When the apple falls far from the tree: Southern New Guinea phonologies now and back through time

Nicholas Evans
Linguistics, School of Culture, History and Language, College of Asia and the Pacific, Australian National University

Monday January 26, 2026, 5:15 - 6:30. Schwarzman Centre 30.445, Linguistics Large Seminar Room.
 

Until rising seas around 10,000 bp, the two present-day hemi-continents of Australia and New Guinea were joined in the erstwhile continent of Sahul. For the first 55,000 years of human habitation in Sahul, there was thus no geographical continuity and in fact the now-submerged zone of Torres Strait and the northern Gulf of Carpentaria would have been an ecologically alluring landscape of lakes and deltas. It is thus an enduring enigma that the languages, and cultures, of Australia and New Guinea appear so different, apparently bearing no trace of 55,000 years of continental cohabitation.
 

Today, The Papuan languages of Southern New Guinea (SNG)_, especially those of the Yam and Pahoturi families, are spoken just a few miles away from the northernmost Australian languages, and further back in time one could walk from Southern New Guinea to Australia. Yet if we compare the phonologies of modern languages, such as Nen, with their Australian neighbours, they appear worlds apart: Nen lacks such typical Australian features as velar nasals, multiple coronals series forming a 'long flat' phoneme inventory, and a rich liquid inventory, while on the other hand it possesses such 'un-Australian' features as fricatives and coarticulated labial-velar consonants.
 

But have things always been like that? In this talk I will show how systematic data-gathering on a variety of Yam languages, coupled with the application of the comparative method to reconstruct earlier phases of these languages, produces a very different picture of what their ancestral phonology looked like, with it drawing much closer to the phonological patterns found in Australian languages. Velar nasals, retroflex stops, and an interdental series all appear, through the power of the comparative method to recover lost structures by pooling information from multiple descendant languages. I conclude by posing a key question for diachronic phonological typology: what factors determine why descendant languages stick close to their inherited phonologies, while others diverge wildly from them?