a postal service were equally present in the early American Republic, Britain, and France of the late eighteenth and early nineteenth centuries. However, the degree of literacy, the social practices of newspaper reading, the relative social egalitarianism as opposed to elitism, the practices of political suppression or subsidy, and the extent of the postal system led to a more egalitarian, open public sphere, shaped as a network of smaller-scale local clusters in the United States, as opposed to the more tightly regulated and elitist national and metropolis-centered public spheres of France and Britain. The technical platforms of mass-circulation print and radio were equally available in the Soviet Union and Nazi Germany, in Britain, and in the United States in the 1930s. Again, however, the vastly different political and legal structures of the former created an authoritarian public sphere, while the latter two, both liberal public spheres, differed significantly in the business organization and economic model of production, the legal framework and the cultural practices of reading and listening-leading to the then still elitist overlay on the public sphere in Britain relative to a more populist public sphere in the United States.
Mass media structured the public sphere of the twentieth century in all advanced modern societies. They combined a particular technical architecture, a particular economic cost structure, a limited range of organizational forms, two or three primary institutional models, and a set of cultural practices typified by consumption of finished media goods. The structure of the mass media resulted in a relatively controlled public sphere-although the degree of control was vastly different depending on whether the institutional model was liberal or authoritarian-with influence over the debate in the public sphere heavily tilted toward those who controlled the means of mass communications. The technical architecture was a one-way, hub-and-spoke structure, with unidirectional links to its ends, running from the center to the periphery. A very small number of production facilities produced large amounts of identical copies of statements or communications, which could then be efficiently sent in identical form to very large numbers of recipients. There was no return loop to send observations or opinions back from the edges to the core of the architecture in the same channel and with similar salience to the communications process, and no means within the mass-media architecture for communication among the end points about the content of the exchanges. Communications among the individuals at the ends were shunted to other media-personal communications or telephones-which allowed communications among the ends. However, these edge media were either local or one-to-one. Their social reach, and hence potential political efficacy, was many orders of magnitude smaller than that of the mass media.
The economic structure was typified by high-cost hubs and cheap, ubiquitous, reception-only systems at the ends. This led to a limited range of organizational models available for production: those that could collect sufficient funds to set up a hub. These included: state-owned hubs in most countries; advertising-supported commercial hubs in some of the liberal states, most distinctly in the United States; and, particularly for radio and television, the British Broadcasting Corporation (BBC) model or hybrid models like the Canadian Broadcasting Corporation (CBC) in Canada. The role of hybrid and purely commercial, advertising-supported media increased substantially around the globe outside the United States in the last two to three decades of the twentieth century. Over the course of the century, there also emerged civil-society or philanthropy-supported hubs, like the party presses in Europe, nonprofit publications like Consumer Reports (later, in the United States), and, more important, public radio and television. The one-way technical architecture and the mass-audience organizational model underwrote the development of a relatively passive cultural model of media consumption. Consumers (or subjects, in authoritarian systems) at the ends of these systems would treat the communications that filled the public sphere as finished goods. These were to be treated not as moves in a conversation, but as completed statements whose addressees were understood to be passive: readers, listeners, and viewers.
The Internet's effect on the public sphere is different in different societies, depending on what salient structuring components of the existing public sphere its introduction perturbs. In authoritarian countries, it is the absence of a single or manageably small set of points of control that is placing the greatest pressure on the capacity of the regimes to control their public sphere, and thereby to simplify the problem of controlling the actions of the population. In liberal countries, the effect of the Internet operates through its implications for economic cost and organizational form. In both cases, however, the most fundamental and potentially long-standing effect that Internet communications are having is on the cultural practice of public communication. The Internet allows individuals to abandon the idea of the public sphere as primarily constructed of finished statements uttered by a small set of actors socially understood to be "the media" (whether state owned or commercial) and separated from society, and to move toward a set of social practices that see individuals as participating in a debate. Statements in the public sphere can now be seen as invitations for a conversation, not as finished goods. Individuals can work their way through their lives, collecting observations and forming opinions that they understand to be practically capable of becoming moves in a broader public conversation, rather than merely the grist for private musings.
How is private opinion about matters of collective, formal, public action formed? How is private opinion communicated to others in a form and in channels that allow it to be converted into a public, political opinion, and a position worthy of political concern by the formal structures of governance of a society? How, ultimately, is such a political and public opinion converted into formal state action? These questions are central to understanding how individuals in complex contemporary societies, located at great distances from each other and possessing completely different endowments of material, intellectual, social, and formal ties and capabilities, can be citizens of the same democratic polity rather than merely subjects of a more or less responsive authority. In the idealized Athenian agora or New England town hall, the answers are simple and local. All citizens meet in the agora, they speak in a way that all relevant citizens can hear, they argue with each other, and ultimately they also constitute the body that votes and converts the opinion that emerges into a legitimate action of political authority. Of course, even in those small, locally bounded polities, things were never quite so simple. Nevertheless, the idealized version does at least give us a set of functional characteristics that we might seek in a public sphere: a place where people can come to express and listen to proposals for agenda items-things that ought to concern us as members of a polity and that have the potential to become objects of collective action; a place where we can make and gather statements of fact about the state of our world and about alternative courses of action; where we can listen to opinions about the relative quality and merits of those facts and alternative courses of action; and a place where we can bring our own concerns to the fore and have them evaluated by others.
Understood in this way, the public sphere describes a social communication process. Habermas defines the public sphere as "a network for communicating information and points of view (i.e., opinions expressing affirmative or negative attitudes)"; which, in the process of communicating this information and these points of view, filters and synthesizes them "in such a way that they coalesce into bundles of topically specified public opinions."1 Taken in this descriptive sense, the public sphere does not relate to a particular form of public discourse that is normatively attractive from some perspective or another. It defines a particular set of social practices that are necessary for the functioning of any complex social system that includes elements of governing human beings. There are authoritarian public spheres, where communications are regimented and controlled by the government in order to achieve acquiescence and to mobilize support, rather than relying solely on force to suppress dissent and opposition. There are various forms of liberal public spheres, constituted by differences in the political and communications systems scattered around liberal democracies throughout the world. The BBC or the state-owned televisions throughout postwar Western European democracies, for example, constituted the public spheres in different ways than did the commercial mass media that dominated the American public sphere. As advertiser-supported mass media have come to occupy a larger role even in places where they were not dominant before the last quarter of the twentieth century, the long American experience with this form provides useful insight globally.
In order to consider the relative advantages and failures of various platforms for a public sphere, we need to define a minimal set of desiderata that such a platform must possess. My point is not to define an ideal set of constraints and affordances of the public sphere that would secure legitimacy or would be most attractive under one conception of democracy or another. Rather, my intention is to define a design question: What characteristics of a communications system and practices are sufficiently basic to be desired by a wide range of conceptions of democracy? With these in hand, we will be able to compare the commercial mass media and the emerging alternatives in the digitally networked environment.
Universal Intake. Any system of government committed to the idea that, in principle, the concerns of all those governed by that system are equally respected as potential proper subjects for political action and that all those governed have a say in what government should do requires a public sphere that can capture the observations of all constituents. These include at least their observations about the state of the world as they perceive and understand it, and their opinions of the relative desirability of alternative courses of action with regard to their perceptions or those of others. It is important not to confuse "universal intake" with more comprehensive ideas, such as that every voice must be heard in actual political debates, or that all concerns deserve debate and answer. Universal intake does not imply these broader requirements. It is, indeed, the role of filtering and accreditation to whittle down what the universal intake function drags in and make it into a manageable set of political discussion topics and interventions. However, the basic requirement of a public sphere is that it must in principle be susceptible to perceiving and considering the issues of anyone who believes that their condition is a matter appropriate for political consideration and collective action. The extent to which that personal judgment about what the political discourse should be concerned with actually coincides with what the group as a whole will consider in the public sphere is a function of the filtering and accreditation functions.
