Abstract
The features of ethnobiological taxonomies in Australian Aboriginal languages present significant challenges to Berlin et al.’s (1973) universals for folk taxonomy. Authors have shown that environmental/biological, cultural and linguistic features each contribute to the distinctive principles of ethnoclassification in Australian languages (e.g. Waddy 1988, McKnight 1999, Baker 2007), such that they cannot be accommodated in Berlin et al.’s model. Bordulk et al. (In press) is the result of nearly twenty years of field-based collaborative fieldwork between Dalabon speakers, linguists and biologists. The authors document biocultural knowledge on 550 plants and animals in the Dalabon language of south-western Arnhem Land, Australia. This is a relatively high number of species for Aboriginal languages in this latitudinal zone in north Australia, which reveals an unusually dense ethnoclassification system in Dalabon. In this paper, I seek to report on the principles of ethnobiological classification in Dalabon by describing both the familiar and unfamiliar features of this taxonomy. I begin by describing those features of Dalabon ethnobiological classification which are commonly attested in Australian languages (e.g. dominant use of monomorphemic, mononomial names at the species level; few if any genera terms; polysemies based on metonymic relationships between biologically unrelated taxa; onomatopoeia). I then present a linguistic analysis of the categorization principles. I explore the relationship between biological ‘taxa’ and linguistic ‘senses’ and highlight the semantic principles underpinning the polysemy, sign metonymy and (partial) reduplication attested in the Dalabon data. This paper is an example of the depth of analysis which can result from cross-disciplinary documentary fieldwork, and is a contribution to the description of the Dalabon language and of Australian ethnoclassification systems, and to ethnobiological taxonomies more generally.
