Digital Divide
CIVIC ENGAGEMENT, INFORMATION POVERTY, AND THE INTERNET WORLDWIDE
Pippa Norris
Harvard University
CAMBRIDGE
UNTVERSITY PRESS
CHAPTER 1
The Digital Divide
The year 1989 dawned like any other but, in retrospect, it witnessed two major developments of immense historical significance. One was highly visible and widely celebrated: the symbolic dismantling of the Berlin Wall sparking the brushfire of electoral democracy spreading throughout the post-Communist world and beyond. The other was less generally recognized at the time, beyond a few scientific and technical cognoscenti: the invention of the World Wide Web. Dispersed computers communicating via packet-switching networks, and hence a rudimentary version of the Internet, had linked scientific elites for two decades. It took the invention of the Web by Tim Berners-Lee in CERN and the launch of a graphical browser, Mosaic, four years later to popularize this technology. Like a stone dropping into a pellucid pond, the ripples from this invention are surging throughout industrialized societies at the core, as well as flowing more slowly among developing societies at the periphery. With the size of the online community doubling every year, few doubt the potential importance of the Internet for transforming the way people live, work, and play. But, beyond these spheres, what are the causes of stratification in the networked world? In particular - the core focus of this book - will the Internet serve to reinforce or erode the gap between information-rich and poor nations? Will it exacerbate or reduce social divisions within countries? And will it strengthen representative democracy, as many hope, or will it buttress the power of established interests, as others fear?
In exploring these issues, this book focuses on understanding the root causes and the major consequences of inequalities evident during the first decade of the Internet age. The term "digital divide" has quickly become so popular as an instant sound bite that it has entered everyday
Digital Divide
The Digital Divide
speech as shorthand for any and every disparity within the online community. In this study the concept of the digital divide is understood as a multidimensional phenomenon encompassing three distinct aspects. The global divide refers to the divergence of Internet access between industrialized and developing societies. The social divide concerns the gap between information rich and poor in each nation. And finally within the online community, the democratic divide signifies the difference between those who do, and do not, use the panoply of digital resources to engage, mobilize, and participate in public life. To consider these matters, this introduction summarizes the contemporary debate about these issues, and then outlines the book's central argument, framework, and organization.
THE GLOBAL DIVIDE AMONG COUNTRIES
Few doubt the potential impact of digital technologies for reshaping the flow of investment, goods, and services in the global marketplace. Like the Californian Gold Rush of the 1850s, dot.corns have scrambled to stake their claims in the virtual frontier. Productivity and efficiency gains from investments in ICTs remain difficult to gauge but the U.S. Department of Commerce estimates that industries producing computer and communications hardware, software, and services have had a major impact on the U.S. economy.1 These developments fueled an intense flurry of heady speculation about the emergence of a "new" economy breaking the traditional business rules, although, mirroring the fluctuating fortunes of the Nasdaq index and the death of hundreds of dot.com start-ups, more cautious voices have subsequently warned that beyond a few isolated sectors, such as the travel or insurance industries, "bricks and mortar" assets still count for successful business-customer relations, along with old-fashioned notions such as profitability for investors, brand names, sales, and distribution systems.2
In the social sphere, few question the significance of cyberculture for transforming leisure hours, community networks, and personal lifestyles.3 Thousands of Internet sites and over 2 billion web pages cater to every conceivable interest from acupuncture to zoology.'1 Within a decade of its launch, America has become all Internet, all the time. The public has also flooded online in comparable countries such as Canada, Sweden, and Australia.5 The Internet population surged from about 3
million worldwide users in 1994 to more than 400 million in late-2000.6 Yet the potential for this medium, currently reaching about 7 percent of the worlds population, has only started to be exploited. Despite some indications of a possible slowdown in sales of personal computers in the saturated U.S. market, connectivity seems likely to gain momentum in the near future: Metcalf's law suggests that the value of a network is proportional to the square number of people using it: the more people link to the Internet, the greater its utility, the more it attracts.7
But what has been, and what will be, the impact of digital technologies on poorer countries? Surf at random, click on this, click on that, and whose voices do you hear around the globe? There are many plausible reasons why the emerging Internet age may reinforce disparities between postindustrial economies at the core of the network and developing societies at the periphery.8 As many warn, the basic problem is '"lb them that hath shall be given". If investment in digital technologies has the capacity to boost productivity, advanced economies such as Sweden, Australia, and the United States at the forefront of the technological revolution may be well placed to pull even farther ahead, maintaining their edge in future decades. A few middle-level economies like Taiwan, Brazil, and South Korea may manage to leverage themselves profitably into niche markets within the global marketplace, servicing international corporations based elsewhere by providing software development or manufacturing silicon chips. But most poorer societies, lagging far behind, plagued by multiple burdens of debt, disease, and ignorance, may join the digital world decades later and, in the long-term, may ultimately fail to catch up.9
International organizations have sounded the alarm. The OECD warns that affluent states at the cutting edge of technological change have reinforced their lead in the new knowledge economy but so far the benefits of the Internet have not yet triclded down far to Southern, Central, and Eastern Europe, let alone to the poorest areas in Sub-Saharan Africa, Latin America, and Southeast Asia.10 The UN Development Report argues that productivity gains from information technologies may widen the chasm between the most affluent nations and those that lack the skills, resources, and infrastructure to invest in the information society: "The network society is creating parallel communications systems: one for those with income, education and literally connections, giving plentiful information at low cost and high speed; the other for those without connections, blocked by high barriers of time, cost
Digital Divide
and uncertainty and dependent upon outdated information."11 Echoing these concerns, UNESCO emphasizes that most of the world's population lack basic access to a telephone, let alone a computer, producing societies increasingly marginalized at the periphery of communication networks.12 Leaders in the World Bank, European Union, United Nations, and G-8 have highlighted the problems of exclusion from the knowledge economy, where know-how replaces land and capital as the basic building blocks of growth.13 Initiatives have been launched to address this problem but disparities in the distribution of information and communication technologies are deep seated, suggesting that they will not easily be eradicated or ameliorated. The global flow of such traditional media as news, books, or scholarly research has long displayed center-periphery inequalities, with information flowing primarily from north to south; an issue generating heated debate during the 1980s centered on UNESCO's controversial New World Information Order.14 Technology has always held promise as an engine of economic growth for transforming developing nations - including machines for printing, textiles manufacture, and iron railways in the nineteenth century, and automobiles, oil production, and television in the twentieth - but critics argue that in practice this promise has often mainly served to benefit the industrialized world.15
Yet at the same time if technological diffusion can be achieved in poorer societies, and it is a big "if," then many observers hope that the Internet provides multiple opportunities for socioeconomic and democratic development. Digital networks have the potential to broaden and enhance access to information and communications for remote rural areas and poorer neighborhoods, to strengthen the process of democratization under transitional regimes, and to ameliorate the endemic problems of poverty in the developing world. With connectivity as the umbilical cord, enthusiasts hope that the Internet will eventually serve multiple functions as the world's favorite public library, school classroom and medical database, post office and telephone, marketplace and shopping mall, channel for entertainment, culture and music, daily news resource for headlines, stocks and weather, and heterogeneous global public sphere. In the heady words of the G-8 Okinawa Charter: "Our vision of an information society is one that better enables people to fulfill their potential and realize their aspirations. To this end we must ensure that IT serves the mutually supportive goals of creating sustainable economic growth, enhancing the
The Digital Divide
public welfare, and fostering social cohesion, and work to fully realize its potential to strengthen democracy, increase transparency and accountability in governance, promote human rights, enhance cultural diversity, and to foster international peace and stability."16 The Internet may allow societies to leapfrog stages of technological and industrial development. On the production side, if Bangalore companies can write software code for IBM or Microsoft, and if Costa Rica can manufacture chips for Intel, then potentially entrepreneurs can offer similar services from Malaysia, Brazil, and South Africa. The Internet encourages market globalization: small craft industries and the tourism industry in Bali or the Maldives can deal directly with customers and vacationers in New York and London, irrespective of distance, the costs of advertising, and the intermediate distribution chains of travel ■ agents and retail businesses.17 The Internet also offers promise for the delivery of basic social services such as education and health information across the globe, a function that may be particularly important for middle-level professionals serving their broader community.18 Local teachers or community officials connected to the digital world in Lagos, Beijing, or Calcutta can access the same electronic journals, books, and databases as students at the Sorbonne, Oxford, or Harvard. Distance learning can widen access to training and education, via open universities in India, Africa, and Thailand, and language websites for schools.19 Networks of hospitals and health care professionals in the Ukraine, Mozambique, and Stockholm can pool expertise and knowledge about the latest research on AIDS. Peasant farmers using village community centers can learn about storm warnings and market prices for their crops, along with employment opportunities in local towns. Where peripheral regions lack access to the traditional media, the convergence of communication technologies means that the Internet has the potential to deliver virtual local newspapers, streaming radio and television video, as well as other services.
