A team of historians, researchers, and developers have joined forces to establish Freedom on the Move (FOTM), a database focused on materials regarding fugitives from slavery in North America. The free, open-source site, created by an interdisciplinary team of scholars and researchers, compiles a wide collection of “runaway ads”—notices about fugitives, placed in newspapers from colonial days through the end of the Civil War.

The advertisements were placed in newspapers both by enslavers, who hoped to locate self-liberated Africans and African Americans, and jailers, who posted notices describing those they had captured. These often contained detailed information about the people they described—appearance, age, gender, race, language, personality, literacy, possessions, dress, skills, personal history, and location—that have the potential to offer a new, comprehensive source of data on the experiences of enslaved people.

A joint project between Cornell University’s Department of History and the Cornell Institute for Social and Economic Research (CISER), Freedom on the Move invites the public to help transcribe the digitized notices and, by answering questions, add metadata to provide a rich level of detail about a little-documented population. The site is available for use by instructors, scholars, students, genealogists, and citizen historians as a research aid, teaching tool, and resource.

BRINGING RECORDS TOGETHER

Scholars of 19th-century North American slavery wishing to learn more about the voices and experiences of the enslaved have mostly relied on narratives written by fugitive or formerly enslaved people, such as those by Frederick Douglass or Harriet Jacobs, or Works Progress Administration (WPA) interviews with formerly enslaved individuals conducted in the 1930s. That pool, however, is limited.

The runaway ads, on the other hand, often contain a wealth of detailed information—offered in the service of recovering enslavers’ “property” or claiming rewards.

While the advertisements have been used as sources individually or as part of local collections, they have never been collected in a central repository. Historians note that enslaved people often escaped more than once, traveled widely, or changed their names—so a single ad couldn’t tell a full story. "From the historian's perspective, it's very hard to work with these sources because they're scattered through all these newspapers,” explained Mary Niall Mitchell, associate professor of history at the University of New Orleans. “There are regional or statewide collections, but if you're trying to use them to look for larger patterns it would be very difficult and time-consuming."

Mitchell and colleague Joshua Rothman, professor of history at the University of Alabama, had been working together on a similar project when they connected with Edward Baptist, professor of history at Cornell’s College of Arts and Sciences. Baptist had for years been considering ways to crowdsource the ads, and was investigating the idea with CISER director William Block, who had experience working on large-scale data sets for the Integrated Public Use Microdata Series at the University of Minnesota. Together they teamed up with Vanessa Holden, assistant professor of history at the University of Kentucky, and Hasan Kwame Jeffries, associate professor of history at Ohio State University to develop FOTM.

With a small grant from the National Endowment for the Humanities (NEH) in 2016, the partners brought in a team of developers from CISER and began designing the interface. They agreed from the start that it would be designed for students, teachers, and citizen historians as well as scholars, and an initial beta version was tested in a few classrooms and with scholars. A subsequent larger NEH grant and funding from the National Archives’ National Historical Publications & Records Commission allowed further customization of the site, which launched on February 14.

THE RIGHT QUESTIONS

Crowdsourcers who visit FOTM, after a brief tutorial, have the choice of transcribing an ad that has never been worked on or proofreading and editing existing transcriptions. They can also choose by state, date, or both. As they begin transcribing, users are walked through a series of questions about each ad; the answers generate its metadata.

Developing questions that would prompt accurate responses while remaining culturally sensitive was a long process, Mitchell told LJ. Cornell librarians provided much of the necessary input.

For instance, she said, almost all of the ads reflect some sort of racial classification, such as Negro, mulatto, griff (denoting an individual who was three-quarters black), or quadroon. However, the question for the user can’t be phrased, “What is the race of the person in the ad?” Instead, it needs to be framed as, “What racial classification is being assigned to the person in the ad?”

“That’s a big difference,” Mitchell noted. “You have to be careful not to naturalize this very constructed racial language. It took us a long time to get that wording the way we wanted it, to reflect that it's something being imposed on the person.”