Filtering for Potential Political Relevance. Not everything that someone considers to be a proper concern for collective action is perceived as such by most other participants in the political debate. A public sphere that has some successful implementation of universal intake must also have a filter to separate out those matters that are plausibly within the domain of organized political action and those that are not. What constitutes the range of plausible political topics is locally contingent, changes over time, and is itself a contested political question, as was shown most obviously by the "personal is political" feminist intellectual campaign. While it left "my dad won't buy me the candy I want" out of the realm of the political, it insisted on treating "my husband is beating me" as critically relevant in political debate. An overly restrictive filtering system is likely to impoverish a public sphere and rob it of its capacity to develop legitimate public opinion. It tends to exclude views and concerns that are in fact held by a sufficiently large number of people, or to affect people in sufficiently salient ways that they turn out, in historical context, to place pressure on the political system that fails to consider them or provide a legitimate answer, if not a solution. A system that is too loose tends to fail because it does not allow a sufficient narrowing of focus to provide the kind of sustained attention and concentration necessary to consider a matter and develop a range of public opinions on it.
Filtering for Accreditation. Accreditation is different from relevance, requires different kinds of judgments, and may be performed in different ways than basic relevance filtering. A statement like "the president has sold out space policy to Martians" is different from "my dad won't buy me the candy I want." It is potentially as relevant as "the president has sold out energy policy to oil companies." What makes the former a subject for entertainment, not political debate, is its lack of credibility. Much of the function of journalistic professional norms is to create and preserve the credibility of the professional press as a source of accreditation for the public at large. Parties provide a major vehicle for passing the filters of both relevance and accreditation. Academia gives its members a source of credibility, whose force (ideally) varies with the degree to which their statements come out of, and pertain to, their core roles as creators of knowledge through their disciplinary constraints. Civil servants in reasonably professional systems can provide a source of accreditation. Large corporations have come to play such a role, though with greater ambiguity. The emerging role of nongovernment organizations (NGOs), very often is intended precisely to preorganize opinion that does not easily pass the relevant public sphere's filters of relevance and accreditation and provide it with a voice that will. Note that accreditation of a move in political discourse is very different from accreditation of a move in, for example, academic discourse, because the objective of each system is different. In academic discourse, the fact that a large number of people hold a particular opinion ("the universe was created in seven days") does not render that opinion credible enough to warrant serious academic discussion. In political discourse, say, about public school curricula, the fact that a large number of people hold the same view and are inclined to have it taught in public schools makes that claim highly relevant and "credible." In other words, it is credible that this could become a political opinion that forms a part of public discourse with the potential to lead to public action.
Filters, both for relevance and accreditation, provide a critical point of control over the debate, and hence are extremely important design elements.
Synthesis of "Public Opinion. The communications system that offers the platform for the public sphere must also enable the synthesis of clusters of individual opinion that are sufficiently close and articulated to form something more than private opinions held by some number of individuals. How this is done is tricky, and what counts as "public opinion" may vary among different theories of democracy. In deliberative conceptions, this might make requirements of the form of discourse. Civic republicans would focus on open deliberation among people who see their role as deliberating about the common good. Habermas would focus on deliberating under conditions that assure the absence of coercion, while Bruce Ackerman would admit to deliberation only arguments formulated so as to be neutral as among conceptions of the good. In pluralist conceptions, like John Rawls's in Political Liberalism, which do not seek ultimately to arrive at a common understanding but instead seek to peaceably clear competing positions as to how we ought to act as a polity, this might mean the synthesis of a position that has sufficient overlap among those who hold it that they are willing to sign on to a particular form of statement in order to get the bargaining benefits of scale as an interest group with a coherent position. That position then comes to the polls and the bargaining table as one that must be considered, overpowered, or bargained with. In any event, the platform has to provide some capacity to synthesize the finely disparate and varied versions of beliefs and positions held by actual individuals into articulated positions amenable for consideration and adoption in the formal political sphere and by a system of government, and to render them in ways that make them sufficiently salient in the overall mix of potential opinions to form a condensation point for collective action.
Independence from Government Control. The core role of the political public sphere is to provide a platform for converting privately developed observations, intuitions, and opinions into public opinions that can be brought to bear in the political system toward determining collective action. One core output of these communications is instructions to the administration sitting in government. To the extent that the platform is dependent on that same sitting government, there is a basic tension between the role of debate in the public sphere as issuing instructions to the executive and the interests of the sitting executive to retain its position and its agenda and have it ratified by the public. This does not mean that the communications system must exclude government from communicating its positions, explaining them, and advocating them. However, when it steps into the public sphere, the locus of the formation and crystallization of public opinion, the sitting administration must act as a participant in explicit conversation, and not as a platform controller that can tilt the platform in its direction.
Throughout the twentieth century, the mass media have played a fundamental constitutive role in the construction of the public sphere in liberal democracies. Over this period, first in the United States and later throughout the world, the commercial, advertising-supported form of mass media has become dominant in both print and electronic media. Sometimes, these media have played a role that has drawn admiration as "the fourth estate." Here, the media are seen as a critical watchdog over government processes, and as a major platform for translating the mobilization of social movements into salient, and ultimately actionable, political statements. These same media, however, have also drawn mountains of derision for the power they wield, as well as fail to wield, and for the shallowness of public communication they promote in the normal course of the business of selling eyeballs to advertisers. Nowhere was this clearer than in the criticism of the large role that television came to play in American public culture and its public sphere. Contemporary debates bear the imprint of the three major networks, which in the early 1980s still accounted for 92 percent of television viewers and were turned on and watched for hours a day in typical American homes. These inspired works like Neil Postman's Amusing Ourselves to Death or Robert Putnam's claim, in Bowling Alone, that television seemed to be the primary identifiable discrete cause of the decline of American civic life. Nevertheless, whether positive or negative, variants of the mass-media model of communications have been dominant throughout the twentieth century, in both print and electronic media. The mass-media model has been the dominant model of communications in both democracies and their authoritarian rivals throughout the period when democracy established itself, first against monarchies, and later against communism and fascism. To say that mass media were dominant is not to say that only technical systems of remote communications form the platform of the public sphere. As Theda Skocpol and Putnam have each traced in the context of the American and Italian polities, organizations and associations of personal civic involvement form an important platform for public participation. And yet, as both have recorded, these platforms have been on the decline. So "dominant" does not mean sole, but instead means overridingly important in the structuring of the public sphere. It is this dominance, not the very existence, of mass media that is being challenged by the emergence of the networked public sphere.
