Analytically, the most ambitious of these works is that of Jane Fountain, whose project is no less than to reformulate a theory of bureaucracy to encompass recent changes in and applications of ICT. The reorganization of the American government in light of the Internet signals, she argues, represents a profound transformation of the American state. As with e-commerce, the initial euphoria about e-government has been replaced by a growing recognition that the potential of the Internet can only be exploited through fundamental organizational restructuring. Without an empirical theory to understand the relationship between ICT and the bureaucratic organization, there can be no normative theory to guide public decision-making about technology acquisition and deployment. Fountain's intention, therefore, is to fill a much-needed gap, rather than to amend or refute well-developed approaches to this topic, because "[T]here are no well-developed approaches" (17).

To date, American policy-makers have used ICT in ways that leave the deeper structures and processes of government undisturbed: "Organizations tend to patch information systems onto existing structures in ways that may enhance efficiency and capacity but that otherwise maintain the status quo" (19). Much of Fountain's empirical work is designed to substantiate this claim about such "plug and play" applications. Virtual agencies and networks are certainly increasingly prevalent, but they have emerged without a clear understanding of how underlying bureaucratic structures are being transformed, and of the wider implications of those transformations.

Fountain borrows from three streams of scholarship to draw the appropriate theoretical lessons: the political-science literature on governance; the new institutionalism theories from organizational theory and sociology; and the multidisciplinary literature on the relationship between information technologies and organizations. Two introductory chapters on the early attempts by the U.S. government to establish virtual agencies are followed by a more in-depth comparison of Weberian bureaucracy and "virtual" bureaucracy, and by an analysis of the impact of the Internet on organizational networks and the creation of social capital. The Internet acts, Fountain concludes, as a "catalyst for the formation of interorganizational networks by providing a cheap, powerful infrastructure for communication and shared information" (82). Using these insights, she goes on to construct her own theoretical framework designed to explain how "enacted technology" (as distinct from "objective technology") becomes embedded in existing institutional structures and norms.

Unsurprisingly, the theory is founded on some basic assumptions about the reflexive capacity of agents and the complex interplay of structure and agency. A series of propositions about agency behavior is generated by this framework and then used to guide Fountain's analysis of three case studies. The International Trade Data System was designed to develop information systems to integrate the work of over 60 separate federal agencies with jurisdiction over trade matters. The U.S. Business Advisor was the nation's first federal Web portal, linking about 300 government databases and bringing together for the first time information and services of interest to small and medium-sized business. By contrast, Fountain also offers a case study of the development of automated battlefield management systems within the more traditional, hierarchical, Weberian structure of the Ninth Infantry Division of the U.S. Army.

Fountain charts some new ground in this book. It will be of interest to any student of U.S. bureaucracy, and anybody concerned with the impact of ICT on the state. At the end of the day, I am not sure how useful her analytical framework will be for future studies. It rests on some questionable conceptual distinctions, especially that between "enacted" and "objective" technology. I also found that the case studies, while carefully researched, tended to stray from some of the initial theoretical concerns, and that the various loose ends were only partially tied up in the concluding chapter. Nevertheless, I appreciated Fountain's ambition, her thorough knowledge of the relevant literature, and the insightful ways in which she told the stories of the initial attempts by the U.S. federal government to provide services via the Internet.

One of Fountain's concluding predictions is that no virtual state can be built without careful attention to the issue of personal privacy. That is the focus of Alan Charles Raul's book, the gist of which is an examination of the proper balance between information privacy and access to public records. A brief conceptual chapter is followed by an analysis of the federal foundation for U.S. privacy policy, although the overview is far from comprehensive. There is then a more useful chapter on public records privacy, with some helpful lists of the various categories of personal information maintained by federal and state governments that are routinely made available to the public. Raul goes on to assess the impact of e-government initiatives on public records policy, and he concludes with some "Best Practices" recommendations designed to strike a realistic "balance" between access and privacy and to ensure that the most "sensitive" personal information is accorded the strongest privacy protection.

The bulk of the book was obviously written before September 11, 2001, and so Raul was unable to review the important legislation that has been passed in the aftermath. Nevertheless, his book makes few advances in our understanding of privacy-protection policy. On the one hand, I found it improperly conceptualized; a more thorough review of the privacy literature would have immediately exposed the theoretical and practical impossibility of founding a privacy-protection policy on a prior identification of the most objectively sensitive categories of information. Any information can be sensitive and potentially harmful to the individual, depending on the context. There were also some curious omissions from the chapters on U.S. legislation, some very cursory and fragmentary discussion of centrally important issues, and some inexplicable concentration on some issues (such as identity fraud) at the expense of others. There are, in short, several better books about contemporary U.S. privacy protection policy.

Privacy protection is not a central theme in the edited volume by Elaine Ciulla Kamarck and Joseph Nye, the latest in the series of works from the "Visions of Governance in the 21^st Century" project at Harvard's Kennedy School of Government, from where most of the authors are drawn. After Nye's introductory essay, which tries to place the "information revolution" in some historical context and theorize about the impact of this "third industrial revolution" on governance, the bulk of the essays concentrate on the impact of the Internet on the theory and practice of American democracy. We are first treated to Arthur Applbaum's examination of the likely impact of the information revolution on Madisonian democracy, together with a response from Dennis Thompson in the form of a fictitious e-mail from Madison back to Applbaum. The point of contention centers on whether or not the Internet, by diminishing the effects of geographic dispersion, removes one of the chief impediments to the formation of those tyrannical factions with which Madison was so concerned.

One of the more robust contributions in the book is provided by William Galston's analysis of the impact of the Internet on civic life. Four key structural elements of community (limited membership, shared norms, affective ties, and a sense of mutual obligation) are applied to a comparison of online groups and place-based communities. In dispelling some of the myths about online communities, Galston concludes that while they can fulfill important emotional and utilitarian needs, "[T]hey must not be taken as solutions for our current civic ills, let alone as comprehensive models for a better future" (56). A similarly skeptical note is struck in Pippa Norris's empirical investigation of the use of the Internet in recent U.S. elections and by Elaine Kumarck's findings that political campaigning on the Internet is typically confined to the provision of very traditional "electronic brochures." David King's analysis of how the Internet may be used to "catch voters" through sophisticated profiling and marketing techniques has to rest on a quite hypothetical campaign, rather than on a detailed observation of current practice. Jane Fountain's contribution is drawn from the book already reviewed above. Jerry Mechling's discussion of e-government begins with the caution that "[I]f the e-government agenda were a trip to the moon, we would have just cleared the gantry a few seconds ago" (141).

I was fortunate enough to have a sabbatical year at the Kennedy School in 1999-2000, at the height of the dot-com boom where there was enormous scholarly excitement about the ability to ask a lot of old questions about democracy and governance in relation to some very new technological applications. It is, therefore, instructive that so many of these essays strike a skeptical tone, reflecting an obvious "coming down to earth" amid a recognition that the political uses of ICT are not amenable to easy and unidirectional theoretical frameworks. As Fountain reminds us, there is no substitute for careful empirical analysis of the application of these technologies in different contexts, informed by robust theoretical propositions. And, to strike a final critical note, there is also no substitute for comparative work, because there is no reason to assume that the experience of "American" democracy in the Information Age can-and should-be generalized to other advanced democratic societies.