Part II offers an analysis of the Minnesota Electronic Democracy Project (MN E-Democracy), focusing on the MN-POLITICS list archive. The analysis will demonstrate the potential possessed by the Project as a Deweyan system of public opinion formation. Section 2.0 and section 3.0 introduce the reader to the Internet and to MN E-Democracy. This introduction is accomplished through general histories, followed by various perspectives - including an analysis of participants, of usage statistics, and a description of the Project archive as an artefact. Section 4.0 offers a background to some literature on the social effects of CMC that apply within the framework of the Dewey-Lippmann tension developed in Part I. I argue that Deweyan interpretations of social effects, such as the position that the medium can encourage adherence to normative behaviour depending on context, are more accurate than individualist interpretations which claim the medium encourages anti-normative behaviour. Furthermore, an initial survey of the material collected from the Project supports the accuracy of the approach. This survey moves us from a consideration of social effects of CMC to a consideration of political effects. Section 5.0 focuses on the design of the Project and the manner in which boundaries between anti-normative and normative behaviour are constructed as well as how activity outside the boundaries can be managed by organisers and participants. Section 6.0 approaches the mechanics behind the process of participation within the boundaries of acceptable behaviour. The important product that results from participation is a thread. This section also attempts to account for the mechanics of thread generation. Section 7.0 focuses on the threads that were generated over the course of the Project. At different periods there are different patterns of thread generation and these patterns offer insight into the development of the Project and the potential impact on politics. The data suggests that over time the new knowledge generated within the thread emerges to guide the construction of threads as objects, resulting in a deliberation over normative and anti-normative behaviour among participants as well as the opportunity for democratisation. When included in this form of interaction, a candidate for office must be cognisant of the deliberation over normative values or risk failure in the forum. Section 8.0 offers an account of how the media communicated the new knowledge generated during the Project to a mass audience.

The MN E-Democracy Project provides material for a case study of how the Internet has affected, and may continue to affect, the American democratic process. To gain an understanding of the MN E-Democracy Project as well as how the Internet might ultimately effect politics I will begin with a brief history of the Internet itself.

During the 1960s, at the height of the Cold War, the Advanced Research Projects Agency of the United States Department of Defence sought to develop a distributed communications network for the United States military. The result - in collaboration with the private firm Bolt Beranek and Neuman (BBN), the University of California at Los Angeles (UCLA), Stanford Research Institute, the University of California at Santa Barbara and the University of Utah - was ARPANET. For the military, the purpose of ARPANET was to provide a system of communication that would be fully functional even though one or more points might be destroyed by enemy attack. From the point of view of many of the computer scientists who contributed to the Project, such as Paul Baran and JCR Licklider, the goal was to create a means of communication that would allow any user or program on any of the networked computers to be able to utilise any program or subsystem on any other computer without having to modify the remote program.

Three important ideas that underlie the concept of computer networking need to be explained: "packet-switching", "time-sharing", and the connection of the computers to one another through ordinary telephone lines. Packet-switching allows digitised information to be broken down into chunks (the 0s and 1s that make up bits and bytes) at the point of departure from one computer, sent through different paths to a destination in another computer, and reassembled on arrival (Bonchek 1996). Time-sharing enables a computer to devote its resources to multiple tasks within a span of seconds or less, so that a user at the host computer as well as a user at a remote computer can interact with the machine and therefore each other, simultaneously. Third, the decision to connect all the computers on the network through dialup telephone allows a cheap and easy way to implement packet-switching and time-sharing techniques for a limitless number of potential users.

As individuals began to communicate over phone lines through computers a broad range of applications made new forms of interaction possible. The key applications that continue to lie at the heart of the Internet include File Transfer Protocol (FTP), which allows for the sharing of files and other resources between remote computers; telnet, which allows for remote login access; and electronic mail (e-mail), which allows for users on remote systems to exchange personal messages and other information. In 1971 two programmers at BBN who wanted to communicate with one another through personal messages developed electronic mail, arguably the most important application of new technology (Lynch et al. 1993). With computer-mediated communication possible through the phone lines, a new form of mass communication was clearly being developed.

The roots of ARPANET mesh extraordinarily well with Deweyan thinking about the role of scientific investigation as a model for democratic practice. For example, during World War II, Vannevar Bush was the Director of the Office of Scientific Research and was responsible for the work done by physicists on behalf of the war effort. After the war and under the weight of the newly emerging nuclear threat, Bush attempted to inspire the community of physicists to apply their skills to the task of perfecting a new machinery of knowledge. In "As We May Think", a piece that appeared in The Atlantic Monthly in 1945, Bush wrote:

The applications of science have built man a well-supplied house, and are teaching him to live healthily therein. They have enabled him to throw masses of people against another with cruel weapons. They may yet allow him truly to encompass the great record and to grow in the wisdom of race experience. He may perish in conflict before he learns to wield that record for his true good. Yet, in the application of science to the needs and desires of man, it would seem to be a singularly unfortunate stage at which to terminate the process, or to lose hope as to the outcome. (Bush 1945)