The roots of the contemporary industrial structure of mass media presage both the attractive and unattractive aspects of the media we see today. Pioneered by the Dutch printers of the seventeenth century, a commercial press that did not need to rely on government grants and printing contracts, or on the church, became a source of a constant flow of heterodox literature and political debate.2 However, a commercial press has always also been sensitive to the conditions of the marketplace-costs, audience, and competition. In seventeenth-century England, the Stationers' Monopoly provided its insiders enough market protection from competitors that its members were more than happy to oblige the Crown with a compliant press in exchange for monopoly. It was only after the demise of that monopoly that a genuinely political press appeared in earnest, only to be met by a combination of libel prosecutions, high stamp taxes, and outright bribery and acquisition by government.3 These, like the more direct censorship and sponsorship relationships that typified the prerevolutionary French press, kept newspapers and gazettes relatively compliant, and their distribution largely limited to elite audiences. Political dissent did not form part of a stable and independent market-based business model. As Paul Starr has shown, the evolution of the British colonies in America was different. While the first century or so of settlement saw few papers, and those mostly "authorized" gazettes, competition began to increase over the course of the eighteenth century. The levels of literacy, particularly in New England, were exceptionally high, the population was relatively prosperous, and the regulatory constraints that applied in England, including the Stamp Tax of 1712, did not apply in the colonies. As second and third newspapers emerged in cities like Boston, Philadelphia, and New York, and were no longer supported by the colonial governments through postal franchises, the public sphere became more contentious. This was now a public sphere whose voices were self-supporting, like Benjamin Franklin's Pennsylvania Gazette. The mobilization of much of this press during the revolutionary era, and the broad perception that it played an important role in constituting the American public, allowed the commercial press to continue to play an independent and critical role after the revolution as well, a fate not shared by the brief flowering of the press immediately after the French Revolution. A combination of high literacy and high government tolerance, but also of postal subsidies, led the new United States to have a number and diversity of newspapers unequalled anywhere else, with a higher weekly circulation by 1840 in the 17-million-strong United States than in all of Europe with its population then of 233 million. By 1830, when Tocqueville visited America, he was confronted with a widespread practice of newspaper reading-not only in towns, but in far-flung farms as well, newspapers that were a primary organizing mechanism for political association.4
This widespread development of small-circulation, mostly local, competitive commercial press that carried highly political and associational news and opinion came under pressure not from government, but from the economies of scale of the mechanical press, the telegraph, and the ever-expanding political and economic communities brought together by rail and industrialization. Harold Innis argued more than half a century ago that the increasing costs of mechanical presses, coupled with the much-larger circulation they enabled and the availability of a flow of facts from around the world through telegraph, reoriented newspapers toward a mass-circulation, relatively low-denominator advertising medium. These internal economies, as Alfred Chandler and, later, James Beniger showed in their work, intersected with the vast increase in industrial output, which in turn required new mechanisms of demand management-in other words, more sophisticated advertising to generate and channel demand. In the 1830s, the Sun and Herald were published in New York on large-circulation scales, reducing prices to a penny a copy and shifting content from mostly politics and business news to new forms of reporting: petty crimes from the police courts, human-interest stories, and outright entertainment-value hoaxes.5 The start-up cost of founding such mass-circulation papers rapidly increased over the second quarter of the nineteenth century, as figure 6.1 illustrates.
Figure 6.1: Start-up Costs of a Daily Newspaper, 1835-1850 (in 2005 dollars)
James Gordon Bennett founded the Herald in 1835, with an investment of five hundred dollars, equal to a little more than $10,400 in 2005 dollars. By 1840, the necessary investment was ten to twenty times greater, between five and ten thousand dollars, or $106,000-$212,000 in 2005 terms. By 1850, that amount had again grown tenfold, to $100,000, about $2.38 million in 2005.6 In the span of fifteen years, the costs of starting a newspaper rose from a number that many could conceive of spending for a wide range of motivations using a mix of organizational forms, to something that required a more or less industrial business model to recoup a very substantial financial investment. The new costs reflected mutually reinforcing increases in organizational cost (because of the professionalization of the newspaper publishing model) and the introduction of high-capacity, higher-cost equipment: electric presses (1839); the Hoe double-cylinder rotary press (1846), which raised output from the five hundred to one thousand sheets per hour of the early steam presses (up from 250 sheets for the handpress) to twelve thousand sheets per hour; and eventually William Bullock's roll-fed rotary press that produced twelve thousand complete newspapers per hour by 1865. The introduction of telegraph and the emergence of news agencies-particularly the Associated Press (AP) in the United States and Reuters in England-completed the basic structure of the commercial printed press. These characteristics-relatively high cost, professional, advertising supported, dependent on access to a comparatively small number of news agencies (which, in the case of the AP, were often used to anticompetitive advantage by their members until the mid-twentieth-century antitrust case)-continued to typify print media. With the introduction of competition from radio and television, these effects tended to lead to greater concentration, with a majority of papers facing no local competition, and an ever-increasing number of papers coming under the joint ownership of a very small number of news publishing houses.
The introduction of radio was the next and only serious potential inflection point, prior to the emergence of the Internet, at which some portion of the public sphere could have developed away from the advertiser-supported mass-media model. In most of Europe, radio followed the path of state-controlled media, with variable degrees of freedom from the executive at different times and places. Britain developed the BBC, a public organization funded by government-imposed levies, but granted sufficient operational freedom to offer a genuine platform for a public sphere, as opposed to a reflection of the government's voice and agenda. While this model successfully developed what is perhaps the gold standard of broadcast journalism, it also grew as a largely elite institution throughout much of the twentieth century. The BBC model of state-based funding and monopoly with genuine editorial autonomy became the basis of the broadcast model in a number of former colonies: Canada and Australia adopted a hybrid model in the 1930s. This included a well-funded public broadcaster, but did not impose a monopoly in its favor, allowing commercial broadcasters to grow alongside it. Newly independent former colonies in the postwar era that became democracies, like India and Israel, adopted the model with monopoly, levy-based funding, and a degree of editorial independence. The most currently visible adoption of a hybrid model based on some state funding but with editorial freedom is Al Jazeera, the Arab satellite station partly funded by the Emir of Qatar, but apparently free to pursue its own editorial policy, whose coverage stands in sharp contrast to that of the state-run broadcasters in the region. In none of these BBC-like places did broadcast diverge from the basic centralized communications model of the mass media, but it followed a path distinct from the commercial mass media. Radio, and later television, was a more tightly controlled medium than was the printed press; its intake, filtering, and synthesis of public discourse were relatively insulated from the pressure of both markets, which typified the American model, and politics, which typified the state-owned broadcasters. These were instead controlled by the professional judgments of their management and journalists, and showed both the high professionalism that accompanied freedom along both those dimensions and the class and professional elite filters that typify those who control the media under that organizational model. The United States took a different path that eventually replicated, extended, and enhanced the commercial, advertiser-supported mass-media model originated in the printed press. This model was to become the template for the development of similar broadcasters alongside the state-owned and independent BBC-model channels adopted throughout much of the rest of the world, and of programming production for newer distribution technologies, like cable and satellite stations. The birth of radio as a platform for the public sphere in the United States was on election night in 1920.7 Two stations broadcast the election returns as their launchpad for an entirely new medium-wireless broadcast to a wide audience. One was the Detroit News amateur station, 8MK, a broadcast that was framed and understood as an internal communication of a technical fraternity-the many amateurs who had been trained in radio communications for World War I and who then came to form a substantial and engaged technical community. The other was KDKA Pittsburgh, launched by Westinghouse as a bid to create demand for radio receivers of a kind that it had geared up to make during the war. Over the following four or five years, it was unclear which of these two models of communication would dominate the new medium. By 1926, however, the industrial structure that would lead radio to follow the path of commercial, advertiser-supported, concentrated mass media, dependent on government licensing and specializing in influencing its own regulatory oversight process was already in place.
Although this development had its roots in the industrial structure of radio production as it emerged from the first two decades of innovation and businesses in the twentieth century, it was shaped significantly by political-regulatory choices during the 1920s. At the turn of the twentieth century, radio was seen exclusively as a means of wireless telegraphy, emphasizing ship-to-shore and ship-to-ship communications. Although some amateurs experimented with voice programs, broadcast was a mode of point-to-point communications; entertainment was not seen as its function until the 1920s. The first decade and a half of radio in the United States saw rapid innovation and competition, followed by a series of patent suits aimed to consolidate control over the technology. By 1916, the ideal transmitter based on technology available at the time required licenses of patents held by Marconi, AT&T, General Electric (GE), and a few individuals. No licenses were in fact granted. The industry had reached stalemate. When the United States joined the war, however, the navy moved quickly to break the stalemate, effectively creating a compulsory cross-licensing scheme for war production, and brought in Westinghouse, the other major potential manufacturer of vacuum tubes alongside GE, as a participant in the industry. The two years following the war saw intervention by the U.S. government to assure that American radio industry would not be controlled by British Marconi because of concerns in the navy that British control over radio would render the United States vulnerable to the same tactic Britain used against Germany at the start of the war-cutting off all transoceanic telegraph communications. The navy brokered a deal in 1919 whereby a new company was created-the Radio Corporation of America (RCA)-which bought Marconi's American business. By early 1920, RCA, GE, and AT&T entered into a patent cross-licensing model that would allow each to produce for a market segment: RCA would control transoceanic wireless telegraphy, while GE and AT&T's Western Electric subsidiary would make radio transmitters and sell them under the RCA brand. This left Westinghouse with production facilities developed for the war, but shut out of the existing equipment markets by the patent pool. Launching KDKA Pittsburgh was part of its response: Westinghouse would create demand for small receivers that it could manufacture without access to the patents held by the pool. The other part of its strategy consisted of acquiring patents that, within a few months, enabled Westinghouse to force its inclusion in the patent pool, redrawing the market division map to give Westinghouse 40 percent of the receiving equipment market. The first part of Westinghouse's strategy, adoption of broadcasting to generate demand for receivers, proved highly successful and in the long run more important. Within two years, there were receivers in 10 percent of American homes. Throughout the 1920s, equipment sales were big business.