It is hoped that within a few years many of the existing barriers to Jccess will be overcome with the combination of technological breakthroughs, market competition, and state initiatives. Internet has usually been delivered via bulky desktop personal computers tethered to telephone wires, but multiple less expensive devices are rapidly facilitating wireless access, including NTT's DoCoMo mobile phones using I-mode in Japan, Nokia's Communicator using WAP-enabled services m Europe,  and  handheld  personal  digital  assistants such as
Digital Divide
Handspring and Palm Pilots which are popular in the United States.20 Prototype disposable prepaid cell phones and laptops are under development, along with speech-recognition software and voice-activated Internet services. The price of hardware, software, and services has been plummeting, owing to increased competition in telecommunications combined with computer technologies' falling costs, faster speeds, and smaller microprocessors.25 In the 1960s Intel founder Gordon Moore predicted that, for the foreseeable future, chip density, and hence computing power, would double every eighteen months while costs would remain constant. During the last thirty years "Moore's law" has proved remarkably prescient. Every eighteen months, you can get twice as much power for the same cost. Telecommunications bandwidth, the speed at which data can be moved through the phone network, is experiencing similarly dramatic improvements owing to high-speed fiber-optic cable, satellites, and wireless communication technologies, all of which can be used on the same network. There have been parallel developments with computer memory and storage devices such as rewritable CD-ROMs. In 1980, a gigabyte of storage cost several hundred thousand dollars and occupied a room. It now fits on a credit-card device that can be carried in your pocket. As well as technological innovations, public-sector initiatives in developing countries as diverse as Estonia, Costa Rica, and Bangladesh have promoted the infrastructure, skills training, and knowledge necessary to widen use of digital technologies.
The implications of these developments promise to sweep well beyond the economic sphere. Observers hope that digital technologies will shift some of the global disparities in power as well as wealth, by fostering a worldwide civic society countering the role of international agencies, strengthening the voice of the developing world, dissolving some of the boundaries of the nation-state, and reinforcing the process of democratization.22 By directly linking political activists in different countries, and reducing the costs of communication and networking, the Internet may foster new types of mobilization by transnational advocacy networks around the world.23 By connecting disparate social movements, coalitions can be formed that mobilize a global civic society, such as protestors concerned about the World Trade Organization meetings in Seattle and Washington, D.C., the anti-landmine campaign, the anti-sweatshop manufacture of Nike shoes, and opposition movements in Burma, linking indigenous groups in developing societies with
The Digital Divide
I
a diverse melange of Norwegian environmentalists, Australian trade unionists, and European human rights organizations.2'1 The Internet [-:       may facilitate the networking and mobilizing functions of NGOs work-S '      ing across national borders, as a countervailing force to the influence of p:-::     technocratic elites and government leaders running traditional interna-!/.'.'     tional organizations.25 The role of the Internet may be even more important as a force for human rights, providing a global platform for opposition movements challenging autocratic regimes and military dic-liE:   tatorships, despite government attempts to restrict access in countries :        like China and Cuba.26 Therefore many observers have emphasized that lllfc   the emerging years of the Internet Age have generated substantial worldwide inequalities in access and use although, if this could be overcome, it is widely believed that digital technologies will provide multiple opportunities for development.
The role of technology has therefore fueled a debate among opti-llll-   mists envisaging the positive role of the Internet for transforming : -      poverty in developing societies, skeptics who believe tirat new technologies alone will make little difference one way or another, and pessimists who emphasize that digital technologies will further exacerbate '. the existing North-South divide. This debate generates a series of ques-
tions that will be considered in this book. Today which nations around the globe are digital leaders and laggards? What explains variations across countries in Internet use, in particular is it levels of socioeconomic development, investments in human capital, the process of lill democratization, or something else? Does the Internet create new inequalities, or reinforce existing divisions evident for decades in the spread of old communication technologies? Attempts to move beyond speculative theorizing about these questions face major challenges. The World Wide Web remains in its adolescence; any examination of trends is limited to just a decade. Technology continues to evolve rapidly, along with its social uses, so that projected estimates are often rapidly overtaken by events. Yet, despite the need for considerable caution in weighing the available evidence, if we can establish the main drivers behind the diffusion of the Internet, and if these prove similar to the reasons behind the adoption of older forms of information technologies, then we are in a much better position to understand and predict the probable pattern of future developments, the potential consequences of the rise of the Internet age, and also the policy initiatives most likely to overcome the global divide.
8
9
Digital Divide
The Digital Divide
SOCIAL STRATIFICATION WITHIN COUNTRIES
I _
Equally important, many official agencies have expressed concern : about the development of a widening digital divide within societies. | Technological opportunities are often unevenly distributed, even in | nations like Australia, the United States, and Sweden at the forefront f.„ of the information society. As the Internet has become increasingly central to life, work, and play - providing job opportunities, strengthening community networks and facilitating educational advancement - it becomes even more important if certain groups and areas are systematically excluded, such as poorer neighborhoods, working-class households, or peripheral rural communities. Governments in many countries have recognized this issue and developed initiatives designed to tackle this potential problem. The EU prioritized social inclusion as one of the three key objectives when launching the e-Europe Action Plan in Lisbon in March 1999.27 In the United States, a series of studies by the Department of Commerce, Falling Through the Net, have emphasized lower rates of Internet penetration among the poor.28 The 1998 survey found that affluent households (with income of $75,000 and above) were twenty times as likely to have Internet access as those at the lowest income levels, and more than nine times as likely to have computer access.29 In February 2000, President Clinton expressed concern about this situation and proposed a new plan to help bridge the "digital divide," offering private companies a $2 billion tax break, new teacher training programs, and the development of Community Technology Centers in low-income neighborhoods to help close the gap so that the Internet eventually becomes as ubiquitous as the availability of the telephone or television.30 The Department of Commerce has headed this initiative, emphasizing the role of programs to widen public access, promote digital skills, and encourage content that will empower underserved communities. The most common policy strategy has been to wire classrooms, although some warn that by itself this may be insufficient to close the digital divide.31 The survey in August 2000 found that many groups that have traditionally lacked digital opportunities have been malcing substantial gains in connectivity and computer ownership, with the rising Internet tide carrying many boats. Nevertheless notable divides in Internet penetration still exist between Americans with different levels of income and education, different racial and ethnic groups, old
and young, single and dual-parent families, and those with and without disabilities.32 Many industry leaders in the corporate sector have expressed concern that too many people are being left behind in the Information Age, and multiple nonprofit organizations and foundations have highlighted this problem.33 Governments in Finland, Germany, Canada, and Sweden have all announced programs to address access inequalities, often blending private and public resources. The British government, for example, has established a network of city learning centers, introduced a scheme to distribute reconditioned computers to homes in poor neighborhoods, and developed a national grid linking all public libraries to the Internet.34 Will digital inequalities prove a temporary problem that will gradually fade over time, as Internet connectivity spreads and "normalizes," or will this prove an enduring pattern generating a persistent division between info-haves and have-nots? Again the debate divides cyber-pessimists who emphasize deep-seated patterns of social stratification and the growth of an unskilled underclass in technological access, cyber-skeptics who believe that technologies adapt to society, not vice versa, and cyber-optimists who hope that in affluent postin-dustrial societies, at least, the digital divide will eventually succumb to the combined forces of technological innovations, markets, and the state. Positive scenarios suggest that inequalities in Internet access may prove a short-term phenomenon, similar to the type of households that could afford to buy television sets when services were first introduced in the early 1950s. In this perspective, the profile of the online community will probably come to reflect society as a whole given the wider availability of simpler and cheaper plug-and-play technologies and faster broadband services, facilitating delivery of popular mass entertainment including streaming video-on-demand. Some suggest that high-tech companies will compete to connect the public with a speed and efficiency that no government program can match, even in the neighborhoods of the urban poor, if there is mass demand for the services.35 For those with personal computers, free Internet services, email and Web hosting services are already widely available, albeit with advertising strings attached.36 The market may be insufficient to close the gap but the nonprofit sector has also been active. Major American corporations including Microsoft, Intel, Hewlett-Packard, and AT&T have foundations devoted to expanding access to local communities, most often through donating educa-
10
II
Digital Divide
The Digital Divide
tional equipment and fostering training in deprived areas, complementing state initiatives designed to furnish the younger generation with keyboard skills and training in wired schools. Telecommunications policy may play an important role here if the Internet is treated as a public utility, so that access is made widely available through public libraries, community centers, and private homes, much as telephone services were regulated to produce low-cost services and universal access to rural areas.37
The interesting question is not whether there will be absolute social inequalities in Internet access; of course there will be, as in other dimensions of life. Although Alexander Graham Bell's commercial telephone service was launched in the United States in 1877, today in America, more than a century later, there remain pockets of racial inequality in access to household telephones. Cable TV started to become available in the mid-1960s but today, owing to choice or necessity, only two-thirds of American households are connected, along with about half of all households in industrialized nations.38 Given substantial inequalities in the old mass media, it would be naive to expect that the Internet will magically transcend information poverty overnight. The more intriguing series of questions addressed by this book concern whether there are special barriers to digital technologies, such as their greater complexity or costs, and whether relative inequalities in Internet use will be similar to disparities in the penetration rates of older communication technologies.
