Mapping Patterns of Polysemy tree, firewood and fire in the languages of Sahul Rachel Hendery, Antoinette Schapper & Lila San Roque University of Western Sydney, Leiden University/Cologne University, Radboud University/MPI Nijmegen Conclusions  The claims in the literature about polysemy of tree, fire, firewood, etc in Sahul cannot be upheld.  These terms behave quite differently in Australia (where the abb pattern predominates) vs the rest of Sahul, where full differentia- tion is more common.  The distribution of the different patterns corresponds mainly with language family distribution, but there is some evidence of trans- fer of patterns to unrelated languages nearby.  The availability of electronic dictionaries, digital grammars, and- software linguistic mapping software makes it possible to test statements previously based only on anecdotal evidence. References Dixon, R. M. W. The Languages of Australia. 1980. Cambridge: Cambridge University Press. Dryer, M. S. and Haspelmath, M. (eds.). The World Atlas of Language Structures Online. 2013. Leipzig: Max Planck Institute for Evolutionary Anthropology. Available from: http://wals.info/ Laycock, D. Papuan languages and the possibility of semantic classification. Papers in New Guinea Linguistics No. 24, 1986. Pacific. Linguistics A-70: p. 1-10 It is not possible to include all the references our data are drawn from on this poster. Please see the VR overlay for details (via Aurasma app). Background “Some – but by no means all – Australian languages take the principle of having a single term to describe some natural object, and also something that can be made from it, to the extreme of having a single lexeme covering both 'tree, wood' and 'fire'” (Dixon 1980) “The main conflation to look for here is that of ‘tree’ and ‘fire’ – via the intervening concept ‘firewood’. It is found in Foe, and is reported to be common in [Trans-New Guinea Phylum] languages.” (Laycock 1986) There has been no systematic study of the distribution of these polysemy patterns across Sahul. Advances in the use of databases, electronic dictionaries, and digital mapping in the field of linguistic typology in recent years have made it possible to conduct more extensive surveys of such patterns. Methods We surveyed 216 Papuan languages and 40 Australian languages. We gathered terms from electronic dictionaries, word- lists, and grammars. We then used the software provided by the World Atlas of Language Structures Online (WALS) to map the distributions of polysemies. The advantage of using the WALS mapping software is that it allows us to add in layers of other linguistic features, so that we can build on this project in future by looking at correlations between these polysemy patterns and other features of the languages, or checking whether the groupings we find correspond to known genetic or areal groups. Papuan Australian TOTAL aaa full conflation pattern 19 4 23 aab TREE/FIREWOOD conflation 26 1 27 aba FIREWOOD differentiation 9 0 9 abb FIREWOOD/FIRE conflation 70 32 102 abc full differentiation pattern 92 3 95 Table 1: Number of languages with each pattern Why does this matter? If the languages of Sahul share an unusual property (like having a single term for tree, wood and firewood), this adds to the evidence that they have had a long history of contact with each other.