Radio stations, however, were not dominated by the equipment manufacturers, or by anyone else for that matter, in the first few years. While the equipment manufacturers did build powerful stations like KDKA Pittsburgh, WJZ Newark, KYW Chicago (Westinghouse), and WGY Schenectady (GE), they did not sell advertising, but rather made their money from equipment sales. These stations did not, in any meaningful sense of the word, dominate the radio sphere in the first few years of radio, as the networks would indeed come to do within a decade. In November 1921, the first five licenses were issued by the Department of Commerce under the new category of "broadcasting" of "news, lectures, entertainment, etc." Within eight months, the department had issued another 453 licenses. Many of these went to universities, churches, and unions, as well as local shops hoping to attract business with their broadcasts. Universities, seeing radio as a vehicle for broadening their role, began broadcasting lectures and educational programming. Seventy-four institutes of higher learning operated stations by the end of 1922. The University of Nebraska offered two-credit courses whose lectures were transmitted over the air. Churches, newspapers, and department stores each forayed into this new space, much as we saw the emergence of Web sites for every organization over the course of the mid-1990s. Thousands of amateurs were experimenting with technical and format innovations. While receivers were substantially cheaper than transmitters, it was still possible to assemble and sell relatively cheap transmitters, for local communications, at prices sufficiently low that thousands of individual amateurs could take to the air. At this point in time, then, it was not yet foreordained that radio would follow the mass-media model, with a small number of well-funded speakers and hordes of passive listeners. Within a short period, however, a combination of technology, business practices, and regulatory decisions did in fact settle on the model, comprised of a small number of advertiser-supported national networks, that came to typify the American broadcast system throughout most of the rest of the century and that became the template for television as well.
Herbert Hoover, then secretary of commerce, played a pivotal role in this development. Throughout the first few years after the war, Hoover had positioned himself as the champion of making control over radio a private market affair, allying himself both with commercial radio interests and with the amateurs against the navy and the postal service, each of which sought some form of nationalization of radio similar to what would happen more or less everywhere else in the world. In 1922, Hoover assembled the first of four annual radio conferences, representing radio manufacturers, broadcasters, and some engineers and amateurs. This forum became Hoover's primary stage. Over the next four years, he used its annual meeting to derive policy recommendations, legitimacy, and cooperation for his regulatory action, all without a hint of authority under the Radio Act of 1912. Hoover relied heavily on the rhetoric of public interest and on the support of amateurs to justify his system of private broadcasting coordinated by the Department of Commerce. From 1922 on, however, he followed a pattern that would systematically benefit large commercial broadcasters over small ones; commercial broadcasters over educational and religious broadcasters; and the one-to-many broadcasts over the point-to-point, small-scale wireless telephony and telegraphy that the amateurs were developing. After January 1922, the department inserted a limitation on amateur licenses, excluding from their coverage the broadcast of "weather reports, market reports, music, concerts, speeches, news or similar information or entertainment." This, together with a Department of Commerce order to all amateurs to stop broadcasting at 360 meters (the wave assigned broadcasting), effectively limited amateurs to shortwave radiotelephony and telegraphy in a set of frequencies then thought to be commercially insignificant. In the summer, the department assigned broadcasters, in addition to 360 meters, another band, at 400 meters. Licenses in this Class B category were reserved for transmitters operating at power levels of 500-1,000 watts, who did not use phonograph records. These limitations on Class B licenses made the newly created channel a feasible home only to broadcasters who could afford the much-more-expensive, high-powered transmitters and could arrange for live broadcasts, rather than simply play phonograph records. The success of this new frequency was not immediate, because many receivers could not tune out stations broadcasting at the two frequencies in order to listen to the other. Hoover, failing to move Congress to amend the radio law to provide him with the power necessary to regulate broadcasting, relied on the recommendations of the Second Radio Conference in 1923 as public support for adopting a new regime, and continued to act without legislative authority. He announced that the broadcast band would be divided in three: high-powered (500-1,000 watts) stations serving large areas would have no interference in those large areas, and would not share frequencies. They would transmit on frequencies between 300 and 545 meters. Medium-powered stations served smaller areas without interference, and would operate at assigned channels between 222 and 300 meters. The remaining low-powered stations would not be eliminated, as the bigger actors wanted, but would remain at 360 meters, with limited hours of operation and geographic reach. Many of these lower-powered broadcasters were educational and religious institutions that perceived Hoover's allocation as a preference for the RCA-GE-AT&T-Westinghouse alliance. Despite his protestations against commercial broadcasting ("If a speech by the President is to be used as the meat in a sandwich of two patent medicine advertisements, there will be no radio left"), Hoover consistently reserved clear channels and issued high-power licenses to commercial broadcasters. The final policy action based on the radio conferences came in 1925, when the Department of Commerce stopped issuing licenses. The result was a secondary market in licenses, in which some religious and educational stations were bought out by commercial concerns. These purchases further gravitated radio toward commercial ownership. The licensing preference for stations that could afford high-powered transmitters, long hours of operation, and compliance with high technical constraints continued after the Radio Act of 1927. As a practical matter, it led to assignment of twenty-one out of the twenty-four clear channel licenses created by the Federal Radio Commission to the newly created network-affiliated stations.
Over the course of this period, tensions also began to emerge within the patent alliance. The phenomenal success of receiver sales tempted Western Electric into that market. In the meantime, AT&T, almost by mistake, began to challenge GE, Westinghouse, and RCA in broadcasting as an outgrowth of its attempt to create a broadcast common-carriage facility. Despite the successes of broadcast and receiver sales, it was not clear in 1922-1923 how the cost of setting up and maintaining stations would be paid for. In England, a tax was levied on radio sets, and its revenue used to fund the BBC. No such proposal was considered in the United States, but the editor of Radio Broadcast proposed a national endowed fund, like those that support public libraries and museums, and in 1924, a committee of New York businessmen solicited public donations to fund broadcasters (the response was so pitiful that the funds were returned to their donors). AT&T was the only company to offer a solution. Building on its telephone service experience, it offered radio telephony to the public for a fee. Genuine wireless telephony, even mobile telephony, had been the subject of experimentation since the second decade of radio, but that was not what AT&T offered. In February 1922, AT&T established WEAF in New York, a broadcast station over which AT&T was to provide no programming of its own, but instead would enable the public or program providers to pay on a per-time basis. AT&T treated this service as a form of wireless telephony so that it would fall, under the patent alliance agreements of 1920, under the exclusive control of AT&T. RCA, Westinghouse, and GE could not compete in this area. "Toll broadcasting" was not a success by its own terms. There was insufficient demand for communicating with the public to sustain a full schedule that would justify listeners tuning into the station. As a result, AT&T produced its own programming. In order to increase the potential audience for its transmissions while using its advantage in wired facilities, AT&T experimented with remote transmissions, such as live reports from sports events, and with simultaneous transmissions of its broadcasts by other stations, connected to its New York feed by cable. In its effort to launch toll broadcasting, AT&T found itself by mid-1923 with the first functioning precursor to an advertiser-supported broadcast network.
