Professor Simon Kirby explains how his research into the evolution of language has led to artistic collaboration and ultimately the release of a record featuring words about light…

What is the difference between doing science and making art? For the past decade or so, I’ve tried to find out the answer to this question by spending time on both. Just as I’ve collaborated with brilliant colleagues and students to run experiments investigating how language evolved, I’ve also been lucky enough to work with musicians, designers, and artists to create interactive installations, mysterious mechanical contraptions, and the occasional vinyl record release. Each time, I try and introduce my artistic collaborators to the ideas that run through our academic work, and they in turn give me new perspectives on how to understand my scientific investigations.

For my latest such collaboration, Sing the Gloaming, I worked with two artists and musicians, Tommy Perman and Rob St John, to create a series of sound installations and a record inspired by a diagram from a paper that my collaborator Chrissy Cuskley and I wrote. The paper was about the ways in which words in a language might not be as arbitrary as they first appear. In particular, the diagram looked at a lot of words in English that relate to light: gloom, glow, glass, glaze, glare, glitter, glisten, gleam, glimpse, glimmer, glance, glint. Why do all these words start with “gl-”?

In this case, the answer lies – at least in part – in their shared history. They all stem from a root form in a 5000 year old language, Proto-Indo-European, that is the ancestor of around 450 languages spoken around the world today. Languages are passed on over time through a repeated process of learning and use by many generations. Each time a word is learned and used there is the possibility for change in that word’s sound, and in its meaning, leading to a proliferation of related words that are inherited by later generations scattered across time and space. This process of cultural evolution is what drives language change, but also, as research by my group at the Centre for Language Evolution has shown, explains why languages have the particular structure that they do. Patterns of regularity in the lexicon are one aspect of that structure in language – a structure that is both the result of accidents of history but also a reflection of the many brains through which language has passed on its way to us.

We decided to explore these ideas artistically by recreating Chrissy’s diagram in sonic form. We asked some of Scotland’s greatest vocalists to sing the “gl-” words and then used these as the starting point for new compositions based around how these words evolved over the centuries. We asked the singers themselves to come up with their own interpretation of the words we gave them to sing. But we wanted the process of composition to reflect the way words themselves evolve. My collaborator, Tommy, after hearing about some of the lab experiments  that we do in the Centre, hit upon the idea of letting each singer hear the recordings of the previous singer. This creates a chain of transmission from one singer to the next, just like the ones down which language itself is transmitted. And just like languages, melodies emerged and evolved in this process.

We started first with a 24-hour sound installation in a Scottish forest, in which we placed a series of singing boxes among the trees, allowing the audience to wander at will and hear for themselves the evolution of the word forms as they walked from site to site. Later, we created a miniature version of our installation in a shop window for the Dundee Design Festival, literally making the glass of the windows sing to passers-by.

For our record release, we wanted to take our compositional approach further, and bring in other technologies to arrange the music and transform the sound of the voices. We used data from the Google Books corpus to extract the detailed histories of each of the words. Then we used these histories to create simulated reverberant spaces inside the computer within which the singers’ voices were placed. When you listen to the record, you’re hearing the words sung through hundreds of years of language evolution. We also used the original diagram from our paper to create a compositional score by writing a computer simulation of different possible routes through the histories of the words. In addition, we broke the words down into their constituent parts, based on their phonological structure and spectral characteristics, to create new instruments to be played on modular synthesisers.

The end result is a record whose creation was informed at every step by the research that was its inspiration. However, this process was more than merely an artistic interpretation of scientific results. I have learned that the practice of art-making is fundamentally one of asking questions, of investigating and probing ideas, of throwing new light on the familiar to reveal the unexpected. All of these are important aspects of the scientific process too. So, although I started out by asking what the difference is between doing science and making art, I’ve now realised that I am further than ever from an answer to that question.

And not knowing the difference is exactly where I want to be.

Simon Kirby is Professor of Language Evolution at the University of Edinburgh.

Sing the Gloaming is released on vinyl and digitally 31 July 2020, and is available to pre order from Blackford Hill Records.