Bush’s vision continued to be influential among the generation of scientists working on computer networking in the 1960s. For example, Fernando Corbato and Robert Fano, both at MIT, developed the concepts such as time-sharing at the foundation of the new technology. Like Bush they saw the positive potential in creating a new machinery of knowledge. They also envisioned how networking would enable this machinery to work for the benefit of community collaboration. They wrote:

The time-sharing computer system can unite a group of investigators in a co- operative search for the solution to a common problem, or it can serve as a community pool of knowledge and skill on which anyone can draw according to his needs. Projecting the concept on a large scale, one can conceive of such a facility as an extraordinarily powerful library serving an entire community, a sort of intellectual public utility. (Corbato and Fano 1966, Hauben and Hauben)

Finally, in the ARPANET Completion Report, written in 1978, the authors stated that a new communications medium comparable to the telegraph and printing press had successfully been developed. Pointing to the tremendous impact the new medium would have on society at large, specifically, in this instance, in relation to e-mail, the authors wrote, "The largest single surprise of the ARPANET program has been the incredible popularity and success of the network mail. There is little doubt that the techniques of network mail developed in connection with the ARPANET program are going to sweep the country and drastically change the techniques used for intercommunication in the public and private sectors." From inception to completion, visionaries in the scientific community seemed set on, as John Dewey might say, altering the modes of conjoint behaviour through computer networking.

While military goals were central to the Project’s original purpose, the undertaking soon became a general research project without a specific application. One reason for this is that ARPA (which changed its name to DARPA in 1972) did not and does not perform research or have laboratories but rather funds research performed at other locations. Furthermore, although the Pentagon supported the project, JCR Licklider in turn gave a great deal of support to the computer science community in academia. Thus a variety of scientists working in a variety of situations for a variety of reasons drove the development of computer networking. Furthermore, as host sites were added to the network it increasingly became a tool for the academic research community. Beginning in 1969 with 4 sites, the network had 15 host sites by 1971, 45 by 1973, 111 by 1978, and 4,000 by 1983. In 1980, MILNET was formed for the military and split off ARPANET, thus disappearing from public view while ARPANET would continue to be used for research and development.

Many innovations and stories about bringing the new medium to more people would drive the development of the new technology from the early and mid-70s to the almost omnipresent Internet of today. Among the early innovations was the development of Transmission Control Protocol/Internet Protocol (TCP/IP), a protocol that would allow various types of computers operating various software applications to exchange information if part of the network. Also important was the combination of the release of Digital Equipment Corporations relatively low-cost VAX family of mini-computers and the development of UNIX, an operating system specifically designed to allow for the easy development of applications in and for the open, interactive arena of computer networking. Cheaper computers and easier to understand operating systems made computer networking feasible for many in the research community and private sphere (Hauben and Hauben, Internet Unleashed 1994).

Of course, compared to the mass public the community of users in the 1970's was still minuscule. The principles of user-friendliness being developed at the Xerox Palo Alto Research Center (XeroxParc) were, therefore, an important parallel development to computer networking. Xerox Corporation funded a research and development division which drew some of the most talented individuals in computer science. Although Xerox was not quick enough to capitalise on their work, they designed key interface concepts such as the "desk-top", "files", "folders", "trash", the "pull-down menu", the "mouse", "point-and-click", and "windows". It was left to others such as Steven Jobs, the founder of Apple, and Bill Gates, the founder of Microsoft, to capitalise on these ideas and over a period of two decades bring the personal computer (PC) to a mass market of billions. Without such a critical mass the Internet would not be on the verge of universality, connecting the worlds satellites and TVs. Instead, it would still be an arena for computer specialists (Cringely 1993, Levy 1994).

Another series of stories involves the way in which computer networking led to the emergence of a variety of networks formed for different purposes. In 1979, an academic group that had been using the slower UNIX to UNIX Control Program (UUCP) for computer networking, created the Computer Science Research Network (CSRN) and attached it to the faster ARPANET through a single gateway, thus beginning the web-like interconnections among various networks that has come to characterise the Internet. Also in 1979 Steve Daniel and Tom Truscutt, two students at Duke University, created Usenet News. This application eventually became so successful that a new protocol called Net News Transfer Protocol was created to carry the Usenet News feed over the TCP/IP connections that tied host sites into the Internet. Usenet News is an almost completely open networking space where anyone with access can participate in ongoing conversations on several thousand topics. Although the future of Usenet News is uncertain, its existence has contributed a great deal to the anarchic reputation of the Internet.