THE DEMOCRATIC DIVIDE
The last challenge, and perhaps the most intractable, concerns the potential impact of the digital world on the distribution of power and influence in political systems. Even if we assume, for the sake of argument, that Internet penetration rates will gradually widen throughout society there is growing awareness that a substantial democratic divide may still exist between those who do and do not use the multiple political resources available on the Internet for civic engagement. What will be the impact of digital technologies in the public sphere?
The Internet has generated deeply contested alternative visions about the future. Cyber-optimists emphasize the Panglossian possibilities of the Internet for the involvement of ordinary citizens in direct democracy. Digital technologies hold promise as a mechanism facili-
SStating alternative channels of civic engagement such as political chat ■0oms, electronic voting in general elections and for referenda issues, and the mobilization of virtual communities, revitalizing levels of fjll*    mass participation in public affairs.39 The use of the Internet by f*$ffgroups and social movements is often believed to exemplify digital .....^politics. This view was popular in the mid-1990s and the revolution-
' 'arT Potentia^ °f digital technologies continues to be expressed by iyjj ^-many enthusiasts such as George Gilder.40 Yet as the Internet evolved, a darker vision has been articulated among cyber-pessimists who - regard digital technology as a Pandora's box unleashing new inequali-' ties of power and wealth, reinforcing deeper divisions between the information rich and poor, the tuned-in and the tuned-out, the "activists and the disengaged. This account stresses that the global and social divides already discussed mean that Internet politics will disproportionately benefit the elite.'11 In this perspective, despite the potential for technological innovations, traditional interests and established authorities have the capacity to reassert their control in the virtual political sphere, just as traditional multinational corporations have the ability to reestablish their predominance in the world of '  e-cornmerce.42 Finally, cyber-skeptics argue that both these visions are exaggerated, because so far the potential of the Internet has not had a dramatic impact on the practical reality of "politics as usual," for good or ill, even in countries at the forefront of digital technologies.43 For ■■-■■■- example, during the 2000 American presidential campaign the major candidates used their Web pages essentially as glossy shop-windows, as fund-raising tools, and as campaign ads, rather than as interactive "bottom-up" formats  for  public  comment  and discussion.44 Technology, in this view, is a plastic medium that flows into and adapts to preexisting social molds. The demise of many dot.coms in [■the business world has reinforced the skeptical view.
Each of these viewpoints reflects an element of truth depending, like a Rorschach test, on whether studies are focusing on different multifaceted components of digital technologies. As with the blind men of Indostan in Hindu legend, observers touch different parts of an elephant - the tusks, the tail, and the trunk - and report their experience with absolute conviction as though describing the whole of the digital world.45 Yet it requires a considerable stretch to get our arms around this beast. Multiple warnings should be posted before entering this territory. Deep-rooted hopes and fears about the poten-
ans
[2
13
Digital Divide
tial for technology often outweigh dispassionate analysis. Powerful myths and vivid anecdotes commonly appear as plausible as concrete observations. The best forecasts often seem little more than intelligent guesses. "Facts" commonly exhibit a shelf life of weeks or months. And hucksters in the guise of market research hype the industry's wares.
THE CORE ARGUMENT, CONCEPTUAL FRAMEWORK, AND STRUCTURE OF THE BOOK
How can we move beyond speculative theorizing toward more systematic evidence on these issues? The overall structure of the book can be summarized as follows. Chapter 2 considers approaches to understanding the causes and consequences of Internet access and use, and the major challenges that arise owing to the rapid pace of technological and social development, the limitations of cross-national comparative evidence, and the need for a multimethod research design. The chapter concludes that the most effective way to meet these challenges is to develop a comparative multilevel research design covering a wide range of political system.
In this study, the conceptual framework used to understand these issues distinguishes among three nested levels of analysis, as illustrated in Figure 1.1. The national context, including the macrolevel technological, socioeconomic, and political environment, determines the diffusion of the Internet within each country. The institutional context of the virtual political system provides the structure of opportunities mediating between citizens and the state, including the use of digital information and communication technologies by governments and civic society. Finally, the individual or microlevel of resources and motivation determines who participates within the virtual political system. Most studies arc limited to only one level. In contrast the more holistic approach used in this book compares the national context of Internet access in 179 countries around the globe, as well as the virtual political system within these nations, and then explores patterns of online civic engagement among individual citizens in Western Europe and the United States. The nested framework assumes that the national context, such as the process of technological diffusion, influences the development of the virtual political system. In turn, the core institutions of the political system available in the digital world provide the
national Context
The Digital Divide Political Institutions
individual-level
Socioeconomic Development
—A
Technological Diffusion
Online ICitizensl
Groups 1—,
Parties
Media
Executive Legislature Judiciary
Resources
Online Civic Engagement
Motivation
Figure 1.1. The Internet Engagement Model.
systematic context within which individual citizens have opportunities to participate online. Which particular citizens choose to take advantage of these opportunities is determined by their personal resources (like time, money, and skills) and their motivation (like interest, confidence, and efficacy).
