Back to listing

Director weekly highlights 11 June

Nicholas Evans, Outreach

Date: 11 June 2021

This has been one of those weeks when a whole lot of ships come in, in terms of publications. There are too many to do justice to all of them — six papers across a range of journals, a further two now out in the rolling special edition on Australian Languages Today of the journal Languages, a massive grammar of Marind, and a textbook on corpus linguistics. You can find all the details below, and on behalf of CoEDL I congratulate all authors. I’d like to comment a bit more on a couple of them here.  

First, Bruno Olsson’s epic grammar of Marind, just published in the Mouton Grammar Library. Bruno wrote his PhD on this fiendishly complex language of the Merauke district of Papua province, Indonesia. Bruno submitted an earlier version of this grammar for his PhD at CoEDL’s partner institution, Nanyang Technological University. During his postdoc, though primarily working on other topics like the related language Yaqay, he has revised it by around 20%. Drabbe (1962: 9), the depth and breadth of whose knowledge of Papuan languages was unparallelled, considered Marind “in every respect more complex than all the languages of New Guinea which have until now been studied”1 , and sixty years later his judgment still stands. But it’s a mistake to see complexity as the most interesting issue — rather, it’s what that complexity does, and among the many fascinating features of this language are its grammaticalised systems of engagement and focus. This grammar joins Hannah Sarvasy’s earlier Nungon grammar and Christian Döhler’s Komnzo grammar (as well as several PhDs, still unpublished) in setting the gold standard for showing us how Papuan languages work and providing the sorts of reference grammars that go on being used for many decades. 

Another book, which I’ll only mention briefly (in sneak-peek mode, because it hasn’t fully come out yet), is Danielle Barth and Stefan Schnell’s new textbook Understanding Corpus Linguistics, in Routledge’s innovative and forward-looking Understanding Language series. I mention it now because you can already pre-order it (see below), something that will help boost promotion and print runs. Once it is properly out, I will say more about this foundational text. 

One of the big tasks CoEDL set itself is to ‘Transform the science of language’, and it is through textbooks like this that a whole lot of growing tips get joined onto one branch of knowledge. We can be very proud that of the 13 titles in the Understanding Language series so far, two are by CoEDL Chief Investigators and/or postdocs (this one, plus Felicity Meakins, Jenny Green and Myf Turpin’s Linguistic Fieldwork, which appeared in 2018), one by Partner Investigator Caroline Rowland (Understanding Child Language Acquisition), and one by former Advisory Committee chair Kate Burridge with Alexander Bergs (Understanding Language Change). 

It was also a great pleasure for me to see a couple of papers come out that I’ve been involved in through some rather long pipelines.  

One, with Louise Baird and Simon Greenhill, seeks to give a first quantified answer to the ‘Bird-Himmelmann problem’ — how much documentation do we need for a given language? We deliberately set the bar extremely low, to make it more easily measured — we simply measure what length of text is needed in order to have a close to 100% chance of attesting at least one allophone of every phoneme in a language. (In the battle for disciplinary survival, linguistics won’t count if linguists won’t count). To answer that we used a resource that has been built up over many decades, the famous ‘Illustrations’ of the Journal of the International Phonetic Association, which give comparable texts for 160 languages from around the world that we could use to measure the growth of ‘phoneme capture’ though text-time. Trying to make this work involved some very careful data-scrubbing by Louise, complex curve-fitting by Simon, and an unexpected detour into a clutch of Czech bibles (read the paper to find out why!). In the end we are able to at least give an answer to our question — about three and a half hours of text is needed to meet this very low bar of at least one allophone of every phoneme, if we want to allow for the worst-case scenario set by the slowest-recovery language in the sample (Cicipu, from North-western Nigeria). 

The other, which merged members and databases of our CoEDL Parabank team (Wolfgang Barth, Simon Greenhill, Kyla Quinn and myself) with the Bristol-based Varikin project (Sam Passmore and Fiona Jordan), tackled another fundamental problem — how far are languages ‘systems where everything hangs together’? In this case, how far do choices in one part of the kinship system (parent and uncle terms) predict, and get predicted by, choices in another part (sibling and cousin terms)? Many of us who have taught how different kinship systems work are used to saying things like ‘because you call your father’s brother by the same term as your father, your father’s brothers children are treated as siblings, just like your father’s children’. But does this actually stack up? Our paper looked at this question, drawing on a large database of 1,107 kinship systems from around the world, and found that, against our comfortable belief, things don’t hang together nearly as much as we had thought. Sure, there is some co-selection between design choices made at different points in the system, but way less than our classic curriculum would have us believe. This paper came out of the workshop on the Evolution of Kinship Systems organised by Kim Sterelny, Steve Levinson and myself in early 2019, and the full collection of papers will be coming out in the next month or so — stay tuned. 

It’s hard to know where to stop here, with such a rich slate of publications this week. Alistair Harvey’s paper is especially interesting for the fact that the summary and introduction are in the Kala Kawaw Ya of the Western Torres Strait, continuing a proud Western Torres Strait tradition of using the Indigenous language in journal publications2. And Jane Simpson’s wide-ranging sleuthing into the history of early women linguistic researchers in Australia, finally out after a long publishing delay, is a fascinating if shocking read full of poignant detail.  

I’d like to finish by mentioning another paper that I just found out about this week, courtesy of Robin Hide, which gives a firm numerical base for arguments about the vital role of language diversity in transmitting ethnobiological knowledge. This is a study by Cámara-Leret and Bascompte (2021) who tallied up the total known ‘medicinal plant services’ — pairings of a medicinal plant species and a medicinal subcategory (e.g., Ficus insipida + digestive system) — across three regions with high biocultural diversity — North America, New Guinea, and Northwest Amazonia. . The survey took in 236 indigenous languages and identified 12,495 such ‘medicinal plant services’, 75% of which were linguistically unique — in other words, just one language in this sample records this connection. They conclude: ‘Our finding of high uniqueness in indigenous knowledge and strong coupling with threatened languages suggests that language loss will be even more critical to the extinction of medicinal knowledge than biodiversity loss’. Click here for the full study. 

Finally, on a bittersweet note, our amazing Finance Officer Susan Jiang will be finishing up next week, having been recruited to a position at a higher level (though still within the same organisational unit) at ANU. We have been privileged to have Susan work with us for nearly two years. Quietly and competently behind the scenes, she has done a huge amount during that time to keep our finances on a firm footing. On behalf of CoEDL I would like to thank Susan for her superb job and wish her all the best in her new job. 

Have a good week everyone — I’ll be taking some time off over the coming weeks, and Jane Simpson will kindly be taking over these weekly introductions for that period. 

Nick Evans


1 “in ieder geval ingewikkelder dan al de talen van Nieuw-Guinea waarvan tot nu toe studie gemaakt werd” (Drabbe 1955: 7) 

2 Another Western Torres Strait variety, Kala Lagaw Ya, was also represented in a linguistic publication in 1987 in the late Ephraim Bani’s article “Garka a ipika: Masculine and feminine grammatical gender in Kala Lagaw Ya” in the Australian Journal of Linguistics. 

  • Australian Government
  • The University of Queensland
  • Australian National University
  • The University of Melbourne
  • Western Sydney University