The reality of human history, said historian Fabrício Prado, is much messier than code. But data fluency can make the work of humanists go further.

Associate professor of history Fabrício Prado.
Associate professor of history Fabrício Prado.

After publishing his second book, Prado, an associate professor of history at William & Mary, was left with a huge database and a story yet to tell about modern nation-building in the Americas. A few years later, this database served as foundation for the W&M Global Americas Lab – a collaborative project combining historiographical work with STEM opportunities for students.

Through its ever-expanding collection of shipping records, the Global Americas Lab examines political and social networks between North and South America during the Age of Atlantic Revolutions. It also engages its students in opportunities beyond the essay format – such as database creation, machine learning projects and geographic information systems map design. 

“I realized that students were eager to engage in research,” said Prado. “W&M students really want to be hands-on and acquire more of a professional experience.”

“These skills are going to be useful in the job market, but what is also interesting is that data science students fell in love with the digital humanities,” said Prado, in full alignment with the data and careers initiatives from William & Mary’s Vision 2026 strategic plan.

“That’s very much a W&M thing,” he said. “We combined teaching and research: That’s the only way this project could come to life.”

Making the humanities accessible

The Global Americas Lab has a data and computation specialist in Ethan Meidinger ’25, a data science major. Working on the project gave him the opportunity to engage with historical data from his home country of Argentina – while developing new skills in computer vision, a field enabling computers to characterize and identify different patterns in objects. 

Notably, Meidinger developed a machine learning model capable of transcribing historical handwritten text in Spanish. 

Daniel Vasiliu, an associate teaching professor of data science cooperating with the Global Americas Lab, said that the goal is creating an algorithm that can be applied to any language using Latin script.

 “We will soon be able to search within manuscripts, and I have the equivalent of 10,000 pages of that,” said Prado.

handwritten records from Buenos Aires in the 19th century.
An example of the historical records the Global Americas Lab works with.

According to Bennett Snyder ’25, a history and government major, such tools make it much easier and quicker for historians to read documents and search for important phrases, names or places. 

“Extensive data will show scholars the importance of U.S.-Latin American relations,” said government major Amy Weitzman ’24. “My favorite aspect of the coordination is that data science makes the humanities accessible.”

Global Americas data will eventually be available to a wider audience, she said, highlighting the volume of trade between North and South America in the 18th and 19th centuries and its impact on historical events.

“When we use data in conversation with qualitative historical sources, we can describe the fuller picture of what was going on at a particular time,” said Snyder. “By using historical shipping data, we are able to see broad patterns that would not be evident if we just looked at merchant letters or ship records.”  

Snyder, who is studying U.S. trade with Haiti after the country’s independence from France, said that Global Americas allowed him to think more globally. He uncovered histories of people treated inhumanely through the trans-Atlantic slave trade “at literally the exact same time Haiti was becoming the world’s first Black-led nation outside of Africa and a center for Black freedom and anti-slavery.”

Xincheng Hou ’24, a history major in the St Andrews William & Mary Joint Degree Programme, uses GIS to map trading activities between the U.S. and East Asia in the early 19th century. He highlighted the lab’s focus on both qualitative and quantitative skills – but also the interdisciplinary possibilities allowed by a digital humanities approach in general.

“By familiarizing themselves with statistical software and other digital technologies, historians can more efficiently and accurately present comprehensive pictures of the topics they study,” he said. As an example, he mentioned the potential of 3D modeling for urban historians, and the dialogue it could open with experts in architectural design, soundscape and environmental-related fields.

A common language for the digital humanities 

“Rigorous research now requires historians to be fluent in this type of digital humanities,” said Prado. 

In the beginning, he said, humanists, social scientists and data scientists in the Global Americas Lab had to find a common language – for example, when agreeing on what a “good” database looked like for the project’s purposes. In the end, a synergy was established among all lab participants.

“For example, there are humanists and social scientists learning Python; they are getting into the groove of how to fix their code,” said Prado. Data scientists, on the other hand, started considering not just the technical aspects, but the context in which data was created and applied.

“Our data scientists understood that the machines need to adapt and adjust to the human experience,” he said.

The Global Americas Lab represented a learning opportunity for its lead, too.

“I was a humanist who knew how to use tables,” said Prado. “Through my experience in the lab, I can say that I feel I am now a digital humanist.”

When approaching his own research work, he now thinks about the architecture of the software he will need to use to balance efficiency with historical rigor.

Prado plans to apply for external funding to expand the remit of his project. He noted support from the Charles Center – an “instrumental incubator” that allowed this project to come to life – and the collaborative nature of his lab, which also involves the Developers Club from the W&M computer science department.

Meidinger urged humanists and data and computer scientists to seize opportunities for collaboration when they arise.

 “It’s important to be a specialist in what you do, but also to pick up skills from other disciplines to become a more well-rounded and capable academic or professional,” he said.

Snyder, too, reflected on career opportunities.

“My involvement in Global Americas has made me see history as a viable career path for myself,” he said. “I have always loved history and this lab has shown me how rewarding it is to uncover things from our past that we did not know or were silenced or hidden before.”

, Senior Research Writer