Those preferring to go directly to the meat and potatoes of the evidence can turn directly to Chapter 3 which establishes what we know about the global divide in the networked world, drawing upon aggregate indicators to map the spread of digital technology around the globe, and then considers the causes of cross-national differences in Internet connectivity. The evidence indicates that some developing nations such as Malaysia, Brazil, and Taiwan have made substantial progress in the knowledge economy. But average rates of Internet penetration have grown sluggishly, at best, in most developing nations. Chapter 3 demonstrates that the global divide in Internet access is substantial and expanding: About 87 percent of people online live in postindustrial societies.46 The contrasts worldwide are sharp: : More than half of all Americans now surf the Internet compared with 0.1 percent of Nigerians. There are currently twice as many users in Sweden than across the vast continent of Sub-Saharan Africa.''7 In considering alternative explanations of this phenomenon, the evidence strongly suggests that economic development is the main factor driving access to digital technologies, so that the Internet reflects and reinforces traditional inequalities between rich and poor societies. Once we control for levels of economic development, then democrati-
14
15
Digital Divide
zation plays an insignificant role in the process of technological diffusion. Far from a new pattern, the global spread of the Internet reflects existing patterns of access to the traditional mass media including television, newspapers, and radios, disparities that have existed for decades and that show no sign of gradually closing over time. Striking inequalities are evident worldwide: Half a billion people living in Sub-Saharan Africa share 14 million phone lines, fewer than in Manhattan or in Tokyo.48 In Sub-Saharan Africa, for every 100 people there are only 17 radio sets, 5 televisions, and 0.5 percent mobile phones.49 On this basis it seems likely that, despite initiatives by state and international agencies, and despite technological developments in the marketplace, the global digital divide will probably continue in the foreseeable future, driven by world poverty, even if new forms of Internet transmission eventually become inexpensive and as easy as pushing the power button on a radio.
Chapter 4 goes on to analyze the extent and the causes of .social inequalities in digital opportunities within different countries, focusing on Internet penetration rates broken down by social class, education, gender, and generation. The composition of the online population is analyzed using representative surveys in Western Europe and the United States. The study concludes that unequal rates of Internet penetration are due to deep divisions of social stratification within postindustrial societies - such as patterns of household income, education, and occupational status - that shape not just digital opportunities but also access to other common forms of mass communications including cable and satellite television, VCRs, and fax machines. Far from narrowing as the information society expands, the income gap in Internet penetration is currently greatest in societies such as Sweden and the Netherlands where access to digital technologies has become most widespread. Of course considerable caution is needed in projecting from current patterns to future trends. The rosy scenario suggests that digital opportunities could eventually become more socially inclusive under certain conditions: if costs continue to fall dramatically in the marketplace, if the technology becomes simplified, and if policy initiatives by the state widen Internet access, training, and keyboard skills. Through inexpensive cell phones or handheld personal assistants, use of a stripped-down version of the Internet, for example just email and some headline news services could eventually become as ubiquitous in postindustrial societies as the availability of household television sets. Genre-scrambling
£_j>_ .. 1 he Digital Divide
■ sss; technologies converging broadband access, the Internet, telephony, and TV entertainment promise to alter conventional forms of content deliv-f"-'    ery and also inputting devices. The long-term process of generational t    replacement should eventually lead to greater familiarity with computers throughout society.
J*ut m me short-term these rosy projections, while not impossible, jtc    involve multiple "ifs." At present, affluent households with multiple T consumer durables designed for traditional forms of home entertain-|i -   ment and communications are also most likely to possess networked " .    personal computers. Poorer families are excluded from digital oppor-^ "    tunities, and hence access to online employment vacancies, educa-* '     tional resources, and social networks. Moreover, even if basic access to if;:-:;   email becomes ubiquitous, say as common as public telephones in Europe and North America, the marketplace for technological inno-r       vations will continue to generate ever faster, smaller, and better machines, spawning new applications and multiple levels of function-.        ality. The chameleon-like capacity of digital technologies to morph, converge, and reappear in different guises, as cell phones can play music files, personal digital assistants can take photos, and computers carry radio waves, makes the Internet dissimilar to earlier machines like television sets. Even if the basic digital divide shrinks gradually over time, it is naive to believe that the virtual world can overturn fundamental inequalities of social stratification that are endemic throughout postindustrial societies, any more than it is likely to over-:    come world poverty.
THE VIRTUAL POLITICAL SYSTEM
Part II compares the institutional context for representative democracy focusing on three issues: Where and what type of political organizations worldwide have adapted to digital technologies? What are the functions of these websites for maximizing transparent information and interactive communications? And what explains the rise of digital politics, in particular the relative significance of socioeconomic development, technological diffusion, and the process of democratization? Chapter 5 expands upon theories of cyber-democracy and considers the potential capacity of the Internet for strengthening civic society and the institutions of representative democraq' around the world. Although many
16
17
Digital Divide
The Digital Divide
specific case studies describing cyber-politics in particular nations are becoming available, and a burgeoning literature is developing in the United States and Western Europe, it remains difficult to find systematic typologies and evidence comparing digital politics across a wide range of countries at different levels of social, economic, and political development. Subsequent chapters compare the way that the institutions of representative democracy have responded to digital politics, drawing on evidence about the distribution and function of websites for different types of political organizations from around the world. Chapters 6 through 9 analyze which countries have forged ahead in digital politics, where the Internet has been used for information and communication by governments and civic society, and die socioeconomic, technological, and political factors driving the adaptation of organizations to digital politics. As noted earlier, there are many reasons to be cautious in any analysis. The first decade of the emerging Internet age has seen a process of restructuring and adaptation as political institutions have learned what does, and doesn't, work using digital technologies. Yet precisely because this is a period of experimental transition and institutional change it is particularly important to draw the appropriate lessons based on the available evidence, to map the current state of play, and to consider how the Internet functions in a wide range of political systems, including but also beyond the United States and Western Europe.
The optimistic claims that the interactive capacities of digital technologies will facilitate a new era of direct democracy, characterized by widespread citizen deliberation in affairs of state, like a virtual Agora, while attractive as a normative ideal, is ultimately implausible in practice as soon as we understand who becomes involved in digital politics. As we will see, the cross-national survey evidence indicates that those who take advantage of the opportunities for electronic civic engagement are activists most likely to participate via conventional channels. As a medium of choice par excellence, it seems improbable that digital politics will reach the disengaged, the apathetic, and the uninterested, if they choose to spend their time and energies on multiple alternative sites devoted to everything from the stock market to games and music. In this regard, die Internet seems analogous to die segmented magazine market, where some subscribe to The Atlantic Monthly, The Economist, and Foreign Affairs, but others pick Golfing Weekly or Playboy. The available studies of politically oriented discussion groups, bulletin boards, and online chat rooms have found these largely fail as deliberative fora,
IIP
WSSI
IPttl
- h
instead serving as places to reinforce like-minded voices.50 Claims for the potential of digital direct democracy to revitalize mass participation can find few crumbs of support from these studies. At the same time, die skeptics' claim that nothing much will change in the political system, as most established political institutions will adapt digital technologies to facilitate existing functions, while admittedly more realistic, * overlooks the occasional indications that, here and there, now and then, Tike a faint sporadic seismic tremor, some disruptive threats to politics as usual are already becoming evident.