The alliance members now threatened each other: AT&T threatened to enter into receiver manufacturing and broadcast, and the RCA alliance, with its powerful stations, threatened to adopt "toll broadcasting," or advertiser-supported radio. The patent allies submitted their dispute to an arbitrator, who was to interpret the 1920 agreements, reached at a time of wireless telegraphy, to divide the spoils of the broadcast world of 1924. In late 1924, the arbitrator found for RCA-GE-Westinghouse on almost all issues. Capitalizing on RCA's difficulties with the antitrust authorities and congressional hearings over aggressive monopolization practices in the receiving set market, however, AT&T countered that if the 1920 agreements meant what the arbitrator said they meant, they were a combination in restraint of trade to which AT&T would not adhere. Bargaining in the shadow of the mutual threats of contract and antitrust actions, the former allies reached a solution that formed the basis of future radio broadcasting. AT&T would leave broadcasting. A new company, owned by RCA, GE, and Westinghouse would be formed, and would purchase AT&T's stations. The new company would enter into a long-term contract with AT&T to provide the long-distance communications necessary to set up the broadcast network that David Sarnoff envisioned as the future of broadcast. This new entity would, in 1926, become the National Broadcasting Company (NBC). AT&T's WEAF station would become the center of one of NBC's two networks, and the division arrived at would thereafter form the basis of the broadcast system in the United States.
By the middle of 1926, then, the institutional and organizational elements that became the American broadcast system were, to a great extent, in place. The idea of government monopoly over broadcasting, which became dominant in Great Britain, Europe, and their former colonies, was forever abandoned. The idea of a private-property regime in spectrum, which had been advocated by commercial broadcasters to spur investment in broadcast, was rejected on the backdrop of other battles over conservation of federal resources. The Radio Act of 1927, passed by Congress in record speed a few months after a court invalidated Hoover's entire regulatory edifice as lacking legal foundation, enacted this framework as the basic structure of American broadcast. A relatively small group of commercial broadcasters and equipment manufacturers took the lead in broadcast development. A governmental regulatory agency, using a standard of "the public good," allocated frequency, time, and power assignments to minimize interference and to resolve conflicts. The public good, by and large, correlated to the needs of commercial broadcasters and their listeners. Later, the broadcast networks supplanted the patent alliance as the primary force to which the Federal Radio Commission paid heed. The early 1930s still saw battles over the degree of freedom that these networks had to pursue their own commercial interests, free of regulation (studied in Robert McChesney's work).8 By that point, however, the power of the broadcasters was already too great to be seriously challenged. Interests like those of the amateurs, whose romantic pioneering mantle still held strong purchase on the process, educational institutions, and religious organizations continued to exercise some force on the allocation and management of the spectrum. However, they were addressed on the periphery of the broadcast platform, leaving the public sphere to be largely mediated by a tiny number of commercial entities running a controlled, advertiser-supported platform of mass media. Following the settlement around radio, there were no more genuine inflection points in the structure of mass media. Television followed radio, and was even more concentrated. Cable networks and satellite networks varied to some extent, but retained the basic advertiser-supported model, oriented toward luring the widest possible audience to view the advertising that paid for the programming.
The cluster of practices that form the mass-media model was highly conducive to social control in authoritarian countries. The hub-and-spoke technical architecture and unidirectional endpoint-reception model of these systems made it very simple to control, by controlling the core-the state-owned television, radio, and newspapers. The high cost of providing high-circulation statements meant that subversive publications were difficult to make and communicate across large distances and to large populations of potential supporters. Samizdat of various forms and channels have existed in most if not all authoritarian societies, but at great disadvantage relative to public communication. The passivity of readers, listeners, and viewers coincided nicely with the role of the authoritarian public sphere-to manage opinion in order to cause the widest possible willing, or at least quiescent, compliance, and thereby to limit the need for using actual repressive force.
In liberal democracies, the same technical and economic cost characteristics resulted in a very different pattern of communications practices. However, these practices relied on, and took advantage of, some of the very same basic architectural and cost characteristics. The practices of commercial mass media in liberal democracies have been the subject of a vast literature, criticizing their failures and extolling their virtues as a core platform for the liberal public sphere. There have been three primary critiques of these media: First, their intake has been seen as too limited. Too few information collection points leave too many views entirely unexplored and unrepresented because they are far from the concerns of the cadre of professional journalists, or cannot afford to buy their way to public attention. The debates about localism and diversity of ownership of radio and television stations have been the clearest policy locus of this critique in the United States. They are based on the assumption that local and socially diverse ownership of radio stations will lead to better representation of concerns as they are distributed in society. Second, concentrated mass media has been criticized as giving the owners too much power-which they either employ themselves or sell to the highest bidder-over what is said and how it is evaluated. Third, the advertising-supported media needs to attract large audiences, leading programming away from the genuinely politically important, challenging, and engaging, and toward the titillating or the soothing. This critique has emphasized the tension between business interests and journalistic ethics, and the claims that market imperatives and the bottom line lead to shoddy or cowering reporting; quiescence in majority tastes and positions in order to maximize audience; spectacle rather than substantive conversation of issues even when political matters are covered; and an emphasis on entertainment over news and analysis.
Three primary defenses or advantages have also been seen in these media: first is their independence from government, party, or upper-class largesse, particularly against the background of the state-owned media in authoritarian regimes, and given the high cost of production and communication, commercial mass media have been seen as necessary to create a public sphere grounded outside government. Second is the professionalism and large newsrooms that commercial mass media can afford to support to perform the watchdog function in complex societies. Because of their market-based revenues, they can replace universal intake with well-researched observations that citizens would not otherwise have made, and that are critical to a well-functioning democracy. Third, their near-universal visibility and independence enable them to identify important issues percolating in society. They can provide a platform to put them on the public agenda. They can express, filter, and accredit statements about these issues, so that they become well-specified subjects and feasible objects for public debate among informed citizens. That is to say, the limited number of points to which all are tuned and the limited number of "slots" available for speaking on these media form the basis for providing the synthesis required for public opinion and raising the salience of matters of public concern to the point of potential collective action. In the remainder of this chapter, I will explain the criticisms of the commercial mass media in more detail. I then take up in chapter 7 the question of how the Internet in general, and the rise of nonmarket and cooperative individual production in the networked information economy in particular, can solve or alleviate those problems while fulfilling some of the important roles of mass media in democracies today.
Mass Media as a Platform for the Public Sphere
The structure of mass media as a mode of communications imposes a certain set of basic characteristics on the kind of public conversation it makes possible. First, it is always communication from a small number of people, organized into an even smaller number of distinct outlets, to an audience several orders of magnitude larger, unlimited in principle in its membership except by the production capacity of the media itself-which, in the case of print, may mean the number of copies, and in radio, television, cable, and the like, means whatever physical-reach constraints, if any, are imposed by the technology and business organizational arrangements used by these outlets. In large, complex, modern societies, no one knows everything. The initial function of a platform for the public sphere is one of intake-taking into the system the observations and opinions of as many members of society as possible as potential objects of public concern and consideration. The radical difference between the number of intake points the mass media have and the range and diversity of human existence in large complex societies assures a large degree of information loss at the intake stage. Second, the vast difference between the number of speakers and the number of listeners, and the finished-goods style of mass-media products, imposes significant constraints on the extent to which these media can be open to feedback-that is, to responsive communications that are tied together as a conversation with multiple reciprocal moves from both sides of the conversation. Third, the immense and very loosely defined audience of mass media affects the filtering and synthesis functions of the mass media as a platform for the public sphere. One of the observations regarding the content of newspapers in the late eighteenth to mid-nineteenth centuries was the shift they took as their circulation increased-from party-oriented, based in relatively thick communities of interest and practice, to fact- and sensation-oriented, with content that made thinner requirements on their users in order to achieve broader and more weakly defined readership. Fourth, and finally, because of the high costs of organizing these media, the functions of intake, sorting for relevance, accrediting, and synthesis are all combined in the hands of the same media operators, selected initially for their capacity to pool the capital necessary to communicate the information to wide audiences. While all these functions are necessary for a usable public sphere, the correlation of capacity to pool capital resources with capacity to offer the best possible filtering and synthesis is not obvious. In addition to basic structural constraints that come from the characteristic of a communications modality that can properly be called "mass media," there are also critiques that arise more specifically from the business models that have characterized the commercial mass media over the course of most of the twentieth century. Media markets are relatively concentrated, and the most common business model involves selling the attention of large audiences to commercial advertisers.
