The aim of this project is to investigate qualifying processes of normative change that arise when novice scientific authors collaborate in distributed virtual environments.
These processes will be mirrored in the empirical data obtained in the course of the investigation as changes in individual textual norm systems and positioning behaviour. The data consists of many different types of texts from different types of participation and observation situations. They include log and observation materials which are not directly categorisable as either "written" or "oral" texts, but which lie on a cline between these two extremes.
Part of my approach is grounded notions adopted from sociotextological theory (see Berge 1990, 1993). Sociotextology has its roots in the Prague structuralist school, the theories of Jan Mukarsovsky' and the Russian Bakhtin-Volosjinov school, and sociolinguists and social anthropologists such as Labov and Waletzsky. Sociotextology attempts to understand, describe and explain, how textual norm systems are constituted in socio-cultural contexts. This is studied from the perspectives of both text-internal functions which cause text components to be put together in some kind of coherent whole (endotextual functions), and text external functions that position actors in a communicative situation in a normatively controlled, interpersonally and socially binding relationship to one another (exotextual functions). These obligatory relationships are related to a medium (e.g writing or speaking) because they must be communicated via a medium. There are also certain normative expectations with regard to how the medium shall be structured in order for the obligational relationships to become manifest.
A textual norm is developed as a result of some cultural labour carried out by some actors in a social field (S), who all attempt to reach one or more communicative objectives (M) with one or more functions (F). The norm N is consequently developed as a generative system adequately adapted to the objectives and functions of the social field. This relationship may be formulated formally in the following way:
Ms->Fs->Ns
A social field is more generally defined a system of praxis relationships which is attained by more or less spesialised agents and institutions in the struggle for acquisition of some kind of symbolic capital (Berge 1990, p. 112)
This theoretical approach is being used and developed at the present time in a study of the evaluation criteria and student writing strategies that are developing as new examination question types for final examinations in the general studies curriculum in Norwegian at secondary school level in Norway (see Berge, 1994a, 1994b and in press). The aim of that particular project is to understand, describe and explain the prevailing text norm systems that actually are in operation among candidates and censors in this rather limited social field (i.e. the field of Norwegian culture related to the various actors involved in some way or other in the examination evaluation process) which, due to the national educational reform process is itself undergoing rapid processes of change. This means that also official documents and memos sent to schools by the Norwegian Department of Education are considered as relevant messages within this particular field. Also academic debate and discussion articles in periodicals and journals published by institutional actors in the field: teacher organizations and special interest groups in disciplines such as pedagogy and education are relevant texts in this context.
The virtual environment is constituted at a metalevel by the technology of the mediating system on the basis of relationships between stratified sets of semiotic systems mediated by means of different codes (e.g. the programming code, user-accessible codes that regulate non-verbal and verbal functions and the written language code used to communicate messages as texts). These codes and systems are in continuous interplay with each other and contribute each in their own way to the constitution of systems of endo- and extra-textual functions in the virtual environment. They will consequently set important parameters for (or rather constitute an own meta-norm system related to), how actual (but nonetheless virtual) socio-cultural processes of norm constitution and change are realised as texts in the virtual environment.
Socio-semiotic systems like languages and cultures (and this might just as well be applied to the social fields that are now being constituted in virtual environments: the so-called digital realm of culture) a re dynamic open systems, with a common characteristic of being "metastable" (Halliday 1987, Lemke 1993). Metastable systems survive and develop only because they are in a constant process of transition and change; and this inherent dynamicity is maintained only through continual interactions with other systems that constitute their immediate environment. According to Halliday the constructed metalanguages of science (such as those of formal logic and generational grammars) are too deterministic and rigid to cope with and describe complementarity relations, which is precisely what spoken language has evolved in order to do. It is what is used to cope with categorizing the flow of changing and developing meanings that arise as we move around in, and interact with other living and material processes in the real world. Scientific writing tends on the other hand to reify living processes. Historically, the growth and development of scientific writing shows that this is achieved through processes of grammatical metaphor (Halliday 1995), where there is a move from descriptions of observations of naturally occurring processes to the production of nominals and nominal groups which function as specialised terms related to (and often only understandable through socialisation over time into the discourses of) the various scientific disciplines to which these terms belong. This has over time formed the basis for what Kuhn (1995) refers to as "the incommunsurability problem": communicating knowledge between and across all fields of modern science has become a highly complicated (and some say, almost impossible) translation and interpretation process.
What is unique about the kinds of social communities that develops in distributed virtual environments, besides that fact that they to all extents and purposes are based on the medium of writing (or in a wider sense, text production and interpretation), is that participants are also able to qualify themselves through writing to take part in the construction of the text- based virtual environments that they are "inhabiting" and which other actors can "enter into". Everything that happens in the virtual world, both construction and development of the "physical" and "social" environment, and the various kinds of interactions between actors may be logged by the database software[8] of the MOO at any given time. All communicative events and acts: verbal and non verbal interactions between participants, actions related to the construction of the virtual environment, and at a higher level of abstraction: all processes of meaning making that are going on during the constitution of the textual norm-systems of the virtual environment are stored as textual artefacts as long as anything is happening there. In this way a running ("on-line") "transcript" which is a textual representation of the semiosis of the virtual space is obtained. This gives a unique possibility of studying participant interactions through transcripts where norm constitution processes are actualised as texts at the same time as they are instantiated and construed by participants.
In studying the development of interactional norms systems actualised as texts in the virtual environment I will adapt a communicationally oriented phenomenology which combines the idea of metacommunicative frames (Bateson & Ruesch 1968) with descriptions of how the borders between various communicative spheres (e.g. the story world, the sphere of narration, the sphere of conversation etc.) influence participants' ontological experiences of the situation of communication (Young 1982). This basic approach must however be modified in order to allow the treatment of textual materials of other types such as citations, written commentaries, larger or smaller text fragments and memos and notes presented in response groups, pointers to other texts and discourses which run into one another in the log files.
The above considerations are one important reason why I have chosen to report from this particular project by means of an ethnographic narrative. This makes it also easier to incorporate reports and observations from othe participants into the process of describing what is "going on".