Rejecting the view that either everything will change as direct democracy comes to replace representative governance, or that nothing will change as the digital world merely replicates "politics as usual," this book argues that digital technologies have the capacity to strengthen the institutions of civic society mediating between citizens and the state. Established political institutions, just like major corporations, can be expected to adapt the Internet to their usual forms of communication, providing information online, but not reinventing s'themselves or rethinking their core strategy in the digital world, unless successfully challenged. In contrast, insurgent organizations traditionally have fewer political assets, fewer traditional advantages, but also fewei inhibitions about adapting flexibly to the opportunities for information and communication via the Internet. If this account is essentially correct, digital politics may have most impact in leveling the . playing field, not completely but at least partially, for a diverse range of challengers, such as transnational advocacy networks, alternative social ■; movements, protest organizations and minor parties, such as those concerned with environmentalism, globalization, human rights, world . trade, conflict resolution, and single-issue causes from all shades of the political spectrum, ranging from genetically modified food and anti-.: fuel taxes to animal rights and anti-sweat shops. The Internet does not drive these movements - these causes are triggered by deeper passions .: - but it facilitates their organization, mobilization, and expression.51 Information and the mechanisms for delivering it are the Hfeblood and sinews of the body politic. Some power comes out of the barrel of a gun. Some power can be bought with the resources of wealth and income. Some may be inherited by sultans and princelings. But in democratic systems the primary coinage of the realm - the resource that persuades, that influences, that swings votes - is information. Information" comes in all shapes and forms, from the publication of
18
19
Digital Divide
official documents by government departments to brief news bulletins on the hour, from lengthy parliamentary debates to 30-second campaign ads, and from demonstrations by new social movements to informal conversations over the water cooler. Political organizations are essentially designed as control systems for the transmission of information, binding together the activities of all members within the unit and communicating priorities to the external world. Some information exchanges are brief and transitory; others use rich and well-developed channels. The explosive growth of connectivity via the Internet alters the transmission of information among networks, shrinking costs, maximizing speed, broadening reach, and eradicating distance. Potentially these changes can have profound consequences for altering the balance of resources and power between outsider challengers and established organizations within the political system. Hierarchical communication channels, typical in bureaucratic organizations like government departments and international agencies, are less effective and slower mechanisms of information transmission than horizontal networks shared by informal coalitions of alternative social movements. National boundaries to information flows dissolve, allowing global networks to flourish. Independent upstarts and multiple sources of "news " where immediacy outweighs authority, threaten the legitimacy of traditional journalism in the newspapers and television. Communication costs fall, and information costs plummet even faster. With wider and easier access to official sources, opposition groups and social movements can challenge the authority and expertise of government ministers, civil servants, and elected officials on their own turf.
The main democratic potential of digital information and communication technologies lies in strengthening organizational linkages and networking capacities in civic society. Strengthening these bonds, it will be argued, has the capacity to produce sudden disruptions to politics as usual, especially for flash coalitions mobilizing suddenly like a guerrilla army then dissolving again, exemplified by events such as the anticapi-talism violent protest in the City of London in June 1999, direct-action campaigns against the World Trade Organization on the streets of Seattle and Quebec, antiglobalization protests against the World Bank/International Monetary Fund in Prague and Washington, D.C., and the poujadist fuel price revolt by farmers and truckers that swept the European continent in October 2000. Such occurrences remain relatively rare, but they can have immediate impact on the policy process, and they are important as indicators of the disruptive potential of digi-
|* The Digital Divide
tal politics. Some flash protests are temporary phenomenon. Other transnational advocacy networks manage to sustain longer-term elec-
? * tronic coalitions, such as the International Campaign to Ban Landmines that resulted in a treaty signed by 122 nations in 1997. Global protest movements and direct-action demonstrations spreading across national ' borders have existed for decades, such as the antinuclear movement in the 1950s and the anti-Vietnam war protests of the 1960s, or even farther back the antislavery and the suffrage movements in the nineteenth century. The phenomenon is far from new but these movements are facilitated in an environment of minimal-cost instantaneous global communications. Governments, like British redcoats lined up in perfect formations, seem unsure how to respond. They are flustered when sud-
' denly outmaneuvered by the ad hoc coalitions of truck drivers and fuel-tax protestors, the environmental activists and animal-rights lobbies,
- j      the anticapitalists and anti glob alist forces. It is true, as cyber-skeptics
— - claim, that most established political institutions prefer to co-opt the \~—-    capacities of new technologies to preexisting functions, rather than
~, ~   being forced to reinvent themselves in the Internet age. But it is also true v    that the capacities of the Internet are adapted more easily by smaller, more flexible organizations, a process that is particularly important for the process of democratic consolidation, and for opposition move-~ Z     ments seeking to challenge authoritarian rule around the globe.
* ' " THE IMPACT ON CIVIC ENGAGEMENT
-     What will be die impact of this process for civic engagement among '     ordinary citizens? Part III goes on to examine the nature of the cybercul-tuie and the influence of digital politics on public participation, and then summarizes the core thesis argued in this book. Chapter 10 analyzes
^___ political attitudes in the United States and Western Europe. Many have
~ concluded that as the Internet population has gradually normalized in America, the digital world has come to reflect the general population.53 Nevertheless a more detailed examination of the values and attitudes of the online community in America and Europe, where we have survey evidence, suggests the existence of a distinctive cyberculture, one favorable toward the "new" left on the social agenda and the "old" right on the economic dimension. Just as Internet enthusiasts sympathize with non-regulation in the sphere of personal lifestyles, so they favor freedom from government in the economic sphere. Moreover, this cyberculture is not simply a by-product of the social profile of those who go online,
20
21
Digital Divide
The Digital Divide
since this pattern remains distinctive even after controlling for the usual demographic factors such as the age, education, sex, and income of the online population. Such a cyberculture is one broadly sympathetic to the alternative social movements that use digital technology most effectively for direct action and protest demonstrations.
Will the Internet have the capacity to revitalize public participation in conventional politics, such as levels of party membership, electoral turnout, or activism in civic and voluntary organizations? Chapter 11 suggests that digital politics reduces some of the information and communication costs for individual citizens who are interested in public affairs, but at the same time the Internet probably has the least impact on changing the motivational basis for political activism. In this way, digital politics functions mainly to engage the engaged. For those with access and motivation, the Internet facilitates opportunities for civic engagement, increasing the ability to drill down and compare multiple news sources on an issue, to forward articles and clippings to colleagues, friends, and family, to donate funds electronically to causes or election campaigns, to support groups mobilizing around particular issues, to organize within local neighborhoods, and to discuss politics online, as well as to research official documents and legislative proposals, to access government services and download official forms, and to contact public servants about particular problems of health or housing. Reduced information and communication costs lower some, although not all, of the barriers to civic engagement. Costs can only be expected to fall with the expansion of online political resources, giving grounds for optimism about the ability of digital politics to revive activism among the active. Yet the evidence also suggests that, at least in the short term, at individual level, altering the structure of opportunities and the balance of relevant resources probably has minimal impact on changing the motivational basis of political participation and interest among the mass public. Digital politics thereby contributes toward the vitality of representative democracy, but it also largely bypasses the disengaged.
In this regard, the role of the Internet is similar to the impact of traditional forms of mass media. Previous work has established a consistently positive association between use of the news media and indicators of civic engagement in the United States and Europe.54 Those who watch the news and current affairs on television, read newspapers, and listen to radio news were found to be more politically informed, trusting, and active than average, even with the usual controls for social
I  "   background such as age, gender, education, and income. The evidence I       in this book confirms that, along similar lines, those already most inter-|- -    ested and involved in public affairs take most advantage of the new |-- --    opportunities for information, expression, and political mobilization I -     available via the Web. Environmentalists, for example, are most likely to J-      surf the Greenpeace website, just as Republicans are most likely to check f**-     www.Bu5h2000.org, and women are most likely to click on www. I        Oxygen.com. Like discussing gun control or abortion over dinner with P   '    like-minded friends, reading liberal op-ed pages on problems of health ~        care or affirmative action in schools, or attending a protest rally about genetically modified food, this experience can be expected gradually to '        reinforce political attitudes and strengthen the involvement of the par--        ticipants. This process remains important, functioning to encourage the '        involvement of ordinary citizens in democratic government through representative channels. Yet it disappoints those who hope that the Internet will function as a deliberative public forum, drawing the less engaged into civic life, replacing representative institutions, and thereby | "      strengthening direct, plebiscitary, or "strong" democracy.
