British Library: transcribing the Dunhuang manuscripts
In October 2023 I took a break from my PhD research to do a placement at the British Library. I’m taking part in a project to work on Handwritten Text Recognition (HTR) of classical Chinese manuscripts from the Dunhuang collection. The British Library holds some fabulous resources for delving into Chinese history and literature and the ancient Dunhuang manuscripts are arguably foremost among them. They include handwritten versions of classical Chinese texts dating back to the Tang dynasty (618–907) and beyond. These are naturally of great value as historical objects but they are also really useful for the study of literature.
Conversion of these manuscripts into digital images has been underway for many years and has done much to improve their accessibility. Indeed, a recent project digitised nearly 800 Lotus Sūtra manuscripts from the collection! It is a considerable achievement but these remain simple images and can only be read by the human eye, requiring at least some familiarity with a range of early calligraphic styles. Use of HTR to transcribe these images into digital text could open up a world of potential, allowing us to search and analyse these early writings using the tools of the digital humanities. It is an exciting prospect!
At the British Library I work across two departments, supervised by Adi Keinan-Schoonbaert in Digital Research and Melodie Doumy at the International Dunhuang Programme (IDP). Working with my new colleagues at the library has been inspiring. I’ve had a chance to view some of the Dunhuang manuscripts in person – a rather magical experience! I’ve learned a lot about HTR and the technical processes behind it. I’m also exploring tools for digital analysis of the transcribed texts, ways of identifying some of the unique features of these handwritten works. In the coming weeks I will offer more posts relating to this project. Some of these will appear here and there should be one or two posts on a British Library blog in the (un)foreseeable future.
With thanks to OOC DTP, AHRC, and Bailley Gifford for funding and support throughout this placement.