In 1983 a relative of the Usenet Newsgroup and a core technology of the MN E-Democracy Project, the Internet listserv, gained popularity through a service developed at the City University of New York called Bitnet. The listserv is a mode of group communication by e-mail. A program - "majordomo" in the case of MN E-Democracy - is located at a central computer (a server) that distributes e-mail messages to every individual who "subscribes" to a particular listserv. The individual joins by sending an e-mail message to the program at the central server informing it the individual wishes to subscribe. Once subscribed, the individual will receive every message sent to a central list address. Thus, if 10, 100 or 1,000 machines subscribe to the list, all 10, 100 or 1,000 will receive each message sent to the central address. This creates a method of interaction that exemplifies computer interactivity. It is an example of the new technological concepts that I propose will, in Deweyan terms, alter our methods of conjoint behaviour, be a catalyst for the formation of a new public, influence the selection of our representative officials, and alter the political process. These technologies will do this by rendering apparent false dualities, as well as enabling sophisticated processes in replacement.

By the early 1980s local computer networks began to spring up. These created a way for enthusiasts to network from their homes through modem. As increasingly the population absorbed the technology, thousands of independent computer networks emerged and eventually merged into the Internet. For example, in the mid-1980s a group of 1960s activists developed the Whole Earth ‘Lectronic Link (the WELL). The WELL was one of the first and continues to be one of the most successful electronic communities in which computer networking enthusiasts could interact with others around the world. Perhaps part of its success has been that it is firmly rooted in the San Francisco region. Nonetheless, the success of the WELL began an era in which computer networking became increasingly big business (Rheingold 1994). America Online, Compuserv, Prodigy, Delphi all emerged from a pack of computer networks that appeared in the late 1980s and early 1990s, offering customers the virtues of cyberspace and "virtual" - non-geographically rooted - "community", for a subscription fee. At first these independent networks sought to restrict customers to the use of their closed, or proprietary, networks. As the non-proprietary Internet becomes the standard, this is no longer possible. Proprietary networks increasingly have to open themselves to the Internet, becoming filters and Internet Service Providers (ISPs). In 1990, as a result of the commercial success of computer networking, in any event, the Department of Defence disbanded ARPANET and, for a short time, used other government funded networks (such as the NSFNET) as the principal backbone of what was, by that time, known as the Internet. In the fall of 1995, because the Internet follows a mesh design and a central backbone is not necessary, NSFNET and the other government funded networks ceased to provide public support for the maintenance of the Internet. The future of the Internet was, thus, turned over to many private networks working, loosely, in concert.

While the business of creating "virtual community" has been popularised as the Internet is increasingly supported by the private sphere, the local character of the early history has led to a community or civic networking movement. Local issues and consideration of the public good, rather than profit, feature heavily in this movement. Because the Twin Cities Freenet hosted the MN E-Democracy Project, the freenet movement and the community networking movement as a whole are particularly relevant. In 1986 Thomas Grundner, an employee at a medical clinic in Cleveland, Ohio, decided to put medical information on-line to better serve the community. The service proved so popular that Grundner expanded beyond solely medical information. He began to supply general community information, including government records, chat lines, educational information, etc. Several hundred thousand people in the Cleveland metropolitan region joined the facility, giving birth to the Free-Net and community networking movement. For a time Grundner tried unsuccessfully to guide the movement through his National Public Telecommunications Network (NPTN), modelled after National Public Radio. He also sought to guide others about how to form a freenet in their communities. As a result, there are now over a hundred freenets and many more networks based on the freenet model spread across the world. Even though NPTN has not itself survived, the contribution of Grundner and his freenet to the culture of the Internet is significant (Miller 1996, Schwartz 1996).

Key to pulling the Internet together in the late 1980s and early 1990s was the development of a number of tools to allow less computer literate individuals to use the concepts behind computer networking, allowing them to exchange personal messages and information. The continuing development of these tools has driven and continues to drive the use of the Internet. The number of users is reported to be growing at a rate of 10% each month. "Hypertext", the "World Wide Web", "Mosaic", and - subsequent to the MN E-Democracy Project 1994 - the "Netscape Navigator" and "Microsoft Explorer" have replaced early tools such as "gopher-space". Ted Nelson, influenced by Vannevar Bush’s conception of a world of readily available knowledge, developed the concept of Hypertext in the 1970s. Hypertext was a method to access this knowledge by creating links between ideas on a computer network, so that an individual might be looking at a given text, see a link in the text to another idea or point of interest, touch the link, or point-and-click with a mouse, and be moved through the network to a new text. Of course, the concept of links can be and has been used equally well to connect pictures, sounds and video presentations. During the 1980s many in Silicon Valley sought unsuccessfully to capitalise on the hypertext concepts. A hypertext system designed to distribute multi-media scientific information across computer networks was not successfully developed in the United States. Rather, it was developed at the European Particle Physics Laboratory (CERN) in Switzerland. In 1989 Tim Berners-Lee circulated a proposal and the World Wide Web (WWW or Web) was introduced to the High Energy Physics community as a reality in 1991. Since then the Web has helped bring the Internet to the attention of the general public in many parts of the world. This is because it is relatively easy to create a document on the Web and embed links in the document to take the user anywhere else on the Web. In a sense the Web has provided a glue that holds the Internet together for the general user. The development of increasingly user-friendly Web "browsers", or tools on a hard drive that enable the user to view the Web, have made increasingly the Internet popular. In 1993 the Web browser Mosaic, developed by Marc Anderssen, then a graduate student at The National Center for Supercomputing Applications at the University of Illinois, Champaign-Urbana, was released (December 1994). It was the standard during the MN E-Democracy Project, which was located on the Web at the Twin Cities Freenet.