Therefore the theory developed In this book attempts to strike a balance between more pessimistic claims that the development of the • Internet will serve to reinforce the voices of the powerful, the more
[ skeptical claims that it will merely reflect "politics as usual," and the more optimistic claims that cyber-democracy will transform governance as we know it and restore levels of mass political participation. Instead, the book concludes that the restructured opportunities for information and communication available via digital politics will potentially have positive consequences for civic society, altering the balance of relevant resources and slightly leveling the playing field. The primary beneficiaries are likely to be marginal groups such as minor and fringe parties, loose coalitions of protest organizations, and alternative social movements, particularly those advocating causes that are most conducive to the cyber-culture. Reducing the costs of information and communication minimizes some, although not all, of the signiii-cant barriers to effective political participation at individual level; it becomes easier for ordinary citizens to learn about public affairs, if they are so inclined, and to express their views and to mobilize. This process is most important in many consolidating democracies, stranded midway between an authoritarian past and stable democratic future. The wider diffusion of digital technologies can play a significant role in
22
23
Digital Divide
strengthening civic society in countries such as Taiwan, Brazil, and South Africa if e-governance improves transparency and openness in the policymaking process, if parliaments and parties use new media to strengthen their internal organizations and their links with the public, and if opposition movements develop virtual coalitions to challenge the predominance of the government's message in television, radio, and newspapers. But whether the Internet can ever encourage the less engaged to take advantage of these opportunities at mass level remains doubtful, because as the medium of choice par excellence, it becomes even easier for people to tune out from public life.
Of course, as discussed in the next chapter, there are strong grounds for caution in any prognostication about future developments. This discussion relates to the use of digital technologies during the first decade of the emerging Internet age, and the long-term consequences of these developments cannot be predicted with any accuracy at this stage. History furnishes numerous examples of the failure to foresee the ultimate uses of technologies at the time when they were first introduced. Newfangled telephones were first thought of as channels of musical entertainment, not personal communications. In the nineteenth century, modest electric shocks were believed the novel cure perfect for Improving the healthy constitution. When wireless amateurs started broadcasting before World War I, most saw radio as an active medium of communication, a hobby for young boys, not a passive listening experience. Forecasts often fail to predict the weather, the election results, or the stock market for the day after tomorrow, let alone for decades from now. Contemporary estimates for the impact of the Internet may be similarly misplaced. Digital politics has evolved rapidly during the last decade, and multiple developments will probably occur within the next, such as online registration and voting. The long-term impact of digital technologies could ultimately produce different consequences to their effects during the emergent era. But despite the importance of considerable caution, the pattern of global, social, and democratic inequalities described in this study fits what we already know about the impact of traditional forms of political communications, like newspapers, radio, and television, and also receives support from the comparative evidence in the emergent Internet age, so that the evidence deserves to be examined with an open mind to contrary data and countervailing indicators. By systematically comparing the diffusion of digital politics around the world, including both the leaders and
The Digital Divide
laesard nations, this account can be tested to see whether it provides useful insights into the spread of the Information Society in recent years In conclusion, Chapter 12 recapitulates and expands on the core theory at the heart of this book, summarizes the evidence for this interpretation, and considers the broader implications for understanding digital politics in the Internet age.
24
25
CHAPTER 2
Understanding the Digital Divide
Debate about the impact of the rise of the Information Society has produced deeply contested visions predicting the future direction of trends. Optimists hope that the development of the Internet has the capacity to reduce, although not wholly eradicate, traditional inequalities between information-rich and -poor both between, and within, societies. In contrast, pessimists believe that the digital technologies will reinforce and exacerbate existing disparities. Skeptics suggest that both the fears and hopes are exaggerated, with technologies adapting to the social and political status quo, rather than vice versa. What evidence would help to settle these claims? How can we move from the Frank Capra and the Ingmar Bergman visions toward a more systematic understanding of the impact of the Information Society? It remains difficult to sort the facts from the hype, despite the burgeoning literature on all aspects of the Internet ranging from Web design, software development, and e-commerce to the sociology of the network society, group identities, and virtual culture. Studies in any discipline assessing the impact of the Internet face three main challenges: the problems of studying a phenomenon undergoing rapid change; the limitations of the available cross-cultural evidence allowing us to generalize beyond the experience of the United States; and the difficulties of developing and integrating triangulated methodologies drawn from different disciplines.
THE RAPID PACE OE CHANGE
The first challenge is the rapid pace of technological innovation and social adaptation so that studies of the impact of info-tech represent blurred snapshots of a moving bullet.1 The genesis of the Internet was
mm
Understanding the Digital Divide
initially fairly slow but postindustrial societies are currently experiencing a sharply accelerating 'S* curve of diffusion. The birth of computer-mediated networked communication can be traced back to ARPANET in 1969, an experimental four-computer network, established by the Advanced Research Projects Agency (ARPA) of the U.S. Defense Department to develop a secure form of communication via multiple destinations in the event of nuclear war. Information was split up into "packets" that were then transmitted via several dispersed routes: if one link was unavailable then, like a delta river with numerous tributaries, information simply traveled through alternative routes before the packets were reassembled at the destination. In the 1970s, dispersed communication networks spread email among a select community of scientists ■ and scholars at elite universities and research centers. In 1971 ARPANET linked about two-dozen computers ("hosts") at 15 sites, including MIT and Harvard, and a decade later more than 312 hosts were networked. This process was accelerated in 1986 by the National Science Foundation's development of a high-speed backbone network to link science and engineering, although, other than email, this remained mainly the domain of computer-science aficionados happy to struggle with unforgiving lines of computer programming and printouts. Data were still routinely delivered on magnetic tape mailed in large tin cans like movie reels. Beyond linking communications between research institutions and scholars, the most popular uses of the new networks were financial transactions in electronic banking and email for business.
The Internet as we know it today came about with the invention in 1989 of the World Wide Web and a hyper-text language for global information sharing, by Tim Berners-Lee at CERN in Geneva, and the subsequent release in 1991 of the first client browser software for accessing materials on the Internet. At this time about twenty countries were connected to the network, mostly in North America and Western Europe. The decisive technological breakthrough popularizing the medium occurred in 1993, when the National Center for Supercomputing Applications released Mosaic, the first graphical Web browser, made available for Unix systems, then for Microsoft Windows and the Apple Macintosh. The graphical browser removed the need for any technical expertise in accessing the Web beyond the ability to point and click, making it instantly accessible to a five-year-old. The remarkable rise of the Internet as a new mass medium came in October 1994 when Netscape Communications released the Netscape Navigator browser, built on Mosaic technology and distributed free. Microsoft awoke rela-
26
27
Digital Divide
Understanding the Digital Divide
tively late to the opportunities of the Internet but eleven months later, in August 1995, Internet Explorer was released, bundled with the launch of Windows 95.
In postindustrial societies, the Internet wildfire during the last decade has been, as everyone observes, remarkable. The earliest estimates suggest that in 1994 there were about 3 million users worldwide, mostly living in the United States.2 The following year this number had risen to 26 million. The online population has subsequently roughly doubled every year since then, reaching an estimated 407 million people by late-2000 (see Figure 2.1). The first-ever American opinion poll on this issue, conducted by Louis Harris and Associates, found that one-third of the public had heard of the Internet in June 1994 but only 7 percent had ever used it.3 Pew surveys estimate that the following year the proportion of users had doubled to about 14 percent of all Americans, but by mid-2000 more than one-half of all Americans used the Internet (54 percent).4
Therefore as a form of information and communications spreading beyond the scientific and technical elite, the Internet as a mass medium remains a relatively recent development. Computers have been around for about fifty years, and distributed computer networks for about thirty years, but the popular point-and-click World Wide Web, as we know it, has only existed since 1993. Predictions suggest that the familiar Internet experience of the first decade - with email and Web pages delivered through wired umbilical cords to beige desktop boxes - will probably not be the familiar Internet experience envisaged for the next decade, with at least a cut-down version enabled through wireless cell phones like DoCoMo services in Japan, pagers, digital televisions, handheld personal assistants like Palm Pilots, even streamed in headline versions through ATM banners and screens fitted in elevators, bus stops, and airports, with online automobiles so that we are All Internet, All the Time. For technophiles, the Web is promised to arrive through everything from our toasters to our televisions. Although it is difficult to sort out the reliable estimates from the industry hype, market research forecasts suggest that by 2005 more than one-half of all Americans online, and almost three-quarters of worldwide users, may have digital Web appliances to download information.5 Novel "killer apps" are predicted to transform information technologies.6 Yet predictions are in constant danger of being overtaken by events ("So 1998"), as well as being exaggerated into hyperbole by the industry in its own interests. Despite predictions that smaller firms will thrive in the new economy, multiple
500 i
400
c
; a
300
200
1 00
1 996
1 997
1998
1 999
2000
Figure 2.1. Worldwide Trends in the Online Population, 1995-2000. Source: "How Many Online", www.NUA. ie, November 2000.
entrepreneurial Internet start-ups have fallen by the way, while corporate mergers producing multinational companies are more fashionable today than ever before. Like the Nasdaq, past irrational exuberance surrounding digital technology stocks may prove an unreliable guide to future performance.
