Spurred by the current anti-racist and Black Lives Matter movements, several universities have renewed or ignited their passion for addressing the question: How do you accurately and empathetically describe the lives of the enslaved individuals bound to a university or institution of higher learning?

However, doing this work independently brings about a fundamental shortcoming: efficiency. This was one of the driving reasons Sharon Leon, associate professor of digital history at Michigan State University, East Lansing, had for building an ontology and database specifically tailored to house, catalogue, and disseminate information regarding the lives of enslaved individuals and how they intersect with the history of universities.

Leon saw all the work she and other historians were doing and grew concerned that their varied approaches to organization would stall the development of a macro view of this history. She reached out to friends, and together they formed the national team for On These Grounds: Slavery and the University (OTG). OTG is an experimental data aggregation model that attempts to create standardization in data cataloguing of historical documents detailing enslavement, enslaved individuals, and their intersection with universities and university land.

Two things this initiative does differently than previous university initiatives: center the story around enslaved people and not slave owners, and focus on events. These may seem like small changes, but they make an impactful difference. Traditionally, archival approaches to slavery attempt to outline a person’s life through records of life (birth, marriage, death, etc.). An event-focused approach attempts to outline how events impacted the lives of the enslaved to show how they lived, rather than just when and where they lived.

BUILDING A FOUNDATION

After Leon reached out to fellow historians from the University of Virginia (UVA), Charlottesville, and Georgetown University, Washington, DC, about the prospect of building an ontology model and open-source database, Georgetown sought out funding. The Andrew W. Mellon Foundation awarded OTG $550,000 to be spent on staffing as well as $50,000 for open-source web publishing platform Omeka to build an aggregation model. The grant funding period spans from July 2020 to December 2022.

The national team—consisting of digital historians, archivists, and historians of slavery from Michigan State, Georgetown, and the UVA—together with the Omeka team began the task of creating a foundational framework for the organization and eventual dissemination of information regarding their respective universities’ histories of enslavement.

OTG designed, tested, and edited two organizational models to discover the most precise and accessible framework. Through quantitative testing they uncovered several extraneous meta data headings, necessitating the use of the simpler model.

The work is planned in stages. The first stage will tackle creating a foundational framework to standardize the organization of data cataloging regarding historical documents of slavery and the university. This will be achieved through writing and revising an ontology that will guide the organization and create the metadata used in cataloging these specific historical documents. Once built, this framework can be disseminated and replicated in any institution willing to take on the work. The second stage will focus on centering the stories of the enslaved individuals and building a new way of interacting with this type of historical material through collecting and organizing university records. Within this stage, testing partners will reach out to community members and descendants of enslaved people to ensure the model does not only refocus the story of slavery in theory but in action. Through open dialogue and collaborative building, OTG aims to foster necessary adjustments to language and descriptors used for archives of enslavement.

ACQUIRING TESTING PARTNERS

In selecting testing partners, members of the national team wanted to mirror their team dynamics. Professor Leon noted that an integral criterion for accepting an application was a clear “collaboration of librarians and historians, as well as a grasp of the scope of the collection and a willingness to do a different kind of work” from the applying universities. Among the applicants, five stood out as having the necessary elements to successfully aid in the construction of a working ontology and applicable framework: University of North Carolina (UNC)–Chapel Hill; Hampden-Sydney College, VA; Rutgers, the State University of New Jersey, New Brunswick; University of Georgia, Athens; and Washington and Lee University, Lexington, VA.

LJ spoke with UNC–Chapel Hill’s project team leader Chaitra Powell about the experience of working with OTG. Powell noted that she hopes to “change the way libraries keep records of slavery, to remember that these pieces of paper are not just records but people.”

UNC–Chapel Hill’s approach to this work follows the lens of radical empathy and the ethics of care. This means looking at “what it means for the people working with these pieces—how it affects them and how they view the world and work.” Through forming reading groups to review things like Cleveland police archives, and Michelle Caswell’s articles on radical empathy and symbolic erasure, the team employs the ethic of care to engage the staff working with these documents.

