Decoding Hidden Heritages in Gaelic Traditional Narrative with Text-Mining and Phylogenetics

This project will fuse deep, qualitative analysis with cutting-edge computational methods to decode, interpret and curate the hidden heritages of Gaelic traditional narrative. We will provide the most detailed account to date of similarities and differences in the countries' narrative traditions and, by extension, a novel understanding of their joint cultural history. Using recent advances in language technology, we will digitise, convert and make available a vast collection of folklore manuscripts in Irish and Scottish Gaelic. In turn, this new digital resource will catalyse ongoing research into Gaelic speech technology.

The main question that this project poses is: What can a large, digital collection of traditional narrative tell us about historical cultural exchanges between Scotland and Ireland? While we will examine different kinds of traditional narrative, we will concentrate upon International Tales in Irish and Scottish Gaelic. International Tales are folk narratives that are widely scattered across the world, such as Cinderella or Snow White. Some are amongst the oldest known facets of human oral culture, dating back 5000 years or more. Due to their age and their tendency to vary geographically, they can tell us a lot about relationships between different settlements of people. In this study, we use them as a vehicle for understanding early communication networks between Scotland and Ireland, as well as those that existed within each country.

Through an approach known as text-mining, we will use artificial intelligence to search the tales for similar topics, phrases and other linguistic patterns. The matches that we find between texts will be correlated with what we know about the texts themselves. Using another approach, known as phylogenetic network analysis, we will comb relationships between the texts' themes and the people who produced them: for example, their genders, where they lived and what they did for work. In the end, we will combine the two approaches towards a unified account of Scottish and Irish oral narrative. This work will transform our understanding about Gaelic oral culture and disseminate unique archival material online to a diverse set of end-users. Finally, it will positively impact the sustainability of Gaelic-speaking communities through the creation and further stimulation of important language technologies (e.g. handwriting recognition, machine translation and automatic speech recognition).