Presented in person and online, details below.

In many parts of the world, the matching of reconstructable words from ancestral 'proto-languages' to archaeological evidence of plant presence and use (ranging from palynological studies of species ranges to fossil evidence of plant use, such as charred endocarps) have played a key role in bringing together dates and locations for inferred ancient languages and the cultures they transmitted. Edible plant food suites from proto-Dravidian (Pejros & Shnirelman 1992, Pejros 1997), evidence of early maize cultivation in proto-Uto-Aztecan (Hill 2001), and evidence for localising the homeland of proto-Indo-European on the basis of tree species distributions (e.g. birch, beech, oak; Friedrich 1970) are three well-known examples. Hooking together archaeobotanical and linguistic evidence in this way gives vital insights into the agriculture, plant-based technologies and ecological and geographical settings of speech communities many thousands of years ago.

Yet in the Australian context there has been much pessimism about the possibility of applying these techniques. Nash (1997:190), who carried out an exemplary study of comparative flora terminology in Central Australia, concludes pessimistically that, beyond the regional level 'the prospects seem slim that wider comparison will produce clearly relatable flora vocabulary'. More recently Harvey & Mailhammer (2024), in what is the most up-to-date reconstruction of the proto-Australian ancestral language, give around 50 reconstructed vocabulary items: their list does not contain a single plant term.

However, I will argue in this talk that there are in fact a substantial number of plant terms in Australian languages that reconstruct all the way back to the most remote linguistic ancestor, proto-Australian, from which all or nearly all modern languages are descended. Interestingly, these terms include the majority of plant species excavated from ancient hearths in Madjedbebe (Florin et al, 2020), dated in the 53-65kya range. So far there appear to be at least ten reconstructable plant terms, including those for fruits and berries like Terminalia Carpentariae, Terminalia Arostrata/Grandiflora, and Vigna Lanceolata, tubers like Diascorea Alata and DiascoreaTransversa, plants used for mats and baskets, such as Livistona Humilis (two terms), and others used for shade and medicine including the milkwood (Alstonia Actinophylla) and for digging sticks (Erythrophleum Chlorostachys) and as a mosquito repellent (Guttapercha). A reconstructable term for two orchid species (Cymbidium Caniculatum and Dendrobium Dicuphum) may appear puzzling until one realises its widespread and long-established use as a paint fixative in rock paintings. Perhaps the richest reconstructable set concerns words for Pandanus Spiralis,  pandanus words, with no fewer than five reconstructable sets: the most likely explanation for this embarrassment of lexical riches is that these terms were specialised for e.g. the tree itself, the fruit, the individual corms etc, suggesting a central economic role for this species. 

I conclude with two points: firstly, that the ancient words unearthed so far are likely to represent just a fraction of what historical linguistics can tell us about ancient plant use, and secondly that we face a seeming dilemma between the orthodox 'maximum reconstructability horizon' for languages, regularly asserted to be about 10kya, and the most parsimonious interpretation of the data presented in this talk, that the terms we reconstruct here – and hence the language containing them – may have an antiquity corresponding to the archaeological dates for their earliest use in Australia, and hence attest to a much deeper time horizon for the ancestral language, proto-Australian, than linguists have been willing to accept.

Speakers:
Nicholas Evans  investigates linguistic diversity and what this tells us about the nature of language, culture, deep history, and human creativity. He has carried out extensive fieldwork on the languages of Northern Australia (Kayardild, Bininj Kunwok, Dalabon, Iwaidja) and Papua New Guinea (Nen, Idi), with interests ranging from grammar-writing, linguistic typology, translation and historical linguistics. He has published A member of the Australian Academy of the Humanities, the Australian Social Sciences Academy and the British Academy, he has been awarded the inaugural Anneliese Maier Forschungspreis,  a Ken Hale Award from the Linguistics Society of America, and the Neil and Saras Smith Medal from the British Academy.

His major publications include grammars of Kayardild and Bininj Gun-wok, dictionaries of Kayardild, Dalabon and Nen, the award-winning and widely translated crossover book Words of Wonder: What Endangered Languages Tell Us, and edited volumes on Archaeology and Linguistics (with Patrick McConvell), Non-Pama-Nyungan Languages, Grammar-writing (with Felix Ameka and Alan Dench),  Insubordination (with Honore Watanabe), Polysynthesis (with Michael Fortescue and Marianne Mithun) and Papuan Languages (forthcoming, with Sebastian Fedden). 

Alexandra Marley is a linguist with research interests in language variation, change and evolution in Australian Indigenous languages. She closely documented language change and variation in Bininj Kunwok, a Gunwinyguan language spoken in the Northern Territory, in her doctoral work, and during her post-doctoral fellowship at the Australian National Herbarium at CSIRO, expanded into non Pama-Nyungan languages and examined plant names and uses to discover linguistic connections across northern Australia. She is currently on parental leave from her role as senior linguist and AustLang manager at AIATSIS and will begin a new role lecturing and researching at UQ in 2027.

Zoom details: 

https://anu.zoom.us/j/85125361297?pwd=KdCsabOwtGmQnDnL0MXJuFjZAnacoN.1
Passcode: 755214