Ilana Mushin: 'Are there really free word order languages?', 21 August 2015
When: 3.15pm-4.15pm, Friday 21 August, 2015
Where: Room 407, Babel (Building 139), University of Melbourne
One of the parameters of word order typology is the degree to which word orders in particular languages are claimed to be 'fixed' rather than 'free'. Languages with fixed orders are usually taken to be those, like English, where ordering of arguments with respect to their predicates is one way of indicating grammatical relation (eg subjects precede verbs, objects follow verbs). Under this interpretation, free word order languages where the position of heads and dependents is not dependent on the grammatical role played by those constituents. Rather ordering principles should be accounted for by pragmatic principles such as information status (eg. Firbas 1992, Mithun 1987).
In this paper I question what it means for a language to be classified as a 'fixed' or 'free' word order language using data from the Australian language Garrwa. I use conversational data to show how ordering patterns in a so-called 'free word order' language are in fact quite constrained if one takes into account the emergence of talk in real time, which must factor in prosody, timing and speakership. The results thus raise questions for how typologies of word order should be developed.