One of the best ways to understand the rapid pace of change is to monitor trends during the last decade across many different nations at the forefront of the emergent information society, in order to understand how digital politics has evolved in response to the new structure of opportunities. The past decade is likely to prove atypical of subsequent developments as people learn what does, and doesn't, work. In the 1998 U.S. elections, for example, only one in ten of the major party Senate, House, and gubernatorial candidate websites facilitated online campaign donations and almost one-half did not even ask for money.7 In contrast, just two years later John McCain's campaign raised $1 million via online contributions in the 48 hours after his New Hampshire victory in February 2000, or $2.5 million in total online.8 In 1996, grassroots activists and individual voters for or against the major presidential candidates in America set up a handful of homegrown t>ites. In 2000, in contrast, there were almost 7,000 such sites. In 1996,
28
29
Digital Divide
just a handful of dedicated Internet news outlets got to the U.S. presidential party conventions. Four years later there were 80 to 100 such outlets, like Slate and Salon, as well as online coverage by almost every traditional news organization including CNN and C-Span. The key issue with these sorts of developments is how the public responds when digital politics evolves. When government departments go online, how do people use these sites to seek information? When parties, groups, and campaigns use horizontal networks via "virtual" conferences, policy discussions, and innovative feedback mechanisms, does this mobilize supporters? What new formats work, and what don't? Although future developments remain uncertain, the 1990s represents a unique opportunity to capture how the first generation of online users evolved, similar to studies in the 1950s analyzing the early television audience.
One way to think about these issues is to draw upon classic theories of technological diffusion developed by the work of the nineteenth-century French sociologist Gabriel Tarde and by the Harvard sociologist Pitirim Sorokin (1941) and advanced by communications scholars Elihu Katz and Everett Rogers.9 These theories suggest that the adoption of many successful innovations - whether of new strains of seed corn, industrial machinery, or new medical breakthroughs -have commonly followed an S-(Sigmoid) shaped pattern.10 New technologies have often experienced a slow rate of initial adoption, followed by a substantial surge that peaks when penetration levels reach saturation point and demand subsequently slows. Cyber-optimists suggest that the spread of the Internet will follow a normalization pattern, as costs fall, as the technology becomes simplified allowing plug-and-play access, and as the Web increasingly provides mass entertainment and cheap communications via streaming audio and video. In the normalization model illustrated in Figure 2.2, those who adopt the innovations at an early stage will be ahead of the curve, with the resources, skills, and knowledge to take advantage of digital technologies, but in the long term cyber-optimists believe that penetration will become saturated in these societies. Once a high proportion of households have a personal computer and access to the Internet -like owning a refrigerator, automobile, or washing machine - then demand will slow. The theory predicts that given saturated demand, prices will fall further to attract new users, allowing laggards to catch up, so that eventually access to digital technologies becomes pervasive. The initial period of adoption may therefore be expected to widen social inequalities but the normalization hypothesis suggests
Understanding the Digital Divide
Cumulative S Curve at Technological Diffusion Normalization Model
T1
12
T...
Cumulative S Curve of Technological Diffusion Stratification Model
Figure 2.2. The Cumulative S Curve of Technological Diffusion.
that this temporary gap will eventually close. In contrast, cyber-pes-simists emphasize that the stratification model provides a more realistic scenario where groups already well networked via traditional forms of information and communication technologies will maintain their edge in the digital economy.
30
31
Digital Divide
Understanding the Digital Divide
Diffusion theory allows us to compare the growth of personal computers and the Internet with earlier technologies. In the United States, the spread of many previous innovations has usually followed a sigmoid (S-shaped) time path characterized by a slow pace of initial adoption, followed by a significant advance, and then a gradually tapering of demand (see Figure 2.3). Televisions in America experienced a rapid surge of sales in the 1950s, fueled by pent-up demand for consumer goods and the hiatus in TV production and broadcasting during World War II. VCR sales saw a similar surge in America during the late 1980s. In contrast, some other communication technologies took far longer to spread throughout the American population. Sales of radio receivers were initially held back by the technological complexity of crystal sets and the onset of the Great Depression in the 1920s, before experiencing a slow and steady rise, until today there are more radio sets than people in the United States. The telephone, which had been available as a commercial service since 1877, only took off for the majority of American households after World War II. Automobiles, as big-ticket household items, also experienced a steady climb in sales after 1945 until reaching a plateau in the 1980s. Cable TV saw slow diffusion in America from 1960 to 1980, due to the investment costs of laying cable and the number of stations available to most subscribers, before accelerating in availability. In the United States, the flood of Internet users since the early 1990s has followed an S-shaped curve, and it remains to be seen whether this curve will bottom out with two-thirds access, like cable TV, or more than 90 percent access, like TV. The pattern of American adoption so far has been closer to the rapid surge in television sets and VCRs rather than die slower diffusion of telephones and radios. Although the Internet remains a relatively new phenomenon, the diffusion patterns evident in related communication and information technologies in America provide important evidence about what we might expect to occur with the growth of the online community in future decades and in other postindustrial societies at the forefront of the knowledge economy, such as Sweden and Australia.
Worldwide the comparison of the spread of radios and televisions since the 1950s, and the rise of the Internet since 1995, shows more gradual secular trends rather than a sudden S curve (see Figure 2.4). The growth of the online community has been substantial: For comparison with previous innovations, the telephone took close to 75 years to reach 50 million users worldwide, and television took 13 years, but it took only 4 years for the Internet to reach the same number.'1
Figure 2.3. Twentieth-Century American Technology.
Sources US. Census Bureau: Statistical Abstract of the U.S.,1999; Historical Statistics of BieUS
CROSS-NATIONAL EVIDENCE
ii"J"' hi addition to drawing comparisons historically over time, another major challenge concerns the difficulties of generalizing across many countries based on the limited evidence. Most studies of digital polities focus on the United States yet these findings may well be, in this, asan so much else, exceptional.12 As an industry leader, the United States is certainly atypical in Internet use, even within the universe of postindustrial societies - containing an estimated three-fourths of all
I -11** e-commerce sites worldwide, 79 percent of the world's Internet hosts,
ipp
illltl
59 percent of the world's electronic mailboxes, 54 percent of online buyers, and 38 percent of Internet users.13 If access to digital technologies is heavily contextual, depending on the structure of opportunities available within each society, then the typical experience of Silicon Valley dot-com entrepreneurs, Harvard undergraduates, and New York lawyers will probably have little in common with their counterparts in London, Paris, and Tokyo, still less in Moscow, Beijing, and Johannesburg. A broader analysis, which examines global patterns, contrasting leaders and laggard societies, provides the basis for more reliable generalizations.
32
33
Digital Divide 50 -i---
1950 1955  1960  1965  1970  1975  1980  1985  1990 1995 2000
Figure 2.4. Worldwide Diffusion of Radio, Television, and the Internet, 1950-2000.