Other questions include what languages the enslaved person spoke, when they escaped and where they escaped from, basic appearance—including scars and injuries—possessions they carried, whether they could read or write, and what other skills they had. Locations can be tagged by type (market, levee, city, county).

A second set of prompts asks the reader to distinguish the source of the notice. "There's a distinction between the enslaver ads and the jailer ads,” said Mitchell. “Generally, [the enslaver’s] descriptions of the person are a little more detailed, because they know the person in question. The jailer is dependent on the alleged runaway for information, because they have captured someone they think is a fugitive and they have to rely on that person telling their true story”—which, of course, many of those who were self-liberated did not, she noted, as they had no wish to be returned to their enslavers.

The partners envision much of the transcription coming from classrooms. Each ad goes through several rounds of edits, and they hope that teachers will encourage their students to work freely with the material.

GROWING THE DATABASE

Mitchell has been working to develop portals specifically for schools and museums, and recently spoke at the City University of New York’s American Social History Project.

The classroom setting, noted Block, will be important for the site’s success; he hopes that it will change the conversation about the history of slavery in the United States, particularly at the high school level. FOTM’s support for classroom teachers will include links to other source material, annotations on the ads, a glossary of antiquated language, and lesson plans customized for different age groups—as well as a list of dont's, added Block. “For instance, do not do reenactments of slavery.... We see this happening out there somewhere every week."

The partners also envision FOTM museum kiosks at institutions ranging from the National Museum of African American History and Culture to the Whitney Plantation Museum in Edgard, LA.

The database itself is a work in progress, and will improve as more ads and metadata are entered, but users can currently download CSV files from search results, which can be refined by a number of parameters. For example, said Mitchell, one teacher reported that students are curious about the experience of people their own age. “So she can go in and search, and call up all the ads in the database for 15-year-olds. That kind of tailoring is possible, and then you can download your results."

A FOTM Twitter feed (@fotmproject), launched before the crowdsourcing interface, served to pique people’s interest. The team hopes that by publicizing the database they will bring in contributions from archives and libraries nationwide; FOTM will in turn link back to those local sites, giving them additional exposure as well. “We're not trying to supersede those collections, because they serve a purpose at the local level,” noted Mitchell. “But it's very important that we get as many ads as we can into this larger database.” The team would eventually like to collaborate with similar projects in Latin America and the Caribbean.

Some ads already exist in digital format; others need to be digitized from microfilm. A few have been scanned from books. "We take them however they come,” said Mitchell. “The challenge going forward is figuring out how to find more collections down the road. That means hiring students and people in libraries to help us collect more ads."

A number of the ads repeat, following the same person over the course of their journey. "But the repetition of those texts is in fact itself a kind of data," noted Baptist.

Currently FOTM has more than 20,000 digitized ads, with plans for a total of 40,000 by 2020. Of these, more than 3,500 have been transcribed and enhanced by metadata.

A wide data set of individual stories means that, through linked data, FOTM will eventually be able to uncover patterns of linguistics, migration, material culture, and more. Baptist notes that the site will eventually include some form of mapping capability, as well as GIS data that users can download for their own digital humanities projects. “One of the other things we hope to do is create ways to link this data to other data,” Baptist told LJ, “about enslaved people, about liberated people post-slavery [and] post-emancipation, enslavers, newspapers—there are multiple directions in which these linkages could run. We're trying to explore the best ways to do that."

"I think that eventually, if we're successful,” added Block, “there are all kinds of research related tasks to extend the value and further describe the data that we have, and make it linkable to other data sets. I think that down the road we're going to see that component of the project grow and…provide opportunities for really cool research and connections that don't exist in the original crowdsourced data."

In addition, growing interest from users outside of classrooms is driving the project forward. Descendant communities, originally established by formerly enslaved persons, have expressed enthusiasm, as have genealogists.

The range of details in the stories, Baptist told LJ, is fascinating—from descriptions of clothes and weapons taken from enslavers to the fugitives’ routes of escape to men and women who evaded detection by cross-dressing—FOTM holds stories of “an enormous diversity of ways in which people resisted the institution of slavery."