Skip to content
Home > Blog > Articles > “Creation of Early Music Corpora (CORSICA): Activities and a Workshop About Building Collections of Digital Editions”

“Creation of Early Music Corpora (CORSICA): Activities and a Workshop About Building Collections of Digital Editions”

Frans Wiering, Erik Bergwall, Werner Goebl, David Lewis, Anna Plaksin, Esperanza Rodriguez Garcia, David Smith, David Weigl

There are many challenges in creating and using digital editions, but there are also many opportunities, including a wide pool of individuals and communities actively engaged with music transcription, editing and sharing. Through an EarlyMuse Short Term Scientific Mission, Creation of Early Music Corpora (CORSICA), a group of researchers with diverse historical and technical areas of expertise met on-line, interviewed corpus creators and, from 13-17 May, 2024, gathered together in Utrecht to expand our understanding of the state of the art, and develop a vision for the future of digital music editing.

The Corsica team topics of discussion with the help of Post-It notes

We are clear that digital transcription and editing – just like using the results – is a fundamentally human activity. Over the course of the week, we considered the important roles that contributors and users might make, whether they might be professional performers and musicologists, enthusiasts or casually interested non-experts. We need to make sure that we understand the needs and motivations of contributors and users of our tools and corpora. This was a topic that we explored through our interviews, persona modelling and user stories, and is something that we will need to revisit often.

To ground our discussions with real examples, we were honoured to be invited to the library at Utrecht University to be introduced to a selection of sixteenth- and seventeenth- century manuscripts, most notably, the spectacular Spanish book of instrumental music, the Lerma Codex. We are especially grateful to Bart Jaski for granting us access and for his generosity with his time and expertise.

These wonderful physical objects formed focussing examples for later exercises exploring the encoding of musical content, the availability of – and interconnections between –  catalogue and historical information about the contents, as well as the processes and expertise required to realise successful corpora.

The team is introduced to the Lerma codex by Bart Jaski

We were also guided in our explorations by presentations and discussions from remote visitors – Julie Cumming, Mark Gotham and Cory McKay – and also in-person contributions from local scholars, Anja Volk, Marnix van Berchum, Peter van Kranenburg and Mirjam Visscher.

The team standing on the staircase of the old Utrecht University Academic Building

Perhaps the most important takeaway from our discussions was that the variety of approaches to music encoding and corpus creation is there for a reason. We have tried to map this variety by describing the corpus creation along a number of axes. Corpora created for analytical purposes and for digital editing have to meet different requirements depending on the aims of their creators and users. Some corpora focus on sources, others on works. Forms of music notation differ both visually and in the information they contain. Missing, ambiguous and faulty information can be dealt with in multiple ways that depend on the intended purposes, ranging from musicological study to practical performance and adaptation. Lastly, encoding a piece of music is a creative act, usually done voluntarily, so contributors need a sense of autonomy and respect for their expertise to be motivated to do the work. The future of corpus creation therefore seems to lie not so much in unification as in coordination of diversity.

A potentially helpful insight coming out of this is that it makes sense to distinguish layers of encoding (such as a ‘source’ and ‘error correction’ layer). One encoder’s final product (e.g. an analytical encoding) could thus act as the input for the work of another (e.g. creating a performing edition).

Making the ‘Mission’ work

We wanted to use the five days we would be together effectively, and to gain as much as possible from the STSM. We took a range of approaches to support this. Two pre-workshop online meetings allowed us to share our aims and the literature that influenced our thinking, and also meant that we could collaboratively plan our time in Utrecht. Preparatory interviews with corpus creators also helped us clarify our goals in advance of our physical meeting.

Early in our planning, we identified a venue for sharing our findings – a panel on corpus creation at the Medieval and Renaissance conference in Granada at the beginning of July. Since we needed to submit an abstract well in advance, we were forced to identify core discussion points early. The panel also helped us to start our own workshop. We opened with an approximation of our conference session, showing where we were each starting from and seeding our later discussions.

We timetabled sessions for the week only approximately. After the panel session and our library visit, we collected challenges to explore on Post-It notes written by participants. We then collaboratively organised these spatially into topics, and then planned our time around these emergent topic areas. Dividing into groups as necessary, we explored these topics and then shared our progress in plenary sessions.

We finished our week with another Post-It session, each writing shorter- and longer-term next steps onto notes for combining and grouping. These will form the basis of two written reports: a vision document and an implementation plan.