By 1993 the intricate threads had come together, creating a juggernaut ready to push beyond the consciousness of specialists, into the popular imagination. The principal pathways by which the knowledge of the new technology moved into the mainstream were through political and media institutions. Certain actors receptive to the new technology and operating within these intermediary institutions facilitated this process. Many were beginning to have a relatively more or less shadowy understanding that the new technology would eventually transform the manner in which political and media institutions operate. While the new technology will have an impact on most institutions, political and media institutions are here primary and, therefore, the primary institutions under consideration. Key among the earlier popularisers was Al Gore, who, in the early 1990s, as a Senator from Tennessee and head of the Senate Science Committee, fashioned the metaphor of an Information Super-Highway. This innovation was a derivative of the national super-highway initiative championed in the Senate a previous generation by Gore’s father. Gore carried his belief in the potential of computers and an information super-highway further into the mainstream as a part of the Clinton-Gore literature in the 1992 Presidential campaign. At that time the Independent candidate for President, H. Ross Perot, also brought the concept of an "Electronic Town Hall" into the mainstream. Traditional media institutions gave the rhetoric about the beginning of an information age even more immediacy through reports that the Clinton-Gore and Perot campaigns were beginning to bypass traditional media institutions during the campaign (Schwartz 1994). Increasingly they reached the public through radio call-in, satellite television, cable television, C-Span, talks shows and the like. This caused an as yet unabated interest in stories about new technology, computers and the media business. There was also concern about convergence in the forms of media such as print, audio and video through computer networks, etc. The new administration again furthered interest in these issues when it launched a National Information Infrastructure initiative and then a Global Information Infrastructure initiative, as well as calling for a re-write of the 1934 Communications Act (Browning 1996, Miller 1996, Schwartz 1996).

Steve Clift, a 25-year-old student at the Hubert H. Humphrey Institute of Public Affairs in Minneapolis, Minnesota, founded the Minnesota E-Democracy Project in July 1994. Clift had been active in Minnesota politics since his college days at Winona State University, where he was the Chairman of the Young Democrats of Minnesota. Clift was also interested in studying the impact of new communications technologies on governmental organisations and the political process. Originally, the Project was designed to create a place on the Internet for the public to access information from the candidates and about the candidates running for office in the up-coming State and national elections that November, 1994. Clift drafted a preliminary proposal and sent it to several local electronic discussion lists.

The idea was warmly received by others interested in both information networks and Minnesota politics. Several well-attended meetings were organised. An infrastructure was quickly put in place. Although the Twin Cities Freenet (TCFN) was not yet open to the general public, the main organisers, Scott Fritchie - a systems manager at St Olaf University - and Olaf Holt - at the University of Minnesota - had put TCFN on the Web. As TCFN was itself a new organisation intent on being involved in the community, Fritchie and Holt offered TCFN as the host site for the MN E-Democracy Project. A connection between Steve Clift, who had considerable experience and understanding of Minnesota politics, and the technical people involved with TCFN was forged, developing a pathway for political information to migrate on to the Web. Clift would create contacts with people in the political campaigns, put the information on computer disks and give the disks to volunteers to code so it could appear at the Web site.

Another key innovation of Project organisers was the decision to create an e-mail based public discussion forum using listserv technology. Dennis Fazio, President of Minnesota Regional Network, a non-profit Internet Service Provider, agreed to host the forum, MN-POLITICS, using computers at their site. MN-POLITICS was open to anyone who had the ability to send an e-mail message to "Majordomo@MR.Net" asking to subscribe. Furthermore, anyone could send an e-mail message to all the individuals subscribed to MN-POLITICS by sending a message to "MN-POLITICS@MR.Net". Mick Souder, an Internet educator and student at the University of Minnesota, agreed to be the list-manager, meaning he would watch over the unfolding discussion to make sure the technology and dialogue ran smoothly. Scott Fritchie, who would become the technical co-ordinator of the Project, decided to create an archive of MN-POLITICS at the site. He was able to do this using an application called Hypermail. This application posts the exchanges in an e-mail based listserv discussion to a Web site. As a result, all of the comments from participants in the discussion became available over the entire Internet through hypertext links. In this way a globally accessible conversation about politics in Minnesota was created.

