Database of southern Georgia enslaved peoples

David Carlson's picture

I’m seeking a bit of guidance. For the past year or so, I have been working on a project that needs a bit of direction. For the past year or so, I have combed through newspapers and court records in southern Georgia between 1845 and 1865 to extract information on the region’s enslaved population. This began as a side-project/hobby as I worked on other projects. I am a Civil War historian, not a historian of slavery per se. But as those projects near their end, this one has captured my attention and begun to assume a life of its own.

The newsprint information is similar to the many runaway slave ad projects that are on the web: names, ages, physical descriptions, enslavers’ names, locations, rewards, markings and injuries, etc. Court records are also yielding names, ages, owners, locations, as well as sales and appraisal valuations, inheritance patterns, and enslaved family groupings (sometimes across three generations). Unlike the existing runaway slave ads, I have not recorded these using a content management system like Omeka. Instead, I am creating a relational database in Excel with the hope that it can serve as the basis for an Access database (once I learn it) or some other database management system that would allow those more fluent in computational methods might find useful. To date, I have surveyed half of the available records and amassed over 10,000 enslaved people's names and data.

The issue I am running into is that I see no clear path to completion. I enjoy the process in which I find myself engaged, but, not being a historian of slavery, the possibilities of the end product elude me. How would I “publish” such a database? Is such a database valuable? What kinds of questions are out there that such a database might serve? Is there value in producing such a database simply under the hope that others might find research inspiration from it? In my mind, I can envision an extension of the data that linked detailed demographic information on slaves with white property ownership patterns to produce GIS-maps of slave distributions, ownership transfers, family stability and continuity, but, again, we’re getting into areas about which I know little.

Any suggestions and comments would be greatly appreciated.

The short answer is YES, a database -- particularly a relational database with data focusing on the enslaved drawn from myriad sources -- will be enormously valuable. It will be valuable in ways you can and cannot conceive of now, and that's okay. Simply making information available for other scholars to use is enormously helpful to our shared project of understanding the past more fully and accurately.

That said, having the input of other scholars who have developed or used similar databases might be very helpful as you are in the developmental stage. You are probably already aware of databases of advertisements seeking those who stole themselves free, and of transatlantic slaving ships. Adam Domby and Patrick Sheridan are working a database they are creating documenting the work of enslaved laborers impressed by the Confederacy. Anelise Shrout is creating a relational database from the records of one of the antebellum Black churches in Richmond. I am sure there are other projects out there, or in the works, that might be useful to learn about. You might reach out to places like the Center for History and New Media at George Mason University to see how folks who have worked on a variety of projects approach the creation of new databases.

Perhaps most useful would be for you to put together an advisory board -- informal is fine -- of folks who either 1) have worked on creating/maintaining similar databases; and/or 2) have done research in such databases; and/or 3) do research on the lives of the enslaved in the particular place/time covered by the records you are collecting. Each group will have their own perspective on what would be useful, and taken collectively their input -- along with your considerable work on this project -- will be of great benefit. (I am not quite a neat fit into categories 1, 2, or 3, but I would be delighted to be part of the advisory board.)

Best of luck with this YES VERY VALUABLE project,

Lois Leveen, PhD
public humanities scholar

I have some similar databases being developed in my current project, and I think the best places for them are the newish Journal of Slavery and Data Preservation which publishes peer reviewed data sets and maybe a connection to


Adam Arenson (he/him)
Professor of History
Manhattan College

I am grateful for the replies both in and out of H-South. Most comments share a common focus on and the Journal of Slavery and Data Preservation as outlets for database file(s) once completed. I do, however, like what Stephen Berry and others have done with the digital history projects in making the raw files available but also providing extremely useful graphical interfaces, GIS displays, and traditional research write-ups on stand-alone websites. If anyone were willing to donate some time to look at my database architecture to see if it seems logical and amenable to such things, it would be greatly appreciated.

Mr Carlson,
I am excited about the possibilities of your project. I have been entering slave and slave master information in Excel spreadsheets since 1997. My focus is Upson County, Georgia, compiling and correlating each bit of information available for about 7,000 of the slaves who lived in the county 1824-1865, placing the events in chronological order so as to build whatever biography I can for each person. My method is to data-mine complete datasets (no sampling) of all records from the county, supplemented by specifically-defined extracts of data from other places and datasets that relate to any of the individuals in my Upson data. The connections you can make from record to record, are truly startling and wonderfully instructive. The biographical sketches in my 2009 Article in JSH, "Slaves, Slavery, and Cash in a Georgia Village," were only possible by using the personal data compiled in my spreadsheets.
Unfortunately, when I started the project, I did not know about relational databases--and after I started building ODBC products for my employer, and took some relevant courses, I was too far invested in my current table structure to change it. However, I have designed relational database proposals for other slave data, so I understand many of the challenges. My 2001 article in _The American Archivist_, "A Perspective on Indexing Slaves' Names" (Vol 64, No. 1, pp. 132-142) was actually an argument for making masters' names (as a surname surrogate for the enslaved) the primary key in a relational database.
I recently completed a compilation of all newspaper runaway ads related to Upson County, available as a .pdf doc if anyone wants it. My database was able to provide biographical details (in some cases quite extensive) for some of the fugitive people. (A major triumph of that project was that I broke the code on those letters and numbers that appear at the foot of most antebellum ads.)
Would be glad to exchange ideas with you and everyone else.
David E Paterson
Millington, TN

Hi David,

Another resource for you (or anyone else working on a digital history project) is the NEH Office of Digital Humanities ( They offer a wide variety of grants to help fund the creation, infrastructure, and maintenance of a wide variety of digital projects, as well as workshops and a variety of other things that might help guide your project or help find it a home. Your project is the kind of thing they'd be interested in, although I am not sure what kind of institutional sponsorship they might require for grants.

There are a number of "publishing" options out there - one intermediate step might be to publish a preview of what you're working on in a journal like Current Research in Digital History or similar.

Good luck on the project going forward - it sounds like it has intrinsic value to researchers and the public.