Published online by Cambridge University Press: 12 March 2026
A number of Australian Aboriginal languages have syntactic mechanisms which can be analysed as instances of ‘switch-reference’. These languages have verb suffixes indicating whether or not the subjects of syntactically-related main and subordinate clauses are referentially the same or different. The distribution of switch-reference in various subordinate clause types is examined here, and its formal and functional characteristics are described in detail for a number of language groups. The languages which have switch-reference are all geographically adjacent, but are not obviously related genetically. When the geographical spread of certain formal features such as the association between locative case and some ‘relative clause’ types is discussed, a number of general conclusions can be drawn, among them the suggestion that the syntactic mechanism of switch-reference has been subject to indirect functional diffusion between language groups. It is further suggested that careful attention to the precise nature of syntactic similarities is necessary if any attempts to reconstruct the syntax of proto-languages in Australia are to be undertaken. This paper provides further evidence of the importance of linguistic diffusion in Australia, and raises a number of issues relevant to studies of areal features elsewhere in the world.
This paper was written while I was a post-doctoral fellow in linguistics at UCLA and MIT: I am grateful to the Commonwealth Fund of New York for its generous support through the Harkness Fellowships program. Earlier versions appeared in Austin 1978, 1979. For comments, criticism, and/or provision of additional data, I wish to thank Barry Alpher, Barry Blake, Gavan Breen, R. M. W. Dixon, Ken Hale. Shelly Harrison, Luise Hercus, Rodney Huddleston. Terry Klokeid, Harold Koch, Mary Laughren, Patrick McConvell, Pam Munro, David Nash, Geoff O'Grady, Sandra Thompson, and Wilf Douglas. Errors of fact or interpretation are my responsibility.