The infrastructure for the Project was in place and open to the public by late August, in time for the primaries in early September, when the major parties would choose their candidates. As a result of the contacts made at the campaign offices, position papers on a variety of subjects by candidates running for the United States Senate and Governor of Minnesota had been solicited and made available. Furthermore, by mid-August a group of individuals had signed on to participate in the conversation in MN-POLITICS, creating another important dynamic to the Project. The stated purpose of MN-POLITICS is as follows:

The Minnesota Politics and Public Policy E-Mail Forum (MN-POLITICS) will promote the sharing of information on and discussion of Minnesota politics and public policy during the election season and beyond. The list encourages discussion from diverse political perspectives that is respectful in nature. This forum is more about the presentation of ideas and information than being right with one’s ideology [see APPENDIX 2]

Given the stir that had been building in Washington and around the world about information super-highways and electronic democracy, the early experiment was of interest to the media. On 1 September the metropolitan daily, the Minneapolis Star Tribune, ran a piece in the Metro section, "State Politics Exploring Cyberspace". The subheading of the Star Tribune piece indicated the angle of reporter Bob von Sternberg, "Fringe Parties, Political Junkies Beat Mainstream Candidates to Electronic Bulletin Boards". While the media had an eye on the Project, they were careful to emphasise the distance between the new technology and the mainstream. They did, however, allow Clift to express his view that the new technology could eventually alter the locus of political power. Von Sternberg quoted Clift, "It’s novel and new now, but a few years from now it could help set the agenda, determine how political power is distributed." Although sceptical perhaps, von Sternberg also raised the subject of the democratic potential associated with the new medium, writing, "In effect, the experience of MN-POLITICS bears out what computer aficionados have said all along about the Internet: It’s the ultimate democratizing tool, where everyone (and every idea) is equal."

Based on a press release distributed across the Internet by e-mail in early September the potential of the Project was clear. The effort possessed the features necessary to facilitate a Deweyan system of public opinion formation. As the press release said, "This is the first citizen-based, state-level, multi-candidate election effort that we are aware of in the United States." At the time no other project had the combination of features possessed from the start by the MN E-Democracy Project. It was designed for the electoral process; it was locally based; it was organised by citizens and civic organisations; it sought to distribute political information directly from the candidates; and it featured an interactive public forum. Most important was the combination of the contacts created with candidates and the e-mail based listserv for citizen dialogue being preserved in a Hypermail archive at the Twin Cities Freenet. Finally, the organisers possessed an experimental approach. They wanted to insert this into the political landscape in Minnesota and observe the results. As the press release concluded, "We hope to learn something about how electronic communication can help improve our representative democracy."

I became aware of the existence of the Project on 6 September, 1994 when I received the press release in another e-mail listserv forum in which I was a participant. Realising the potential of the Project, I decided to travel to Minnesota to conduct field research on the development of electronic democracy during the 1994 elections. I was particularly interested in observing the development of MN-POLITICS. Methodologically, it is important to situate my activities beginning with the move to Minnesota in relation to the continuing development of the Project. This is particularly important as, over the course of the Project, I became increasingly involved. Immediately after meeting with Steve Clift the opportunity for involvement was clear and hard to refuse. First, Clift and I possessed a compatible interest in observing and participating in how the new technology would change the distribution of power in the democratic process. It is not insignificant that Clift was a student at the University of Minnesota, working in a programme under the direction of Harry Boyte, who has written extensively on John Dewey and American pragmatism. Second, Clift had the attitude that he was the founder of an infrastructure in which volunteers would have as much latitude to get involved as they desired. Third, Clift had accepted a job at the Minnesota Government Information Access Policy Council, assisting in the design of State information services and the development of State policy. While he would stay involved in the Project throughout, his new position in government required that he step into the background. Not only did the MN E-Democracy Project possess tremendous potential, but organisers were eager for assistance.

The history of the Internet and the early days of MN E-Democracy outlined thus far has described a variety of agents acting in society, designing technology and/or organising systems that would be used in increasing numbers, leading to the design of new technology and/or the organisation of systems. This process would lead to actors implementing previously designed technology and organising the MN E-Democracy Project. In terms of the current discourse, the MN E-Democracy Project is like an uppermost layer of soil and rock covering one area of a massive edifice composed of many layers. When I arrived in Minnesota in September, I would become another agent acting on the edifice. For the sake of the present analysis it will be helpful to continue by offering two historical perspectives. The first is the history of the Project apart from any overt consideration of my agency. The second is the history of the Project in consideration of my agency. The reason that these two perspectives will be helpful is that my agency had a major effect on the development of the Project. Thus, the first history will offer a general overview of the development of the Project. In the second I will isolate some of the decisions I made that influenced the development of the Project.

Between 1 September and 18 October the public forum became an established venue for public discourse about the political season in Minnesota. The combination of the e-mail listserv technology and the Hypermail archive proved to be a very effective unit of political machinery. I will deal with the content of the archive in detail in later sections. Furthermore, for the purposes of the current discourse, the official history of the Project begins on 1 September, when Scott Fritchie created the archive. Although MN-POLITICS was created in early August the messages from participants were not archived automatically until 1 September. Because of the inconsistency of the archive early on, this period will be considered the pre-history of the Project and will not receive detailed treatment.

