Lexicostatistics and Australian languages: Problems and prospects

The present paper begins by pointing out the limitations of more traditional comparative linguistic approaches and why we can expect lexicostatistics to perform better in Australia. Since it has sometimes been argued that the peculiarities of Australian languages could limit the applicability of lexicostatistics, I then draw on previously unpublished work to show (a) why name taboo cannot be expected to affect lexical change in Australia to anywhere near as great extent as has sometimes been imagined, (b) how proposed cases of rapid lexical change are not supported by the available evidence, and (c) how lexicostatistical approaches can be used to detect and cope with cases of heavy borrowing.