1 Introduction

Using telecommunication, any subscriber can in principle make a connection with any other subscriber no matter their location in the world. Communication knows no cultural, religious or political divisions, but spans the globe, space oceans, and land masses through co-operation. In war or peace, using diverse equipment and networks, no matter the language, procedures, or age of equipment, the connections are made. We live in an age of social development dependent on a comprehensive and intricate telecommunication infrastructure. (Espvik 1997)

1.1 The Internet and its surrounding discourse

The Internet forms a rich and complex subject matter for science, technology & society (STS) analysis, being a conglomerate of various technologies that is expected to have far-reaching effects on peoples' everyday lives across the globe, and which has enabled novel forms of human interaction to be developed.

The Internet (or 'the Net' for short) started out as a military inter-city communications network (then called Arpanet) in the USA in the 1960s, designed to withstand a nuclear attack by exhibiting a high degree of redundancy in its mesh of interconnections. The Net was opened up for academic use in 1969, and the first connection outside the USA was to Norway in 1972. The invention of the World-Wide Web (WWW, or 'the Web') at the nuclear physics research centre CERN in Switzerland in 1990 simplified the user interface and paved the way for use of the Net by the general public.

In the 1990s the Internet has been developed at a frenetic pace, and recent developments include increased commercialisation and the advent of multimedia applications offering video and audio transmission in addition to the traditional text-based communications services such as electronic mail (e-mail), discussion forums (Usenet news) and the Web.

The increased social and economic relevance of the Net has resulted in more and more people getting involved with it in one way or another. First there has for several years been a nearly exponential rise in the number of users, i.e. people who treat it as a communication tool at home or at work. There is a large community of technical developers (programmers, engineers and scientist) who design, implement and maintain the technological infrastructure. A number of business developers have entered the field, hoping to find ways of using the Net to generate profit. Along have also come lawyers who struggle with intellectual property rights issues and the contradictions between a fundamentally transnational medium and nation-bound legislation. Politicians have started to realise the importance of the medium, as representing both an opportunity and a threat regarding democracy, the economy and their own way of working. Finally, there has emerged a hoard of analysts that try to grapple with various aspects of the medium (technical, social, political, economic, philosophical, and so on).

As with any new communications medium, the 'nature' or 'impact' of the Internet is far from clear. Many commentators and analysts leap to chance of becoming the first to claim that they have somehow 'understood' the new medium, including its alleged potentials and future impacts. A plethora of hastily concocted narratives thus emerge, providing a blend of analysis and speculation on various aspects of the medium, in particular those relevant for making business (Jones 1999: 10-11). Extrapolation based on barely identifiable tendencies is a common technique.

The danger is that more rigorous studies based on robust empirical foundations (and which tend to result in more sober accounts) may drown in a wave of hype. There is also an element of self-fulfilment in prophesies regarding 'revolutionary' new technologies; if everyone believes they will be out of business unless they jump on the bandwagon, then they will jump. From here the road to technological determinism is not long. This indicates that the discourse around the medium is as important as any 'inherent' properties it may exhibit.

Each of the above mentioned categories of involved people consists of a heterogeneous set of individuals. Developers have different interests and visions for the medium. Users exploit the tool in multifarious ways, resulting in differing views of the capabilities and usefulness of the medium. To some business people the Net is a threat; to others it is a holy grail. Some politicians have ambitions to steer the development of the medium in order to achieve specific socio-political aims; others want to leave this to the business community (the 'market'). So within these broad categories there are subsets of people with a common vision, interest and conceptual framework. These sub-categories form what is in Social Construction of Technology (SCOT) studies called relevant social groups. Their shared conceptual framework is known as a technological frame (Bijker 1995).

The different technological frames results in different interpretations of what the Internet is and where it is heading. The resulting discourse around the medium can be regarded as not to be revealing some a priori existing 'nature' or 'reality' of the medium, but instead as forming this very reality; the 'nature' is the outcome of the discourse. This discourse not only occurs around the Internet as a single entity, but also around the various components of the Net. The Internet has many aspects, both social and technical, and consists of a complex set of technological artefacts. Treating the Net as a single entity is likely to be too coarse-grained.

1.2 The Internet as a platform for social formations

These general remarks about the Internet as a whole also apply to its component parts. There are many ways of 'looking into' the Internet and identifying such components. In purely technical descriptions a conceptual model consisting of hierarchically layered components is often used. Many socio-technical or socio-economic accounts also employ such a model. The Bangemann report, for instance, uses the model depicted in fig. 1.1 (HLGIS 1994: Ch. 4).

The bottom layer represents the physical networks, i.e. the network nodes (e.g. computers, telephone exchanges and satellites) and their interconnections (e.g. fibre-optic cables, copper wires and antennae). The networks are the ultimate carriers of the information. Basic services are the fundamental software components that allow people to use the networks. Examples are e-mail, WWW, search engines and video transfer. Applications are higher-level multimedia systems that use the basic services as components. Examples are videoconferencing, distance learning and teleworking.