On 19 October an official announcement was made about the two political debates that would be hosted in MN-POLITICS on the Internet. The on-line debates were called Electronic Debates or E-Debates. They occurred between 23 October and 5 November. The first E-Debate involved candidates running for Governor of Minnesota; the second involved candidates for the United States Senate. A separate e-mail listserv, MN-DEBATE, was created to provide a secure candidate-only platform for formal debate. While MN-POLITICS was open subscription and open submission so anyone could join and contribute, MN-DEBATE was open subscription and moderated submission. Anyone could join MN-DEBATE, and therefore read submissions, but a moderator was in place who permitted only submissions pertaining to administrative matters and submissions from candidates. Not only did this provide a secure platform for the candidates but it allowed citizens the option of viewing the E-Debates without themselves participating. Of course as the E-Debates were forwarded into MN-POLITICS, citizens also had the option of joining in and commenting on the E-Debates along with fellow participants.

Each E-Debate lasted five days. The first ran from Monday, 24 October to Friday, 28 October; and the second from Monday, 31 October to Friday, 4 November. Five candidates participated in each E-Debate, including all major party candidates. The candidates were given three debate questions on the Saturday before the E-Debates. They were required to send a 500 word response to the first question to MN-DEBATE on the following Monday between 9:00 am and 12:00 pm. They were required to send a rebuttal to an opponent’s original response that afternoon. The candidates were required to send in responses and rebuttals to the second and third questions in like manner on Wednesday and Friday.

From the time of my meeting with Steve Clift I became an active volunteer, meeting other volunteers and working in Minneapolis. I was not always active in the subject of inquiry, MN-POLITICS, however. The general history of the Project breaks down into two periods - the development of MN-POLITICS and the history of the E-Debates. From my perspective, however, there were three qualitatively different periods. During the first period, from 6 September (when I joined the public forum) to 27 September, I observed MN-POLITICS. I submitted only a couple of messages seeking to add value to the work of other organisers. During this period, as MN-POLITICS developed, it was possible to learn about the new technology apart from my input.

During the second period, from 28 September to 18 October, I became an active participant in MN-POLITICS. I intended my contributions to experiment with what I discerned to be the potential of the medium and the particular implementation of the medium. For this purpose I created my own vehicle for communication within MN-POLITICS, a journal called AGORA - THE MN E-DEMOCRACY CHRONICLE. The name "agora" is from the name of the marketplace in which citizens discussed the issues of the day in the classical Athenian City-State. In the Statement of Purpose of AGORA I listed three aims: "1. To facilitate the democratic process in MN. 2. To discuss the construction of a new public space, a proto-type for an electronic townhall. 3. To explore the relationships between the new electronic medium and the traditional media." Through AGORA I aided the further development of the Project into a Deweyan system of public opinion formation. The primary strategy was to increase awareness of the presence of MN E-Democracy in the community and to involve actors in the political process in MN-POLITICS. The first entry was the story of my effort to establish the MN E-Democracy Project as an official member of the media by securing a White House Press Pass to a rally for Ann Wynia, one of the candidates for the Senate, at which President Bill Clinton would be the featured speaker. Subsequent entries included a critical account of the campaign process by a former campaign manager; a report of my experiences on the campaign trail with Republican candidate for Senate and eventual victor, Rod Grams; a piece written by a woman running for Lt. Governor of Wisconsin; and press releases in electronic form from Minnesota candidates. The last two offerings were part of a deliberate strategy to involve candidates in the deliberation taking place in MN-POLITICS. This strategy culminated in a challenge to the candidates to contribute to AGORA. Eventually, Steve Clift and I agreed that it would be possible to organise the E-Debates. Approximately four weeks before the election, I wrote a proposal for the E-Debates that I gave simultaneously to the candidates, the League of Women Voters of MN (who were asked to sponsor the debates) and the media (see APPENDIX 1).

Once the proposal had been submitted to the relevant parties, the other organisers and I began to publicise the E-Debates on MN-POLITICS, elsewhere on the Internet, and in the local and national media. The media took an interest in the Project, the League of Women Voters of MN agreed to co-sponsor the Project, and, one by one, the two major party candidates running for the United States Senate, as well as the candidates running for Governor, agreed to participate.

During the third period, from 19 October, I assumed the title of E-Debate Co-ordinator and was responsible for the design, organisation and co-ordination of the E-Debates. During this time I was no longer an active participant in MN-POLITICS. Instead, I became responsible for the conduct of MN-DEBATE and the flow of information from MN-DEBATE into MN-POLITICS. Thus, I had removed myself from MN-POLITICS and could once again view myself as an observer. This time, however, I was observing an information system I had helped realise.

