Slave Biographies


Slave Biographies: The Atlantic Database Network is an open access data repository of information on the identities of enslaved people in the Atlantic World. It includes the names, ethnicities, skills, occupations, and illnesses of individual slaves. Phase one of a multi-phase project is presented here. Users of the website can access data about slaves in colonial Louisiana and Maranhão, Brazil. They can download datasets, search for ancestors, and run statistical analysis.

For phase two of the project, we invite researchers of slavery in the Atlantic World to contribute new databases. They will be able to share their data widely and be cited for it. They will also be able to store their data securely, use analytical tools we are developing, and collaborate with other researchers. Our goal is to make data about Atlantic slavery widely available to scholars, teachers, and the public. It is also to create a platform that combines multiple, individual datasets in a way that is complimentary and creates a resource for quantitative data analysis and visualizations.

The Project's Beginnings

The Slave Biographies project owes much of its current success to the work of numerous scholars and researchers who, working in isolation during much of the 1980s-1990s, began to collect, catalogue, and organize slave records in the Atlantic region. The initial Slave Biographies data collection is the compilation of the private databases of two of these scholars, Gwendolyn Midlo Hall and Walter Hawthorne.

Gwendolyn Midlo Hall created a landmark database, released in 2000, called the Louisiana Slave Database: the result of Hall's patient and arduous work since 1984. It recorded over 100,000 descriptions of slaves in many different types of documents many of which contained rich information about slaves living in Louisiana between 1719 and 1820. Deservedly, her work received much attention from the national and international media beginning with a front page story in the Sunday New York Times, Identity Restored to 100,000 Louisiana Slaves. It was published on the world wide web with a user friendly search engine in 2001 and also incorporated into the search engine of and featured in a number of books, articles and conferences of several scholarly disciplines.

Walter Hawthorne spent significant time in the Brazilian state of Maranhão and later assembled the Maranhão Inventories Slave Database (MISD). Hawthorne plans to take a second research trip to Brazil's other Amazonian state, Pará, to compile another, compatible dataset. Hawthorne’s dataset contains information about the lives of about 8,500 slaves in Maranhão from the mid-eighteenth century through the early nineteenth century. Data was derived inventories of slaveholders’ possessions that can be found in un-cataloged boxes in the Arquivo Judiciário do Estado do Maranhão in São Luís, Brazil.

A Clear Need

There is a significant need for a collaborative research site about Atlantic slavery. During the past two decades, there has been a seismic change in perception about what we can know about African slaves and their descendants throughout the Atlantic World (Africa, Europe, North and South America). Scholars have realized that, far from being either non-existent or extremely scarce, various types of documentation about African slaves and their descendants throughout the Atlantic abound in archives, courthouses, churches, government offices, museums, ports, and private collections spread throughout the Atlantic World.

Since the 1980s, a number of major databases were constructed in original digital format and used in major publications of their creators, but they lack a platform for preservation and therefore are at risk of being lost as their creators retire. Also, a number of collections of original manuscript documents are beginning to be digitized and made accessible free of charge over the Web.

Our task as historians is more than to preserve images of primary sources; it is to interpret those sources by finding new ways to organize, share, mine and analyze as well as to preserve original materials which might otherwise be discarded or lost. Furthermore, new scholarship is interested in observing the large historical patterns and tendencies. Answering important questions about Atlantic slavery requires the gathering of larger amounts of quantitative data than one individual collection could ever hope to compile and interpret. Slave Biographies fills a real need. It will serve as a collaborative platform for researchers of African slaves in the Atlantic World to upload data they have collected and link it to other datasets, creating a much richer resource than the sum of the individual datasets.

What We've Done So Far

Slave Biographies began officially in 2011 with a $99,994 grant from the National Endowment for the Humanities. This initial funding allowed the development team to define a core set of comprehensive metadata fields for biographical data about slaves and create a database for this information in Kora, the open source digital repository platform that powers Slave Biographies. Two collections were normalized and cross-walked into the comprehensive database fields.

During the initial funding phase, Slave Biographies produced a data guidebook that explains the data structure, describes the types of records in the system, and defines each field, along with controlled vocabularies to standardize data in several fields, as appropriate. Slave Biographies also developed a set of integrated digital tools that enable researchers, students and the public to perform statistical calculations. Usuers can also download the dataset in Excel format and can use search functions to locate individual slaves.

Looking Ahead

Although much has already been accomplished, there is still more to do. Slave Biographies has plans to develop multiple user interfaces that will support the diverse needs of our audiences including genealogists, students, and researchers. Further we plan a visualization layer by developing means of displaying the complex web of kinship and other social network relationships that are well-represented in contributed datasets. Finally we intend to devise a mechanism that will allow the database system to determine the likelihood that slave records from two or more primary source documents describe the same individual.