This paper discusses the development of nasal prestopping in Umbuygamu, a Lamalamic language (Paman < Pama-Nyungan) of Cape York Peninsula in northeastern Australia. Specifically, the analysis covers the origins of prestopped nasals, variation in their realization, and the loss of prestopping in some speakers. While the origins of prestopping are not typologically surprising, the range of variation in its realization is, with preaspirated allophones that as far as I know have not been reported in Australia. I also argue that loss of prestopping in the youngest generations of speakers is not necessarily due to language loss, but may follow a system-internal pathway related to alternations with plain nasal allophones.

Examples (1) and (2) illustrate how prestopped nasals developed from plain nasals, as a new contrastive nasal series. The pathway of development is similar to that observed, for instance, in Arandic languages and their neighbours (Koch 1997, Simpson & Hercus 2004), and consists of initial consonant loss followed by medial strengthening if the original initial was not nasal.

(1)     PP *pama (Hale 1976a)   > apma       ‘person’

(2)     *ngama (Alpher 2019)     > amay       ‘mother’

The Umbuygamu corpus, which covers 12 speakers, shows a surprising range of allophonic variation for the prestopped series. Apart from prestopped realizations, there are also two types of preaspirated allophones, as well as plain nasal ones. The preaspirated realizations are particularly interesting, as they have not, to  my knowledge, been reported in Australia. However, there are interesting parallels with Burmese-style voiceless nasals (Bhaskararao & Ladefoged 1991, Chirkova et al. 2019), and I discuss some typological implications of this parallel.

Nasal prestopping is absent in the youngest generation of speakers, which may suggest that this is due to language attrition. However, there is no evidence of loss of any other marked consonant types in these speakers, which casts doubt on an attrition-based explanation. As an alternative, I argue that there is a system-internal pathway towards loss, driven by a number of contexts where prestopped or preaspirated realizations regularly alternate with plain nasals.

Bhaskararao, Peri & Peter Ladefoged. 1991. Two types of voiceless nasals. Journal of the International Phonetic Association 21: 80-88.

Chirkova, Katia, Patricia Basset & Angelique Amelot. 2019. Voiceless nasal sounds in three Tibeto-Burman languages. Journal of the International Phonetic Association 49: 1-32.

Koch, Harold. 1997. Pama-Nyungan reflexes in the Arandic languages. In Darrell Tryon & Michael Walsh, eds. Boundary rider: Essays in honour of Geoffrey O’Grady. Canberra: Pacific Linguistics. 271-302.

Simpson, Jane & Luise Hercus. 2004. Thura-Yura as a subgroup. In Claire Bowern & Harold Koch, eds. Australian languages: Classification and the comparative method. Amsterdam: Benjamins. 179-206.