There were countless decisions to be taken before the E-Debates had ended. Each decision was necessarily influenced by both practical concerns and a notion of what a democratic machinery required. First among the practical concerns was a calculus about what it would take to persuade the candidates to participate. For example, the primary goal leading into the E-Debates was to involve the major candidates for the Senate race and the Governors race. In focusing on securing the participation of the four primary individuals the fact that the E-Debates would be open to both major and minor party candidates was not emphasised. Once the major candidates agreed to participate the rest of the candidates were invited as well. The importance of this concession to practicality is that it emphasises the power wielded by candidates in negotiating the terms of participation. In other venues the candidates invariably backed out when it became clear that minor party candidates would be involved. They did not back out of this venue, I assume, because the campaign staffs realised that the venue was experimental and believed it unlikely to have a long reach into the population. Furthermore, the candidates were not required physically to type the responses and rebuttals to the questions themselves but were required to be involved in drafting the responses. This was necessary to minimise the effort required by the candidates to take part in the E-Debates and maximise the potential for discussion in the public forum. In fact, for the most part, the major candidates did not draft their own responses. The minor party candidates did and, as we shall see in a later section, this had significant repercussions in the forum and in the reports of journalists covering the forum for other media.

Second among practical concerns are a large cluster of decisions that resulted in the basic format of the E-Debates as described above. For example, there were two E-Debates running consecutively over a two week period. Each E-Debate consisted of three questions, responses and rebuttals, and each debate was stretched over a five day period. The intention was to concentrate on a few specific issues and stretch each debate over a week so that participants in MN-POLITICS would have time to discuss the E-Debates and issues as they developed. Furthermore, the inclusion of the rebuttal feature was an effort to highlight interactivity between the candidates; an encouragement to the candidates to learn how to use the interactive feature of the technology. Some used this feature better than others and therefore had an advantage in the forum. What is more important, in using the rebuttal feature effectively such individuals demonstrated how it can be a powerful tool for candidates in the future. Finally, two lists - MN-POLITICS and MN-DEBATE - were used. The manner in which these lists were implemented and interconnected reflects the democratic theory underlying the Project. As has been described, one list was an open forum for citizen dialogue and the other was a moderated forum for focused candidate debate. The lists were linked because the candidate debate was forwarded to the open citizen dialogue.

Several respondents commented on the lack of a moderator in MN-DEBATE and MN-POLITICS. In MN-DEBATE these respondents wanted someone to make certain the candidates answered the questions. In MN-POLITICS they wanted to impose some sort of editorial function on the many submissions. In MN-DEBATE I thought it unwise to impose strict rules on the candidates. It would be difficult to demand too much from them, especially as the forum reached only approximately 700 citizens directly. MN-POLITICS was not moderated either. There are many reasons for this. One reason is that the technology makes it difficult. Another reason already mentioned is that I wanted to minimise my role in the development of the public forum once it was underway. For methodological reasons I was, therefore, careful to separate my perceptions of events in MN-POLITICS from my function as E-Debate co-ordinator. Organisers did occasionally make suggestions to guide the forum along, however, by urging people to make contributions or setting limits on the number of posts once it became apparent certain individuals might well seek to monopolise the conversation.

Finally, the original intention was to develop the questions used in the E-Debates through consultation with participants in MN-POLITICS. Participants did not provide a substantive contribution at first. Perhaps they were sceptical about whether the debate would actually take place, as the candidates had not yet accepted invitations. I formulated most of the questions used in the E-Debates by gleaning ideas from conversations in MN-POLITICS, in conversation with Steve Clift, and in consultation with the League of Woman Voters of Minnesota. Once it was clear the E-Debates would take place, I again returned to MN-POLITICS to ask for suggestions. The second time, a number of replies were submitted from which the final question of the Senate E-Debate was fashioned. This demonstrates that it is feasible to ask the participants in the public forum in which the debate will take place for the questions if there is some certainty that a debate is, in fact, going to take place. I believe the best way to develop questions is through a consultation between the debate moderator and the citizens-participants.

The notion that underlying theoretical constructs influence the manner in which actor/s design a given technology or information system is here relevant (Orlikowski 1991B, 1992). The more specific notion that a vision of democratic theory underlies the formation of any given democracy project is significant as well (Arterton 1987). From the ideas set out by Vannevar Bush in The Atlantic Monthly after World War II to the development of the ARPANET and the spread of the technologies that would result in the Internet, there have been many actors working on the design and implementation of the new technology. In 1994 a further step was taken in applying the new technology to the political process. The MN E-Democracy Project was a part of this step. As Graeme Browning writes, "The Internet has been evolving in a linear fashion, from point to point, since the 1960s, but on October 18, 1994, (when the E-Debates were announced on the Internet) it took an abrupt turn straight toward the soul of the nation" (Browning 1996).

On 14 November, 1996, approximately one week after the 8 November elections, I submitted a survey to a group of 517 individuals still subscribed to the discussion forums, MN-POLITICS and MN-DEBATE. After my initial submission, I received 93 responses; after submitting the survey on 28 November, I received 40 responses; and after submitting the survey on 19 December, I received 54 responses. In total I received completed surveys from 187 individuals or 36% of the 517 individuals. This section offers a statistical overview of this sample population.

The sample population is a self-selecting group; people who heard about the Project and, for whatever reason, decided to become involved. They are also people who continued their involvement with the project for some time after the election and who, once involved, were the most willing to respond to a survey. One might, therefore, assume that these individuals are particularly motivated in terms of technology and/or politics.

