LEXINGTON, Kentucky (June 17, 2022) † Stephen Davis, an associate professor in the History Department in the College of Arts and Sciences at the University of Kentuckyhas been awarded a 2022 ACLS Digital Justice Seed Grant†
The American Council of Learned Societies (ACLS) Grant Program for Digital Justice is designed to promote and deliver the digital humanities projects that aim to diversify the digital domain, promote justice and equity in digital science practice, and contribute to public understanding of racial and social justice issues, especially those affecting the elevating people’s interests and histories of color and other historically marginalized communities. The program is made possible by a grant from the Mellon Foundation.
“The inaugural ACLS Digital Justice Grantees highlight a wonderful convergence between publicly engaged humanities and digital humanities,” said Keyanah Nurse, ACLS Program Officer of Higher Education Initiatives. “These teams are an example of the shift from extractive practices around data collection and towards collaborative knowledge production with people outside the academy. ACLS is proud to support the robust commitment to data ethics at the heart of these projects.”
Davis specializes in anti-apartheid politics with a specific focus on armed struggle and the practice of political violence. He was awarded the seed grant for “The Personal Writes the Political: Rendering Black Lives Legible Through the Application of Machine Learning to Anti-Apartheid Solidarity Letters.”
The project uses machine learning (ML) models to extract data from an archive of anti-apartheid solidarity letters written primarily by black South African women. The aim is to use newly developed optical character recognition (OCR) and handwritten text recognition (HTR) methods to convert images of handwritten letters into machine-readable text. Once processed, they will train custom ML models to produce triplets, meaning that two or more nouns are related through a verb indicating a qualitative relationship between two categories of data. A knowledge base derived from entity triplets will enable a greater understanding of the lives, struggles and contributions of black women in South Africa.
“We are excited to begin exploratory work on this exciting project. This archive has long held the promise of rethinking our understanding of the role of black women in the fight against apartheid,” Davis said. “With recent advances in machine learning, we can finally unlock that potential to rethink our understanding of popular participation in this important movement.”
“The personal writes the political” is one of eight startup projects awarded ACLS Digital Justice Seed Grants, each of which will receive up to $25,000. All beneficiaries will have the opportunity to work with the Financing Fund for Non-Profit Organizations plan for the long-term management and sustainability of their projects.