This method, paired with the work UNC–Chapel Hill has already done with this collection, allowed the team members to approach their work with OTG in what Powell views as a slightly different perspective than other testing partners. “Some of our peers are just trying to tell the story for the first time,”

As a testing partner, UNC–Chapel Hill’s team members began their work by trying to understand the testing model given to them and finding a workflow that was feasible for their team. The model had a working ontology and organizational framework that the testing partners engage with as their guide for cataloging records. However, they also question and suggest adjustments based on needs that arise and limitations that are uncovered.

Although each team works individually with their own institutions’ records, all testing partners check in monthly with the national team and as often as needed via a Slack channel. The environment is collaborative and allows space for questions to stretch the working model. Powell noted one assumption in the data model that was addressed in the cohort resided within the metadata categorization. In looking at the categories provided, the question emerged “at what point is a crime a crime or freedom seeking or an act of resistance?” Asking these critical questions ensures that librarians are not complicit in perpetuating the same inequities.

RECOGNIZING LIMITATIONS

Among the limitations uncovered is the lack of representation in the national team. Although all team members have an applicable background, they lack cultural diversity. To address this lack, the national team commissioned the assistance of an advisory board whose members (Daina Ramey Berry, professor of history, University of Texas at Austin; Dorothy Berry, digital collections program manager, Houghton Library, Harvard University, Cambridge, MA; Hilary Green, associate professor of history, University of Alabama, Tuscaloosa; Leslie Harris, professor of history, Northwestern University, Evanston, IL; Sony Prosper, PhD candidate, University of Michigan School of Information, Ann Arbor; and Holly Smith, archivist, Spelman College, Atlanta) will serve as a guide, ensuring the voice of the enslaved remains central and respected. The lack of diversity also extends geographically. The universities participating in this round of testing do not span across all former slave-owning states, nor do they encompass all states with universities that have a historical tie to slavery. Professor Leon commented that this is something they hope to address if granted the opportunity to expand this work in future.

For testing partners, understanding the complexity of the model presented a challenge. Working through the model slowed the UNC–Chapel Hill team’s pace when attempting to add records and provide feedback for edits to the model. UNC–Chapel Hill has also faced challenges in lining up stories from descendants of the enslaved with records from the university. The goal is to ensure their stories are heard and incorporated, but working with imperfect records, obscured information, and time constraints makes actualizing this goal arduous.

NEXT STEPS

Once the initial research phase is completed, OTG plans to reapply for grant funds to expand and deepen the testing, and eventually implement the initial model with as many institutions as possible. The vision is to produce a living open data source that is accessible to universities and individuals alike, which provides a space for this work to be connected and these stories to be shared.

For testing partners like UNC–Chapel Hill, being a part of this initiative has already impacted the way they interact with their records, and they see this system being influential for them in the future. They plan on continuing to contribute to the open data source even after their testing period ends. Chaitra also hopes that “this project will help with [offering] tools to use for conscious editing initiatives that revisit [existing] descriptions.”

ADVICE FOR FUTURE PARTNERS

The work is intense and can be labor-intensive for institutions that have not already begun aggregating archives of the history of slavery. Professor Leon’s advice for anyone who may wish to implement this model in the future is to start working now to build a working knowledge of your collection.

Once you are ready to dive in, dividing the work is immensely beneficial. At UNC–Chapel Hill, the team divided the work by task. Librarians, archivists, and historians review the records and make notes, a student aid transcribes the notes, then a metadata expert uploads the information to keep the data records clean. The records are then sent off to Michigan State to aggregate and analyze for data patterns between all the participating universities.

The emotional impact of this work cannot be overlooked. To this end, Powell suggests “moving with intention”: “Think about what it means to uncover these records and who would be impacted by the discovery of these materials.”