It is not easy to compare the group that responded to the majority that did not. One can only assume that among those who did not respond there existed more who did not participate as actively, who were not impressed enough to respond, or who were not interested in participating in academic research. One certainty I should note is that there were many who did not respond for professional reasons. For example, a number of journalists participated and even wrote articles about the Project but did not return the survey. There were also many state and national government addresses among the list of participants. Some of these individuals may have understandably felt disinclined to respond. There were, nonetheless, a number of professional journalists and individuals in government who did respond.

Before turning to the survey data, it is possible to gain some insight into a larger 600 person sample of Project participants through an analysis of e-mail addresses. First, an e-mail address is made of several pieces of information that increase in generality as one moves from left to right. For example, I have two different e-mail addresses, "gsa1001@cus.cam.ac.uk" and "aikens@freenet.msp.sp.us". On the far right is the domain name, ".uk" and ".us", which signals the nature of the e-mail account. ".uk" signals the United Kingdom and ".us" the United States. Individuals and/or organisations request a category when they apply for a domain name for the purpose of creating a customised e-mail and/or Web address. Although country codes are used in these examples there are other categories such as educational institutions in the United States (.edu), companies (.com), organisations (.org), government addresses (.gov), network specific addresses (.net) and military (.mil). The use of country codes is most common among organisations in countries outside of the US. In addition, other countries have their own institutions to process domain names, and alternative abbreviations. For example, in the US, educational institutions are (.edu). In the UK academic institutions are (.ac). Moving to the left, an e-mail address begins to differentiate the nature of the account. In the first example, "cus.cam.ac" informs one that the e-mail address is an academic account (.ac), from the University of Cambridge (.cam), and more specifically from the computer user services (cus) at the University. In the second example, "freenet.mpl.sp" informs one that the account is located in Minneapolis/St Paul (mpl.sp) at a freenet. The symbol @ separates the full domain name on the right from the personal identifier assigned by the service provider to the individual user on the left. Thus, "gsa1001@" and "aikens@" are the personal identifiers assigned to me by the University of Cambridge and the Twin Cities Freenet, respectively. Because there are many complex variations in an e-mail address, I will concentrate on the domain names for the current sample of Project participants. Chart 3.2.1a provides a breakdown of participation by domain name.

Summary:

Domain names: The two most common domain names were ".edu", signifying users from American educational institutions, at 52% of the total, and ".com", signifying users gaining access from private businesses, at 30% of the total. The majority of users from ".edu" accounts, at 43%, were from the University of Minnesota. Other ".edu" accounts included a number from Minnesota schools such as St Olaf and Hamline, as well as a scattering of participants from schools around the country such as Harvard, Duke, Stanford, the University of Wisconsin and Berkeley. The majority of those gaining access through ".com" addresses possessed accounts with national and local Internet Services Providers (ISPs). Among ".com" addresses 34% possessed accounts with American Online, the largest ISP in the country. Other .com addresses were used by individuals with large companies based in the Minnesota region, such as Westlaw.com, Honeywell.com, 3M.com and Cray.com. A handful of the .com addresses were used by media organisations such as the Minneapolis Star Tribune. Nine per cent of the sample had ".us" domain names and these included all freenet accounts, 5 State Senate addresses, 5 State House of Representative addresses, and a handful of K-12 addresses. The ".org" addresses were used by individuals affiliated with non-profits such as Minnesota Public Radio. The ".net" addresses were used by organisations closely associated with the Internet. The ".gov" addresses were used by 4 from the United States House of Representatives, as well as other governmental organisations in the USA and, in one instance, Australia. There were 7 individuals with domain names from foreign countries, including Finland (.fi), the United Kingdom (.uk), Australia (au.) and Denmark (dk.). Finally, two were from the United States military, ".mil".

The e-mail addresses make it clear that the sample includes a mixture of people gaining access to the Internet through a wide range of methods. Besides the general trends that emphasise the importance of both the educational community and companies there are many cases where a user is clearly located in an intermediary political and/or media institution. As has been mentioned many of the domains in the group indicate an affiliation with, for example, state and federal government offices and agencies, The Minneapolis Star Tribune, Minnesota Public Radio, WCCO TV or The Utne Reader magazine, etc. Furthermore, many of the anonymous .com addresses were used by national journalists such as Peter Lewis from The New York Times, Graeme Browning from National Journal and David McIntyre from Deutsche Presse Agentur, as well as national political figures such as Richard Bell, special assistant to the chairman of the Democratic Campaign Committee, and Jock Gill from the White House. Finally, the campaign staffs of the candidates also set up e-mail accounts, whether, like Ann Wynia and Rod Grams through American Online, or like many minor party candidates through a complimentary Twin Cities Freenet account.

In the following I summarise the personal information supplied by the 188 individuals who responded to surveys.

Age: The median age of the sample population was 35.5.