Best Practices for Born-Digital Editions Using Enhanced Imaging (NEH DHA Grant)

Communicating Revealed Texts: Best Practices for Born-Digital Editions Using Enhanced Imaging Across the globe, many important books, both handwritten and print, survive in damaged form, their contents not immediately available to the naked eye. Some of these damaged texts have been widely known for years; others are being identified and discovered in libraries and archives or uncovered through archaeological excavations. Over the past two decades, much progress has been made in non-invasive enhanced imaging techniques to reveal these damaged texts, especially in multispectral (MSI) and XRF imaging, to bring out erased or faded writing, and more recently in micro-CT, to reveal writing in texts too damaged to be opened. The new readings made possible by enhanced images has led to print editions with an accuracy and scope which would not otherwise have been possible. At the same time, the use of enhanced images is best documented and assessed through digital editions, and in this area much work remains to be done. This grant will establish a working group consisting of scholars who are producing digital editions of manuscripts from diverse times and places which are based on enhanced imaging, such as MSI or 3D tomography, and of librarians, curators, conservators, and metadata specialists who are involved in describing and modelling the physical structure of the manuscript and documenting the connections between the material object, images of that object, and the transcription(s) of text based on those images. We have identified a group of participants with ongoing projects in this area, working on manuscripts such as the Herculaneum Papyri, the Sinai Palimpsests, the Dead Sea Scrolls, the Medinet Madi Coptic Manichaean codices, and the Sanskrit Buddhist Sutras. Despite this impressive emerging pipeline of projects, there are almost no examples of born-digital editions which make use of enhanced images to transcribe otherwise unreadable text. On the one hand, enhanced images, especially multispectral, have been used almost exclusively for the preparation of print editions, which do not allow full consultation of the relevant images. On the other hand, most digital publishing platforms for ancient texts, such as papyri.info or the Chinese text project (https://ctext.org/), almost exclusively offer digital versions of texts originally published in print. This working group will meet monthly over Zoom to share and discuss their work, especially as it relates to the convergence of enhanced imaging technologies, digital transcription and editions, and the connection between the materiality of the text, the images of it, and the transcription. At the midway point of the grant, we will assemble for a two-day conference at the University of Iowa Center for the Book to discuss best practices as a full group, and then draft of set of protocols and standards for born-digital critical editions based on enhanced images, to be completed over further zoom meetings during the second half of the grant period. The major outcome of this grant will be a white paper on best practices for creating born digital editions with enhanced images, as well as several “micro-editions” that will serve as examples of these practices. The white paper will specifically address the proper documentation of the inscribed object and its digital representations; how to document for non-specialists various types of image processing and the resulting image files; connecting the digital representations of 3D objects to 2D images of individual pages or sections; improving the accuracy and documentation of scholarly transcriptions of the revealed texts, including mark-up/representation of various kinds of uncertainty, and the linking of uncertain readings to particular images; the integration of multiple proposed readings by various scholars; the integration of other specialized applications, such as digital paleography/handwritten text recognition (HTR) and proposals for missing text based on machine learning; and finally, connecting digital editions to other resources through linked data.