Sources: Television sets and radio receivers per household; annual UNESCO Statistical Yearbooks, UNESCO, Paris; estimates of the Internet population: "How Many Online?" www.NUA.ie,
There is another important reason why we need comparisons beyond the United States. Democracies differ significantly in their core institutions and constitutional features, most notably in terms of majoritarian or proportional electoral systems, the range of competition in party systems, whether executives are parliamentary or presidential, whether state power is centralized or dispersed, and so on. These institutional structures have significant consequences for patterns of political participation such as levels of voting turnout and types of election campaigning, as well as in rates of party membership and activism.'4 If digital technologies adapt chameleon-like to existing political systems, then we would expect to find considerable cross-national differences around the globe. The rapid adoption of the Internet as a lobbying and fund-raising tool in American election campaigns, for example, may reflect the particular form of interest group pluralism and money-driven political campaigns characteristically found in the United States, rather than a model common in many European democracies. The German SDP intranet, as a democratic mass-branch organization, may provide far
Understanding the Digital Divide
rhore opportunities for horizontal interaction and communication among party members than is available in the Japanese leadership-dominated LDP. The online delivery of services for housing and health in Swiss local cantons, with a stronger tradition of decentralized governance, may prove more advanced than equivalent services provided by
T't English county councils. The parliamentary website for the Norwegian
rV - Storting can be expected to be far richer and more interactive than
'?- % those designed for less influential and democratic bodies such as the
f"-" Jordanian National Assembly or Thai Ratha Sapha. And so on. Despite
f - the Internet's growing importance, at present little systematic empirical
f - research compares its spread across nations and its functions in different political systems across the globe. This raises a series of issues: Who
I surfs in Germany, Japan, and Mexico? Who reads online newspapers or
I uses broadband television and radio in the United States, Taiwan, and
>   ' Italy? What information about government services is available in
a- Switzerland, Canada, and South Africa? How is the World Wide Web
'} _ utilized by parties, by networks of alternative social movements, or by
,- lobbyists in France, Sweden, and India? How is email employed to
' mobilize dissident groups, human rights activists, and opposition
r  - movements to challenge the authority of authoritarian regimes in
i_ _ _ Burma, Afghanistan, and China? Scattered case studies of digital politics
]V - are available in many particular countries, and a burgeoning literature
f- is available in the United States, but so far the broader picture across the
!_ - globe remains unclear.
iv-
VÄ TRIANGULATED RESEARCH DESIGNS
- Another major challenge is that research on the Internet needs to inte-;* grate research findings drawn from numerous disciplines including
i those of communications, sociology, anthropology, history, social psy-
J~ chology, market research and business studies, computer studies, and " industrial design, as well as political science.'5 Qualitative methodolo-
gies deconstructing the meaning of digital communications include discourse analysis, literary criticism, rhetorical studies, and textual "r analysis. Quantitative approaches include the standard techniques of
sample surveys representative of the general population and special surveys of the online community, content analysis, focus groups, experimental research designs, and newer market research procedures monitoring user behavior like "click stream data" from cookies measuring activity on websites.16 Yet in the early years, the available data meas-
34
35
Digital Divide
Understanding the Digital Divide
uring Internet use often remain "guesstimates," even with the latest available market research techniques.17 No single methodology can hope to capture the rich complexities of life on the Internet and this study therefore draws on hundreds of studies from different disciplines, as well as empirical evidence from aggregate data, content analysis, and cross-national surveys. The most effective research strategy is to triangulate among diverse sources of evidence, attempting to understand the Internet by piecing together a range of independent studies to see if the evidence points in a consistent direction across different countries. Where the findings conflict, we need to point out the uncertainties and consider some of the reasons leading to these different results. Where the results survive replication, this increases confidence in the reliability of the generalizations.
The book draws upon multiple databases to compare the worldwide patterns of use. Estimates of the online population are provided by NUA, a company that monitors surveys from a wide range of different market research companies.'8 Most data are collected for commercial purposes, to gauge the market for e-commerce. Although use of different surveys limits the reliability of the comparison, nevertheless this source provides the most comprehensive and up-to-date picture of Internet penetration rates worldwide. As discussed in the next chapter, evidence from independent sources serves to confirm the global pattern established in the NUA data, including information about the geographic location of Internet hosts, collected by many international agencies, and data on the distribution of telephones and computer equipment. Worldwide data on websites established by parliaments, parties, government departments, the news media, and interest groups are assembled from multiple sources, providing a comprehensive global map of digital politics. The study compares 179 nation-states worldwide, including in total 5.77 billion people living in societies at all levels of human and political development.19
Representative surveys in the United States and the fifteen-member states of the European Union are used to compare the social background, political attitudes, and behavior of the online community with the general population in these countries. In America, the first occasional opinion poll items on use or awareness of the Internet occurred in mid-1994 but it was only the following year that the population started to be monitored more systematically. The benchmark survey data used in this study are drawn from 1995-2000, which allows us to examine the rapid diffusion process in the emergent era as use of the new information
Mm
technology penetrated the United States and Western Europe. Broader comparisons would have been desirable but unfortunately systematic and reliable cross-national surveys measuring the impact of digital politics are still unavailable in most countries, and this has to await further research. The book draws on the series of American surveys conducted slticc 1995 by the Pew Center for the People and the Press, and also the National Election Study since 1996. For a broader comparison with the fifteen member states of the European Union, the book analyzes the biannual series of Eurobarometer surveys since 1995. Although identical items are not always available, functionally equivalent items allow comparisons to be drawn between America and Europe.
To understand digital politics within each country, the study analyzed the contents of a selected range of parliamentary and party websites around the globe. The aim was not to develop a comprehensive mapping exercise but radier a more limited attempt to isolate and compare some of the key functions of these sites. Using standardized instruments, the coding monitored the presence and depth of informational features (both text and graphics), and communication functions (such as opportunities to email the organization and its representatives, link to Hstservs, bulletin boards, and chat rooms, and other ways to become active). To analyze government websites, comparable data were drawn from the CyPRG group database, which has monitored the content and functions of official departmental sites worldwide since 1997.
Finally, as discussed earlier, the analytical framework used in this study (illustrated in Figure 1.1) distinguishes three nested hierarchical levels of analysis: the macro-level technological and economic environment which determines the availability and social distribution of Internet access within each country; the meso-level context of political institutions which provides the structure of opportunities mediating between citizens and the state including parties, parliaments, government departments, interest groups, new social movements and the news media; and micro-level individual resources and motivation affecting patterns of online civic engagement. This approach requires an analysis of both institutional and individual data. The framework assumes that levels of technological diffusion, such as the proportion of the population online, influences how political institutions have adapted to the Internet environment. In turn, the core institutions of representative democracy that are available in the digital world provide the systematic context within which citizens have opportunities to participate online. Which citizens choose to take advantage of these opportunities is
36
37
Digftal Divide
understood to be determined by their resources (like time, money, and skills) and motivation (like interest, confidence, and efficacy). To develop this framework further, we can go on to examine which nations have emerged at the forefront of the knowledge economy, and which remain laggards in Internet diffusion, and the reasons behind these disparities at macro-level.
Hi
CHAPTER 3
Wired World
h '
n
rhe World Bank, the United Nations, and the G-S have expressed alarm that poorer societies lacking technological investment will drift farther behind their wired rivals in the global marketplace, whereas advanced industrialized societies will surge even farther ahead on the back of dramatic productivity gains. Multiple policy initiatives .■have been proposed, such as investment in technological infrastructure in Malaysia, computer training and education in schools in Latvia, and innovative community-level schemes in Bangladesh. Yet understanding the role of the state and the market in this process, and predicting which initiatives will succeed or fail in widening access in poorer societies, remains difficult unless we understand the reasons for the North-South divide. To unravel this issue we need to map the global spread of the information society and analyze the underlying conditions driving the process of technological transfer.1 Many studies by historians, development theorists, and communication scholars have attempted to characterize the mechanics of the diffusion process, and economists and marketing specialists have attempted to identify the driving factors behind the demand for new products.2 Drawing on this literature, after discussing the theoretical debate, this chapter focuses on four interrelated questions:
' What is the global pattern of Internet diffusion?
' Does this pattern represent the particular characteristics of Internet diffusion per se, or does it reflect similar trends found in the adoption of older forms of info-tech, such as radios, telephones, and televisions?
• In exploring the reasons for inequalities of Internet access, how far do cross-national differences reflect basic economic divisions
38
39