In today’s digital age, data is everywhere. From our smartphones to our voting behaviors, the sheer amount of data generated and collected today is unprecedented.

This data holds the power to reveal profound insights about our world, yet it also raises many questions: What biases influence particular data sets? How do we assign authentic and accurate meaning to data? What are its social and political implications? Where should it be stored, and who should own it? 

These and many other questions are the focus of a year-long seminar series titled “The Data that Divides Us: Recalibrating Data Methods for New Knowledge Frameworks Across the Humanities.” Supported by the Andrew W. Mellon Foundation Sawyer Seminar award and hosted by the Center for Spatial and Textual Analysis (CESTA), the series convenes humanities scholars across disciplines to discuss the pervasive nature of data and to explore its place in the humanities.

“This series is a venue that allows humanists to bring the conversation around data more cogently together across different humanities fields and in relation to data science,” said Giovanna Ceserani, associate professor of classics in the School of Humanities and Sciences, CESTA faculty director, and one of the faculty leads of the series. 

The next seminar will take place on May 30, followed by a symposium on May 31. 

Humanists and data

While data is often associated with the work of engineers, mathematicians, and data scientists, humanities scholars argue that much can be lost when a subject – such as historical events, demographics, or human behaviors – is analyzed strictly by the numbers it produces.

“If we start to understand the world only in terms of data, we may lose the more complex interpretations that come with a traditional humanistic inquiry,” Ceserani said.

Humanities scholarship can analyze subjects and their data through unique lenses, producing novel insights. In the first seminar, “The Place of Data,” scholars Alan Liu and Roopika Risam explored how data can reflect or cause modern social divisions. Their research analyzes data related to geography, race, and gender, examining how this information intersects with social divisions and how data can address these issues.

In another seminar titled “Catastrophe, Data, and Transformation,” historian Jessica Otis discussed her NSF-funded project Death by Numbers. This project involved transcribing, publishing, and analyzing London’s mortality statistics to understand how the lived experiences of plague outbreaks intersected with the mentality of early modern England, while Dagomar Degroot discussed modern climate data.

The fifth part of the series featured Marlene Daut, a professor of French and African American studies at Yale University, in a seminar called “Recuperating Forgotten Narratives,” which explored what happens to text when it is digitized and turned into data.

Daut discussed her work to collect and digitize early 19th-century Haitian print media, such as old newspaper articles and other rare or inaccessible materials, documenting Haiti around the time of the country’s revolution. By making such documents accessible online, she found they often contradicted conventional portrayals of Haiti or corrected false narratives about, for example, France’s involvement in slavery there. She also found that Haitians visiting the site were using it to identify relatives for genealogical research.

“The papers [and] the almanacs contained all these names, and so people are using this digital archive the way that they use municipal archives and physical archives, which is to help complete their family trees,” Daut said.

Discussion of the significance of data for family tree research continued in the following seminar, “The Data of Enslavement,” presented by historians Greg O’Malley and Alex Borucki. They shared their project of the Intra-American Slave Trade Database, an online research tool that documents more than 35,000 slave trading voyages from one port in the Americas to another. In related seminars, scholar Lauren Klein discussed different ethical approaches to the data of the slave trade illustrating how visual choices can humanize the stories data tells, and scholar Ayesha Hardison discussed her work collecting and preserving thousands of novels and other texts by Black authors. 

Approaches to data

Data scientists often acquire a large data set from a particular source, then organize and analyze it. Humanities research tends to be more selective, explained Chloé Brault, a Stanford PhD candidate in comparative literature.

“We’re often in the practice of creating our own data,” she said. “For example, that might look like a literary scholar selecting 100 novels to analyze.”

Brault, who is a Mellon Sawyer dissertation fellow, is working on a dissertation investigating the literary production of 1970s Montreal. Using computational tools and carefully selected texts, she measures how literary language changed, or remained the same, between the 1970s and 2020s. 

Ceserani said that asking “What is the place of data in our work?” forces humanists to look at the objects they study differently and more critically.

“It forces us to ask questions of data scientists, of their received sources, and their evidence, and to clarify what rises to the status of evidence in their inquiries,” she said. "It also forces us to enter into productive conversations with data scientists, and ask questions of their received data sources." 

Upcoming events

The final seminar is titled “Ancient Data and Its Divisions” and presented by Chiara Palladino, assistant professor of classics at Furman University, Eric Harvey, a researcher collaborating with CESTA, and Chris Johanson, associate professor of classics and digital humanities at the University of California, Los Angeles. It will take place Thursday, May 30, from 5:30 to 7 p.m. at Wallenberg Hall. 

The Mellon Sawyer Seminar Series will conclude with a symposium on Friday, May 31, from 9 a.m. to 5 p.m. at Wallenberg Hall. In addition to Ceserani, the faculty PIs for the series are Mark Algee-Hewitt, associate professor of English and digital humanities, Grant Parker, associate professor of classics and African and African American studies, and Laura Stokes, associate professor of history. The series also supports Mellon Sawyer graduate dissertation fellow Matthew Warner, a PhD candidate in English, and postdoctoral scholar Dr. Nichole Nomura, who studies 20th-century American literature and digital humanities. 

In addition to CESTA, the series and symposium are supported by the Stanford Humanities Center (SHC) and the Dean’s Office of the School of Humanities and Sciences.