Loanwords between the Arandic languages and their western neighbours: principles of identification and phonological adaptation

Abstract

This paper summarises the characteristics of loanwords, especially the ways in which they are adapted to the structure of the borrowing language, and surveys the various tests that have been provided in both the general historical linguistics literature and Australianist literature for identifying the fact and direction of borrowing. It then provides a case study of loanwords out of and into the Arandic languages; the other languages involved are especially Warlpiri but to some extent dialects of the Western Desert language. The primary focus is on the phonological adaptation of loanwords between languages whose phonological structure differs especially in the presence vs. absence of initial consonants, in consequence of earlier changes whereby Arandic languages lost all initial consonants. While loanwords out of Arandic add a consonant, it is claimed that loanwords into Arandic include two chronological strata: in one the source consonant was preserved but the other (older) pattern involved truncation of the source consonant. Reasons for this twofold behaviour are presented (in terms of diachronic and contrastive phonology), and the examples of the more radical (older) pattern are individually justified as loanwords, using the criteria discussed earlier in the article.