ARC Centre of Excellence for the Dynamics of Language

We have great pleasure in announcing that the ARC has funded a Centre of Excellence for the Dynamics of Language over seven years. This project will be led by Nick Evans at ANU with a collaborative team from there, the University of Western Sydney, the University of Queensland and the University of Melbourne, and with many partners from other universities and institutions including AIATSIS and  Appen.

We want this to be a centre for collaboration, for generating  ideas and inspiration for linguistics in Australia and the world.  In the New Year we’ll be putting up a web-page to give more information, In the meantime, here’s an overview of what we are planning.

The Centre will be made up of four programs as follows:

Program 1. The Shape of Language explores the design space of the world’s languages. We will investigate a strategic selection of little-known languages of our region, build the first large corpora of Australian Indigenous languages, Papuan languages plus Bislama, and initiate new research on how intergenerational variation can reveal different design solutions evolving in languages to solve similar social communicative problems.

Program 2. Learning examines how children acquire language as a first, or subsequent, tongue, in mono- and multilingual contexts. We will consider first and second language acquisition in complex, understudied contexts in our region, and how acquisition in such an environment interacts with language diversity and change, and with cross-cultural differences in language socialisation.

Program 3. Processing examines psychological, neurological, perceptual and articulatory processes that are used when speaking and listening, and whether the differences between languages affect how we think. To answer these questions, we will examine languages in laboratory and field settings using adapted portable  equipment, from within the frameworks of psycholinguistics, speech pathology, typological linguistics and phonetics.

Program 4. Evolution will use data from the three preceding programs to engage with central questions about the evolution of language. Here we explore what possible structures languages can develop, how learning and processing biases shape the direction of evolution, what is the role of the community in evolution, and how can we use this knowledge to program robots to learn any language.

These four question-focussed programs are interwoven by two threads organised around the technological innovations we need to drive our new research approaches .

Thread 1. New Generation Documentation and Archiving will oversee the unparalleled corpus development in which the Centre will be engaged. We will pioneer new documentation methods (e.g., using time aligned gesture and speech recording and showing how polysynthetic languages are learned), will develop appropriate digital archives to secure an immense cultural heritage, and ensure that documenting and archiving maximises the future potential of the data. This will allow us to chart language change over time in a way not previously possible. The thread supports the programs as follows:

• Develop new-generation language documentation techniques, field-test and disseminate
• Pool and upscale resources for recording, storing, annotating and accessing audiovisual linguistic data, and train researchers as users
• Time-align gesture & speech, test perception & production
• Document multi-modal production, perception and comprehension in under-described regional languages
• Generate diachronic corpora to chart language change

Thread 2. Research Technologies will address issues related to the management of big data, such as phylogenetic methods for examining deep-time historical relationships and for identifying broad-scale patterns in language evolution. It will bring about a massive expansion of the material we can record for little-known languages by adapting lab technologies for field use, such as miniaturised eye-trackers for monitoring the flow of attention during speech, and by developing crowd-sourcing technologies, (e.g., mobile apps to support the recording of languages by members of indigenous speech communities). The thread supports the programs as follows:

• Develop, apply and teach computational methods for solving problems
• Produce statistics from cross-linguistic and variation data, share predictions across Programs
• Model human processing of languages, including infant, aged, and disordered populations
• Predict acquisition sequences, developmental norms, phylogenetic pathways
• Adapt and teach new laboratory- and computational-based technologies for field research
  • Portable speech perception tests; large-scale home recording; eye-tracking; ultrasound.
  • Mobile apps for crowd-sourcing variation, acquisition, and processing data.

Follow