From Lab to the World

Course materials for Michael Dunn's course at the 2018 LOT Winter School

Lecture outline and readings

Software requirements

Exercise 1: Bayesian Phylogenetic Inference with MrBayes
Datafile: Aslian28.nex

Exercise 2: Constraints and multiple partitions
Datafiles: new-britain-swadesh.nex, new-britain-structural.nex

Exercise 3: Ancestral State Reconstruction and the DISCRETE test
Datafiles: Aslian28+structure1.nex, Aslian28.nex.trees, Aslian-complexity.txt, Aslian-semantics.txt

Linking empirical typology to language evolution

This course provides a practical introduction to quantitative linguistic typology with a focus on variation, both synchronic and diachronic. From the synchronic perspective we will examine ways to elicit ecologically valid typological information from language speakers and extract it from natural texts. From the diachronic perspective we will use phylogenetic methods to examine the processes generating this typological diversity we observe in the world. Practical techniques will be illustrated using data from different levels of language from phonology to semantics. As a student you will be guided through some standard analyses (tools and data provided), and an opportunity will be offered to bring your own data/problems.

2018 Schultink Lecture

I have been honoured with an invitation to deliver the 2018 Schultink Lecture, associated with the LOT winter school. I will give a high level, conceptual overview of research on language typology and evolution. Here is the abstract:

Understanding language through language histories

Behind every attempt to explain the diversity of human language lies a question: What is it exactly that we are trying to explain? How can we characterise the nature of linguistic diversity? It seems inconceivable that the attested languages of the world exhaust the possibilities of linguistic diversity. To recycle SJ Gould’s thought experiment of “replaying life’s tape”, if we replayed the tape of the evolution of languages how much of the repetition would look like the original? What developments in language are inevitable, and what are contingent?

In the study of linguistic diversity we have access to a number of small experiments – the de novo development of Nicaraguan Sign Language, transmission chain experiments with artificial mini-languages – and one big experiment, the “natural experiment” that has resulted in the linguistic diversity observable in the world today. In my research I concentrate on the latter, trying to understand the diversity of language typology not only by observing the diversity which exists, but also by inferring the processes that brought this diversity about. In this effort I am part of a small but growing group of linguists, biologists, and cognitive scientists using computational phylogenetic methods to investigate language diversity as an evolved phenomenon. I will present some recent results touching on the different evolutionary trajectories of different linguistic subsystems, and showing the cumulative effects of social and cognitive biases on language structure over thousands of years.