In this paper, I argue that distinguishing between attributions of sentience and agency is a crucial analytical step to take when considering ontological and cosmological differences between groups. Scholars in 'New Animist' and Actor-Network Theory camps sometimes fail to make this distinction adequately and overlook the ways in which characterisations of different kinds of beings, and the implications of these characterisations, might change historically. Examining Australian Aboriginal practices and descriptions of spirit figures and sacralized landscapes, I argue that there is in fact an epistemological incommensurability between indigenous and non-indigenous experiences of human-environment relations, but that there has been a partial transformation of the former, involving a reduction of the previous emphasis on nature- and place-based spirits.
The spirit figures that have endured, and thrive, in Aboriginal Australia are most often human-like and human-oriented, especially ancestors and beings associated with death. I suggest that this ontological shift ultimately derives from changing relationships of Aboriginal peoples to the natural environment, and contributes, however tacitly, to creating new understandings of the category of the human.
Francesca Merlan has been a professor of anthropology at the ANU since late 1995. She has done research in northern Australia since 1976, Papua New Guinea since 1981, and Germany since 1999. She has focused on people-land relationships, attachment and dislocation, colonization and the ethnography and theory of social and cultural change. Her books include Ku Waru:Language and Segmentary Politics in the Western Nebilyer Valley (with Alan Rumsey, 1991); Caging the Rainbow: Places Politics and Aborigines in a North Australian Town (1998); Dynamics of Difference in Australia: Indigenous Past and Present in a Settler Country (2018); together with various ethnographic documentary volumes and works. Although she never intended, really, to enter into the `ontological turn' debates, this paper is about as close as it gets. This paper also draws upon and shows the value of the Australianist ethnographic record, despite all its time-bound qualities and distortions.