Media Concentration: The Power of Ownership and Money
The Sinclair Broadcast Group is one of the largest owners of television broadcast stations in the United States. The group's 2003 Annual Report proudly states in its title, "Our Company. Your Message. 26 Million Households"; that is, roughly one quarter of U.S. households. Sinclair owns and operates or provides programming and sales to sixty-two stations in the United States, including multiple local affiliates of NBC, ABC, CBS, and Fox. In April 2004, ABC News's program Nightline dedicated a special program to reading the names of American service personnel who had been killed in the Iraq War. The management of Sinclair decided that its seven ABC affiliates would not air the program, defending its decision because the program "appears to be motivated by a political agenda designed to undermine the efforts of the United States in Iraq."9 At the time, the rising number of American casualties in Iraq was already a major factor in the 2004 presidential election campaign, and both ABC's decision to air the program, and Sinclair's decision to refuse to carry it could be seen as interventions by the media in setting the political agenda and contributing to the public debate. It is difficult to gauge the politics of a commercial organization, but one rough proxy is political donations. In the case of Sinclair, 95 percent of the donations made by individuals associated with the company during the 2004 election cycle went to Republicans, while only 5 percent went to Democrats.10 Contributions of Disney, on the other hand, the owner of the ABC network, split about seventy-thirty in favor of contribution to Democrats. It is difficult to parse the extent to which political leanings of this sort are personal to the executives and professional employees who make decisions about programming, and to what extent these are more organizationally self-interested, depending on the respective positions of the political parties on the conditions of the industry's business. In some cases, it is quite obvious that the motives are political. When one looks, for example, at contributions by Disney's film division, they are distributed 100 percent in favor of Democrats. This mostly seems to reflect the large contributions of the Weinstein brothers, who run the semi-independent studio Miramax, which also distributed Michael Moore's politically explosive criticism of the Bush administration, Fahrenheit 9/11, in 2004. Sinclair's contributions were aligned with, though more skewed than, those of the National Association of Broadcasters political action committee, which were distributed 61 percent to 39 percent in favor of Republicans. Here the possible motivation is that Republicans have espoused a regulatory agenda at the Federal Communications Commission that allows broadcasters greater freedom to consolidate and to operate more as businesses and less as public trustees.
The basic point is not, of course, to trace the particular politics of one programming decision or another. It is the relative power of those who manage the mass media when it so dominates public discourse as to shape public perceptions and public debate. This power can be brought to bear throughout the components of the platform, from the intake function (what facts about the world are observed) to the filtration and synthesis (the selection of materials, their presentation, and the selection of who will debate them and in what format). These are all central to forming the agenda that the public perceives, choreographing the discussion, the range of opinions perceived and admitted into the conversation, and through these, ultimately, choreographing the perceived consensus and the range of permissible debate. One might think of this as "the Berlusconi effect." Thinking in terms of a particular individual, known for a personal managerial style, who translated the power of control over media into his election as prime minister of his country symbolizes well the concern, but of course does not exhaust the problem, which is both broader and more subtle than the concern with the possibility that mass media will be owned by individuals who would exert total control over these media and translate their control into immediate political power, manufacturing and shaping the appearance of a public sphere, rather than providing a platform for one.
The power of the commercial mass media depends on the degree of concentration in mass-media markets. A million equally watched channels do not exercise power. Concentration is a common word used to describe the power media exercise when there are only few outlets, but a tricky one because it implies two very distinct phenomena. The first is a lack of competition in a market, to a degree sufficient to allow a firm to exercise power over its pricing. This is the antitrust sense. The second, very different concern might be called "mindshare." That is, media is "concentrated" when a small number of media firms play a large role as the channel from and to a substantial majority of readers, viewers, and listeners in a given politically relevant social unit.
If one thinks that commercial firms operating in a market will always "give the audience what it wants" and that what the audience wants is a fully representative cross-section of all observations and opinions relevant to public discourse, then the antitrust sense would be the only one that mattered. A competitive market would force any market actor simply to reflect the range of available opinions actually held in the public. Even by this measure, however, there continue to be debates about how one should define the relevant market and what one is measuring. The more one includes all potential nationally available sources of information, newspapers, magazines, television, radio, satellite, cable, and the like, the less concentrated the market seems. However, as Eli Noam's recent work on local media concentration has argued, treating a tiny television station on Long Island as equivalent to WCBS in New York severely underrepresents the power of mass media over their audience. Noam offered the most comprehensive analysis currently available of the patterns of concentration where media are actually accessed-locally, where people live-from 1984 to 2001-2002. Most media are consumed locally-because of the cost of national distribution of paper newspapers, and because of the technical and regulatory constraints on nationwide distribution of radio and television. Noam computed two measures of market concentration for each of thirty local markets: the Herfindahl-Hirschman Index (HHI), a standard method used by the Department of Justice to measure market concentration for antitrust purposes; and what he calls a C4 index-that is, the market share of the top four firms in a market, and C1, the share of the top single firm in the market. He found that, based on the HHI index, all the local media markets are highly concentrated. In the standard measure, a market with an index of less than 1,000 is not concentrated, a market with an index of 1,000-1,800 is moderately concentrated, and a market with an index of above 1,800 on the HHI is highly concentrated. Noam found that local radio, which had an index below 1,000 between 1984 and 1992, rose over the course of the following years substantially. Regulatory restrictions were loosened over the course of the 1990s, resulting by the end of the decade in an HHI index measure of 2,400 for big cities, and higher for medium-sized and small markets. And yet, radio is less concentrated than local multichannel television (cable and satellite) with an HHI of 6,300, local magazines with an HHI of 6,859, and local newspapers with an HHI of 7,621. The only form of media whose concentration has declined to less than highly concentrated (HHI 1,714) is local television, as the rise of new networks and local stations' viability on cable has moved us away from the three-network world of 1984. It is still the case, however, that the top four television stations capture 73 percent of the viewers in most markets, and 62 percent in large markets. The most concentrated media in local markets are newspapers, which, except for the few largest markets, operate on a one-newspaper town model. C1 concentration has grown in this area to 83 percent of readership for the leading papers, and an HHI of 7,621.
