6.     Conclusion

Conversation in SCMC is quite different in many ways from spoken conversation. It follows that established approaches to spoken discourse analysis do not necessarily map directly on to a novel form of discourse. For example, as shown in this paper, patterns of turn-taking in SCMC are affected by disrupted turn adjacency, itself a characteristic of the discourse setting ? the virtual environment. Hence certain axioms concerning turn taking in spoken discourse do not apply to conversation in the computer-mediated discourse setting. In this paper it has been maintained that the notion of the conversational floor is a useful one in the study of discourse where cohesion is looser than in the spoken mode. Furthermore, a claim has been made that the development of certain floor types is associated with (a) the roles of the participants in the discourse; (b) the topic of the discourse; (c) the current communicative action or, generally speaking, the purpose of the discourse.

The conclusions can be broadened by engaging with the notion of communicative competence (Hymes, 1972; Canale and Swain, 1980). Effective participation in a particular SCMC environment requires a measure of electronic communicative competence. The elements of electronic communicative competence, as they apply to the context and discourse features described above and adapted from the model of Canale and Swain, include the following:

A knowledge of the linguistic system. The Webheads virtual community includes both learners and expert users of English. However, only a minimum level of English is needed to communicate effectively in SCMC discourse; perhaps less than is needed for similar communication in the spoken mode. The speed of turn-taking is slower than in spoken discourse; participants can scroll back up the screen to re-read parts of the conversation, and logs of the text can be saved and studied at a later time. There are thus arguments for the use of SCMC in language teaching.

A knowledge of the discourse patterns involved. The view of cohesion posited here suggests that it operates through the organisation of various types of conversational floor. For participants, managing these floors and perhaps partaking in different floors in parallel, requires new skills. Regardless of one?s level of competence in the language of the virtual environment, the ability to manage threads of SCMC discourse is a primary skill.

A knowledge of the technology. This knowledge encompasses both access to the technology (the computer hardware and an internet connection) but also a technical knowledge enabling a participant to download particular software, to log on to the system, and to join a virtual community amongst other things.

A knowledge of the sociocultural rules of a particular virtual community. Not all SCMC settings are the same. The final aspect of electronic communicative competence includes a knowledge of the roles of participants, the topic range expected in the context, and the broad purposes of communication in the context.

This paper has only touched on these matters by looking at discourse patterns in one particular SCMC context. There is undoubtedly much scope for the investigation of other areas of floor development in SCMC, and more generally, applying any findings to a nascent theory of electronic communicative competence.