Fig. 1.1 represents a techno-centric view of the Internet in which an important social development is invisible. This is the use of the basic services in order not to build a technical component, but to build a social 'component'. Use of e-mail, on-line discussion forums and chat software has facilitated the build-up of social relations between people who have never seen each other and who communicate exclusively via information and communication technology (ICT). This is illustrated in fig. 1.2.

Some of these social formations have many features in common with communities found in the wider society, and have been denoted 'electronic communities'. The two models complement each other, and analyses that only use one of them are bound to represent a partial view. These models will be used as a basis for analysing various aspects of computer-mediated communication (CMC).

A central issue is the relationship between the social formations and the technology. This is an underdeveloped area in social studies of the Net, since the technology is seldom problematised at all. The technology can be viewed as a filter for the symbolic interactions between people communicating via ICT, and the properties of this filter influence the specific form that computer-mediated social relations assume.

Fig. 1.2 only shows one aspect of CMC, namely its dependence on technology, and is thus also to some extent techno-centric. Another issue that this thesis will explore is the relationship between on-line social formations and their socio-cultural environment. The model will therefore be augmented with various social attributes, depending on which aspect is being focussed upon. A central theme is to demonstrate how CMC is rooted in the wider society, and thereby counteract some of the hyped-up and distorted views that characterise much of the discourse around these phenomena. The hype is not only characteristic of popular discourse but seems to have been inherited by many academics, and the result is a sometimes diffuse and incoherent discourse (Hakken 1999: Ch. 1.2.4). I will therefore take a critical look at the presumptions, vocabulary and metaphors used in mainstream analyses.

Fig. 1.1 and 1.2 indicate that there is symmetry between the technical and the social. Such symmetries will be a reoccurring theme in the thesis, as they can be identified in several aspects of Internet-related developments. The focus on these symmetries and on the interplay between the social and the technical represent the predominant STS-aspects of this thesis.

As this disposition indicates, the object of analysis is more on the framework of electronic communities, rather than on these phenomena in themselves. This framework comprises the discourse that surrounds these social formations and their relationship with other social and technical spheres.

1.3 A note on method

I have tried to take a critical and analytic approach to the subject matter. As already stated, the aim has not been to provide a first-hand account of the nature of on-line social formations. The subject matter has rather been other people's accounts of them. My analysis is founded on a hybrid of technical literature and social studies. Technical literature normally exhibits a higher degree of precision in its expositions, which is useful for establishing some analytical foundations that can be applied in an attempt to counteract the diffuseness that is prevalent in many studies.

Attributing explanatory primacy or preference to either the technical or the social has generally been avoided, a feature which is shared with Actor-Network Theory (ANT). ANT has however not been applied since the abstraction level of ANT has been regarded as somewhat of an overkill for this study. The STS framework that is drawn most heavily upon is instead SCOT, a framework suited to analysis of discourse around socio-technical ensembles. The Internet has also proven to be a good example of interpretation activities around a technology, which is central in SCOT.

A problem with SCOT is that it has traditionally been used to study fairly well-defined artefacts. The Internet is more of a complex of artefacts, and still undergoing wide-ranging changes. To try to overcome this problem, the focus will be on the interplay between some of these sub-artefacts, in order to try to identify generic development patterns that are relevant for the study of electronic communities.

CMC and on-line social aggregations lie at the intersection between the technical, the interpersonal and the wider socio-cultural environment. Any reasonably adequate analysis of these phenomena is, at least from an STS perspective, bound to involve multiple social and technical aspects. For a balanced analysis one should take into account as many as possible of these aspects, but it is then not possible to go very far or deep in any one direction, in order to keep the treatment within reasonable bounds. I have attempted to counteract such shallowness by focussing on phenomena that exhibit an increased degree of generality or symmetry.

1.4 Thesis organisation

The dissertation is organised as follows. Chapter 2 takes a critical look at the concept of virtuality, one of the most widely used terms for describing CMC. It is also one of the most problematic, as it leads to certain ontological peculiarities in the treatment of on-line social formations. Associated with the 'virtuality' concept is the notion of mediation, a term whose selective application has led to a problematic analytical chasm between mediated and non-mediated interactions. An argument for a more general applicability of the term will be provided, which enables a more symmetric approach to be taken between the social and the technical.

The results of chapter 2 are subsequently used in a review of the concept of community in chapter 3. As a vaguely defined term associated with images of dubious historical accuracy it is not a readily suitable entity on which to base analyses of new community-like formations. It will be argued that the traditional view of community needs to be somewhat modified to provide a more rigorous basis for analysis. This line of thought leads on to the central element of the thesis, electronic communities. As these have been subjected to numerous studies from a sociological viewpoint, the social content of these communities will only be dealt with briefly. Instead the focus will (in chapter 4) be on the interplay between the social and the technical which I believe is central for accounting for certain characteristics of electronic communities. The analysis will take a critical stance towards traditional face-to-face social interaction, and call for a more symmetric treatment of traditional and on-line communication.

Finally, chapter 5 uses the development of electronic communities as a point of departure for a socio-technical analysis of the Internet in general, where an argument will be presented that social and technical formations essentially follow a common dynamic pattern. This is where SCOT will be applied most explicitly.