Linguistic forms with dedicated evidential meanings have been described for a number of Australian languages (eg. Donaldson 1980, Laughren 1982, Wilkins 1989) but there has been little written on how these are used in social interaction. This paper examines evidential strategies in ordinary Garrwa conversations, by taking into account what we know more generally about the status of knowledge and epistemic authority in Aboriginal societies, and applying this understanding to account for the ways knowledge is managed in `ordinary' interactions.

Hart offers a biologically based explanation for the use of an ‘epistemic positioning strategy’ aimed by speakers/writers at the legitimization of assertions, at persuading addressees of the veracity of the propositions, as a prior condition for the discursive legitimization of actions. This article focuses on various issues addressed in Hart’s article, among them the degree of commitment invoked in speakers/writers’ choice of epistemic stance expressions as legitimization strategies, as well as the expression of subjectivity/intersubjectivity in discourse and the degree to (...) which this involves responsibility and accountability for the speaker/writer. The article also provides an alternative proposal of the categories identified by Bednarek for the conceptual domain of evidentiality. (shrink)