Abstract

Although it is conceded that distinct knowledge domains do presentparticular problems of coming to know, in thispaper it is argued that it is possible to construct a domain independent modelof the processes of coming to know, one inwhich observers share understandings and do soin agreed ways. The model in question is partof the conversation theory of Gordon Pask. CT, as a theory of theory construction andcommunication, has particular relevance forfoundational issues in science and scienceeducation. CT explicitly propounds a ``radicalconstructivist'' epistemology. A briefaccount is given of the main tenets of RC andCT's place in that tradition and the traditionsof cybernetics. The paper presents a briefnon-technical account of the main concepts ofCT including elaborations by Laurillard andHarri-Augstein and Thomas. As part of CT, Pask also elaborated a methodology – knowledgeand task analysis – for analysing the structureof different knowledge domains; thismethodology is sketched in outline