The degree of concentration in media markets supports the proposition that owners of media can either exercise power over the programming they provide or what they write, or sell their power over programming to those who would like to shape opinions. Even if one were therefore to hold the Pollyannaish view that market-based media in a competitive market would be constrained by competition to give citizens what they need, as Ed Baker put it, there is no reason to think the same in these kinds of highly concentrated markets. As it turns out, a long tradition of scholarship has also developed the claim that even without such high levels of concentration in the antitrust sense, advertiser-supported media markets are hardly good mechanisms for assuring that the contents of the media provide a good reflection of the information citizens need to know as members of a polity, the range of opinions and views about what ought to occupy the public, and what solutions are available to those problems that are perceived and discussed.11 First, we have long known that advertiser-supported media suffer from more or less well-defined failures, purely as market mechanisms, at representing the actual distribution of first-best preferences of audiences. As I describe in more detail in the next section, whether providers in any market structure, from monopoly to full competition, will even try to serve first-best preferences of their audience turns out to be a function of the distribution of actual first-best and second-best preferences, and the number of "channels." Second, there is a systematic analytic problem with defining consumer demand for information. Perfect information is a precondition to an efficient market, not its output. In order for consumers to value information or an opinion fully, they must know it and assimilate it to their own worldview and understanding. However, the basic problem to be solved by media markets is precisely to select which information people will value if they in fact come to know it, so it is impossible to gauge the value of a unit of information before it has been produced, and hence to base production decisions on actual existing user preferences. The result is that, even if media markets were perfectly competitive, a substantial degree of discretion and influence would remain in the hands of commercial media owners.
The actual cultural practice of mass-media production and consumption is more complex than either the view of "efficient media markets" across the board or the general case against media concentration and commercialism. Many of the relevant companies are public companies, answerable to at least large institutional shareholders, and made up of managements that need not be monolithic in their political alignment or judgment as to the desirability of making political gains as opposed to market share. Unless there is economic or charismatic leadership of the type of a William Randolph Hearst or a Rupert Murdoch, organizations usually have complex structures, with varying degrees of freedom for local editors, reporters, and midlevel managers to tug and pull at the fabric of programming. Different media companies also have different business models, and aim at different market segments. The New York Times, Wall Street Journal, and Washington Post do not aim at the same audience as most daily local newspapers in the United States. They are aimed at elites, who want to buy newspapers that can credibly claim to embody highly professional journalism. This requires separation of editorial from business decisions-at least for some segments of the newspapers that are critical in attracting those readers. The degree to which the Berlusconi effect in its full-blown form of individual or self-consciously directed political power through shaping of the public sphere will apply is not one that can necessarily be answered as a matter of a priori theoretical framework for all mass media. Instead, it is a concern, a tendency, whose actual salience in any given public sphere or set of firms is the product of historical contingency, different from one country to another and one period to another. It will depend on the strategies of particular companies and their relative mindshare in a society. However, it is clear and structurally characteristic of mass media that a society that depends for its public sphere on a relatively small number of actors, usually firms, to provide most of the platform of its public sphere, is setting itself up for, at least, a form of discourse elitism. In other words, those who are on the inside of the media will be able to exert substantially greater influence over the agenda, the shape of the conversation, and through these the outcomes of public discourse, than other individuals or groups in society. Moreover, for commercial organizations, this power could be sold-and as a business model, one should expect it to be. The most direct way to sell influence is explicit political advertising, but just as we see "product placement" in movies as a form of advertising, we see advertiser influence on the content of the editorial materials. Part of this influence is directly substantive and political. Another is the source of the second critique of commercial mass media.
Commercialism, Journalism, and Political Inertness
The second cluster of concerns about the commercial mass media is the degree to which their commercialism undermines their will and capacity to provide a platform for public, politically oriented discourse. The concern is, in this sense, the opposite of the concern with excessive power. Rather than the fear that the concentrated mass media will exercise its power to pull opinion in its owners' interest, the fear is that the commercial interests of these media will cause them to pull content away from matters of genuine political concern altogether. It is typified in a quote offered by Ben Bagdikian, attributed to W. R. Nelson, publisher of the Kansas City Star in 1915: "Newspapers are read at the breakfast table and dinner tables. God's great gift to man is appetite. Put nothing in the paper that will destroy it."12 Examples abound, but the basic analytic structure of the claim is fairly simple and consists of three distinct components. First, advertiser-supported media need to achieve the largest audience possible, not the most engaged or satisfied audience possible. This leads such media to focus on lowest-common-denominator programming and materials that have broad second-best appeal, rather than trying to tailor their programming to the true first-best preferences of well-defined segments of the audience. Second, issues of genuine public concern and potential political contention are toned down and structured as a performance between iconic representations of large bodies of opinion, in order to avoid alienating too much of the audience. This is the reemergence of spectacle that Habermas identified in The Transformation of the Public Sphere. The tendency toward lowest-common-denominator programming translates in the political sphere into a focus on fairly well-defined, iconic views, and to avoidance of genuinely controversial material, because it is easier to lose an audience by offending its members than by being only mildly interesting. The steady structuring of the media as professional, commercial, and one way over 150 years has led to a pattern whereby, when political debate is communicated, it is mostly communicated as performance. Someone represents a party or widely known opinion, and is juxtaposed with others who similarly represent alternative widely known views. These avatars of public opinion then enact a clash of opinion, orchestrated in order to leave the media neutral and free of blame, in the eyes of their viewers, for espousing an offensively partisan view. Third, and finally, this business logic often stands in contradiction to journalistic ethic. While there are niche markets for high-end journalism and strong opinion, outlets that serve those markets are specialized. Those that cater to broader markets need to subject journalistic ethic to business necessity, emphasizing celebrities or local crime over distant famines or a careful analysis of economic policy.
The basic drive behind programming choices in advertising-supported mass media was explored in the context of the problem of "program diversity" and competition. It relies on a type of analysis introduced by Peter Steiner in 1952. The basic model argued that advertiser-supported media are sensitive only to the number of viewers, not the intensity of their satisfaction. This created an odd situation, where competitors would tend to divide among them the largest market segments, and leave smaller slices of the audience unserved, whereas a monopolist would serve each market segment, in order of size, until it ran out of channels. Because it has no incentive to divide all the viewers who want, for example, sitcoms, among two or more stations, a monopolist would program a sitcom on one channel, and the next-most-desired program on the next channel. Two competitors, on the other hand, would both potentially program sitcoms, if dividing those who prefer sitcoms in half still yields a larger total audience size than airing the next-most-desired program. To illustrate this effect with a rather extreme hypothetical example, imagine that we are in a television market of 10 million viewers. Suppose that the distribution of preferences in the audience is as follows: 1,000,000 want to watch sitcoms; 750,000 want sports; 500,000 want local news; 250,000 want action movies; 9,990 are interested in foreign films; and 9,980 want programs on gardening. The stark drop-off between action movies and foreign films and gardening is intended to reflect the fact that the 7.5 million potential viewers who do not fall into one of the first four clusters are distributed in hundreds of small clusters, none commanding more than 10,000 viewers. Before we examine why this extreme assumption is likely correct, let us first see what happens if it were. Table 6.1 presents the programming choices that would typify those of competing channels, based on the number of channels competing and the distribution of preferences in the audience. It reflects the assumptions that each programmer wants to maximize the number of viewers of its channel and that the viewers are equally likely to watch one channel as another if both offer the same type of programming. The numbers in parentheses next to the programming choice represent the number of viewers the programmer can hope to attract given these assumptions, not including the probability that some of the 7.5 million viewers outside the main clusters will also tune in. In this extreme example, one would need a system with more than 250 channels in order to start seeing something other than sitcoms, sports, local news, and action movies. Why, however, is such a distribution likely, or even plausible? The assumption is not intended to represent an actual distribution of what people most prefer to watch. Rather, it reflects the notion that many people have best preferences, fallback preferences, and tolerable options. Their first-best preferences reflect what they really want to watch, and people are highly diverse in this dimension. Their fallback and tolerable preferences reflect the kinds of things they would be willing to watch if nothing else is available, rather than getting up off the sofa and going to a local café or reading a book. Here represented by sitcoms, sports, and the like, fallback options are more widely shared, even among people whose first-best preferences differ widely, because they represent what people will tolerate before switching, a much less strict requirement than what they really want. This assumption follows Jack Beebe's refinement of Steiner's model. Beebe established that media monopolists would show nothing but common-denominator programs and that competition among broadcasters would begin to serve the smaller preference clusters only if a large enough number of channels were available. Such a model would explain the broad cultural sense of Bruce Springsteen's song, "57 Channels (And Nothin' On)," and why we saw the emergence of channels like Black Entertainment Television, Univision (Spanish channel in the United States), or The History Channel only when cable systems significantly expanded channel capacity, as well as why direct-broadcast satellite and, more recently, digital cable offerings were the first venue for twenty-four-hour-a-day cooking channels and smaller minority-language channels.13
