Legacies of Catalogue Descriptions and Curatorial Voice: Opportunities for Digital Scholarship

"Legacies of Catalogue Descriptions and Curatorial Voice: Opportunities for Digital Scholarship" will develop a platform for a transformational impact in digital scholarship within cultural institutions by opening up new and important directions for computational, critical, and curatorial analysis of collection catalogues. Extensive digital and digitised sets of curatorial descriptions from legacy catalogues are increasingly available and we seek to realise their potential as valuable resources for cross-disciplinary research into curatorial practice, and for enhancing access to and analysis of collections at scale.

Catalogues are fundamental to cultural institutions: they represent their objects, provide the basis of searches for their objects, and communicate knowledge about their objects into the future. Catalogues are also fundamental to the history of cultural institutions, as artefacts of late-nineteenth and early-twentieth century professionalisation that have evolved from physical objects such as printed books to digital databases, online discovery services, and linked open data. Catalogues, then, create a lasting legacy. The digitisation of collections has furthered that legacy with the use of descriptions from legacy print catalogues as a starting point for indexing digitised material.

Thus we see that the writers of catalogue descriptions are powerful interlocutors not only between objects and viewers, but also between the past and now. And when legacy catalogues are reused as the basis for contemporary descriptions of collection items, a powerful and often difficult to detect "curatorial voice" remains. This voice is a product of the historical and social contexts in which the descriptions were written. This has serious consequences for the trust which users can have in federated catalogues, particularly when parts of those catalogues are steeped in unacknowledged or unidentifiable past voices and colonial gathering. There is therefore an urgent need to elucidate and foreground curatorial "voices", and to do this at scale. Curators and researchers alike require methods that can comprehensively articulate the choices, preferences, and omissions made by curators in what are often large bodies of text produced during decades of work.

This project will foreground some of the fundamental ways in which cultural institutions represent their objects and create a pathway to transform the reuse of legacy catalogues for access, scholarship, and research. Our pilot research will investigate the temporal and spatial legacy of a landmark catalogue: the 1.1 million word British Museum 'Catalogue of Political and Personal Satires', which is the basis of related catalogue data at the Lewis Walpole Library and the British Library. We will demonstrate how methods combining corpus linguistic and archival research can be used to produce an empirical account of curatorial "voice" across a large catalogue. We ask questions about the enduring legacies of curatorial labour, methods for defining and highlighting curatorial voice, the role of digital scholarship in responding to the ways in which legacy descriptions work against contemporary ambitions of cultural institutions, and how to develop sectoral capability in digital scholarship. We will co-produce training materials and reports, deliver proofs-of-concept for changing how legacy descriptions are presented to diverse publics, and release transparent code, data, and methods to enable the reuse of our methods.

The project team comprises researchers, curators and technologists from University of Sussex, Yale Digital Humanities Lab, the Lewis Walpole Library and the British Library. The partnerships between the team members, and the wider community, will be developed through pilot research, workshops, and research residencies.

1. *Collaborating Cultural Institutions* will benefit from computational, critical, and curatorial analysis of their collection catalogues. In particular, they will benefit from our focus on elucidating difficult to detect curatorial voices whose continued presence works against the ability of cultural institutions to respond to larger societal shifts. This will impact on the collaborating cultural institutions by, for example, transfers of knowledge about computational approaches to defining features of "curatorial voice" and by giving them agency to co-produce capability building outputs (e.g. a training module) based on their own catalogue data. During and after the project, the British Library and Lewis Walpole Library will benefit from being at the centre of a new transatlantic network and collaboration focused on the temporal and spatial legacies of early-twentieth century Anglophone curatorial labour, and on producing practical, implementable, and public facing responses in light of new knowledge and data about these legacies.
2. *Cultural Institutions with Significant Legacy Catalogue Data* will be benefit from the project as attendees at our capability building and partnership development workshops. UK and US based invitees will be drawn from participants at workshops run during Baker and Salway's recent British Academy "Curatorial Voice" project, and from the networks of British Library and Lewis Walpole Library, for example the British Museum, Cleveland Museum of Art, and Wellcome Collection. For those responsible for delivering access to collections, the workshop will be an opportunity to reflect on curatorial practice and inherited knowledges, and how they impact search and discovery of collections. There is also the potential for a big practical "pay-off" if we demonstrate that descriptions can be classified automatically, e.g. for checking curatorial practice against institutional guidelines, and for selecting (parts of) descriptions to be used as the basis for structured representations of digital images of collection items.
3. *Policy Networks relating to Cultural Institutions* will benefit from interventions that underscore the political urgency of knowing where, from whom, and under what circumstances catalogue data was produced. For example, during the development of our training module we will work with curators and cataloguers to assess the applicability of corpus linguistic techniques to current and future professional practice. Given a set of guidelines for producing curatorial descriptions, corpus techniques can check the extent to which guidelines are being followed at a macro-level, e.g. by identifying what aspects of objects tend to be referred to or not. Further, such analysis can form a basis for plans to enhance a catalogue by providing areas to focus on and estimates on the person time required. A corpus-based characterisation of the language used in an exemplary catalogue could also be used to develop or refine cataloguing guidelines by identifying that catalogue's distinctive linguistic features.
4. *Publics Using Legacy Catalogue Data* will benefit from the production of proofs-of-concept for presenting legacy descriptions and their linguistic features. It is common, for example, for news agencies to clearly "flag" old stories on their websites, but that no such custom exists for legacy descriptions. We will develop implementable pathways for achieving such interventions, disseminated as an easy to digest pamplet. Further, when using a legacy catalogue as the basis for accessing a collection through text-based search, users may benefit from having an overview of the common vocabulary in order to understand what search terms are likely to be effective. Our project considers in what form users of legacy descriptions would find this type of data useful, and creates a pathway towards change.