Table 6.1: Distribution of Channels Hypothetical
|No. of channels||Programming Available (in thousands of viewers)|
||sitcom (1000), sports (750)|
||sitcom (1000 or 500), sports (750), indifferent between sitcoms and local news (500)|
||sitcom (500), sports (750), sitcom (500), local news (500)|
||sitcom (500), sports (375), sitcom (500), local news (500), sports (375)|
||sitcom (333), sports (375), sitcom (333), local news (500), sports (375), sitcom (333)|
||sitcom (333), sports (375), sitcom (333), local news (500), sports (375), sitcom (333), action movies (250)|
||sitcom (333), sports (375), sitcom (333), local news (250), sports (375), sitcom (333), action movies (250), local news (250)|
||sitcom (250), sports (375), sitcom (250), local news (250), sports (375), sitcom (250), action movies (250), local news (250), sitcom (250)|
|* * *||
||* * *|
||100 channels of sitcom (10); 75 channels of sports (10); 50 channels of local news (10); 25 channels of action movies (10)|
||100 channels of sitcom (10); 75 channels of sports (10); 50 channels of local news (10); 25 channels of action movies (10); 1 foreign film channel (9.99)|
||100 channels of sitcom (10); 75 channels of sports (10); 50 channels of local news (10); 25 channels of action movies (10); 1 foreign film channel (9.99); 1 gardening channel (9.98)|
While this work was developed in the context of analyzing media diversity of offerings, it provides a foundation for understanding the programming choices of all advertiser-supported mass media, including the press, in domains relevant to the role they play as a platform for the public sphere. It provides a framework for understanding, but also limiting, the applicability of the idea that mass media will put nothing in the newspaper that will destroy the reader's appetite. Controversial views and genuinely disturbing images, descriptions, or arguments have a higher likelihood of turning readers, listeners, and viewers away than entertainment, mildly interesting and amusing human-interest stories, and a steady flow of basic crime and courtroom dramas, and similar fare typical of local television newscasts and newspapers. On the other hand, depending on the number of channels, there are clearly market segments for people who are "political junkies," or engaged elites, who can support some small number of outlets aimed at that crowd. The New York Times or the Wall Street Journal are examples in print, programs like Meet the Press or Nightline and perhaps channels like CNN and Fox News are examples of the possibility and limitations of this exception to the general entertainment-oriented, noncontroversial, and politically inert style of commercial mass media. The dynamic of programming to the lowest common denominator can, however, iteratively replicate itself even within relatively news- and elite-oriented media outlets. Even among news junkies, larger news outlets must cater relatively to the mainstream of its intended audience. Too strident a position or too probing an inquiry may slice the market segment to which they sell too thin. This is likely what leads to the common criticism, from both the Right and Left, that the same media are too "liberal" and too "conservative," respectively. By contrast, magazines, whose business model can support much lower circulation levels, exhibit a substantially greater will for political engagement and analysis than even the relatively political-readership-oriented, larger-circulation mass media. By definition, however, the media that cater to these niche markets serve only a small segment of the political community. Fox News in the United States appears to be a powerful counterexample to this trend. It is difficult to pinpoint why. The channel likely represents a composite of the Berlusconi effect, the high market segmentation made possible by high-capacity cable systems, the very large market segment of Republicans, and the relatively polarized tone of American political culture since the early 1990s.
The mass-media model as a whole, with the same caveat for niche markets, does not lend itself well to in-depth discussion and dialog. High professionalism can, to some extent, compensate for the basic structural problem of a medium built on the model of a small number of producers transmitting to an audience that is many orders of magnitude larger. The basic problem occurs at the intake and synthesis stages of communication. However diligent they may be, a small number of professional reporters, embedded as they are within social segments that are part of social, economic, and political elites, are a relatively stunted mechanism for intake. If one seeks to collect the wide range of individual observations, experiences, and opinions that make up the actual universe of concerns and opinions of a large public as a basic input into the public sphere, before filtering, the centralized model of mass media provides a limited means of capturing those insights. On the back end of the communication of public discourse, concentrated media of necessity must structure most "participants" in the debate as passive recipients of finished messages and images. That is the core characteristic of mass media: Content is produced prior to transmission in a relatively small number of centers, and when finished is then transmitted to a mass audience, which consumes it. This is the basis of the claim of the role of professional journalism to begin with, separating it from nonprofessional observations of those who consume its products. The result of this basic structure of the media product is that discussion and analysis of issues of common concern is an iconic representation of discussion, a choreographed enactment of public debate. The participants are selected for the fact that they represent well-understood, well-defined positions among those actually prevalent in a population, the images and stories are chosen to represent issues, and the public debate that is actually facilitated (and is supposedly where synthesis of the opinions in public debate actually happens) is in fact an already presynthesized portrayal of an argument among avatars of relatively large segments of opinion as perceived by the journalists and stagers of the debate. In the United States, this translates into fairly standard formats of "on the left X, on the right Y," or "the Republicans' position" versus "the Democrats' position." It translates into "photo-op" moments of publicly enacting an idea, a policy position, or a state of affairs-whether it is a president landing on an aircraft carrier to represent security and the successful completion of a controversial war, or a candidate hunting with his buddies to represent a position on gun control. It is important to recognize that by describing these characteristics, I am not identifying failures of imagination, thoughtfulness, or professionalism on the part of media organizations. These are simply characteristics of a mass-mediated public sphere; modes of communication that offer the path of least resistance given the characteristics of the production and distribution process of mass media, particularly commercial mass media. There are partial exceptions, as there are to the diversity of content or the emphasis on entertainment value, but these do not reflect what most citizens read, see, or hear. The phenomenon of talk radio and call-in shows represents a very different, but certainly not more reflective form. They represent the pornography and violence of political discourse-a combination of exhibitionism and voyeurism intended to entertain us with opportunities to act out suppressed desires and to glimpse what we might be like if we allowed ourselves more leeway from what it means to be a well-socialized adult.
The two basic critiques of commercial mass media coalesce on the conflict between journalistic ethics and the necessities of commercialism. If professional journalists seek to perform a robust watchdog function, to inform their readers and viewers, and to provoke and explore in depth, then the dynamics of both power and lowest-common-denominator appeal push back. Different organizations, with different degrees of managerial control, editorial independence, internal organizational culture, and freedom from competitive pressures, with different intended market segments, will resolve these tensions differently. A quick reading of the conclusions of some media scholarship, and more commonly, arguments made in public debates over the media, would tend to lump "the media" as a single entity, with a single set of failures. In fact, unsurprisingly, the literature suggests substantial heterogeneity among organizations and media. Television seems to be the worst culprit on the dimension of political inertness. Print media, both magazines and some newspapers, include significant variation in the degree to which they fit these general models of failure.
As we turn now to consider the advantages of the introduction of Internet communications, we shall see how this new model can complement the mass media and alleviate its worst weaknesses. In particular, the discussion focuses on the emergence of the networked information economy and the relatively larger role it makes feasible for nonmarket actors and for radically distributed production of information and culture. One need not adopt the position that the commercial mass media are somehow abusive, evil, corporate-controlled giants, and that the Internet is the ideal Jeffersonian republic in order to track a series of genuine improvements represented by what the new emerging modalities of public communication can do as platforms for the public sphere. Greater access to means of direct individual communications, to collaborative speech platforms, and to nonmarket producers more generally can complement the commercial mass media and contribute to a significantly improved public sphere.