Camp Sumter at Andersonville: The notorious Civil War prison

Submitted by Amanda Morrow

Camp Sumter, now Andersonville National Historic Site, located near Americus in western Georgia, opened for the first time to prisoners captured from Union forces on February 24, 1864—150 years ago today. Many of the first prisoners of war (POWs) traveled from Richmond, over the often-unreliable railway system of the Confederate interior. The prison was an open stockade; originally comprising 16.5 acres it was later enlarged by POW labor to 26.5 acres. The original stockade was only meant to accommodate 6,000 men, but at the height of its population (after stockade expansion) it held about 33,000. When the first POWs arrived, the stockade was not even fully enclosed (Marvel 1994).

Almost 13,000 men died at Andersonville. Malnutrition, exposure, and disease were the primary culprits. Surprisingly, the majority of the graves at Andersonville National Cemetery are properly identified. This is almost unique among Civil War Prison cemeteries. By October of 1864, Andersonville was being emptied of all but the sickest prisoners. The Confederacy recognized that the Andersonville site was no longer suitable for a prison, especially one of that magnitude. Many of the surviving POWs were transported to another Georgia prison site, Camp Lawton in Millen.

Many POWs published memoirs after their incarceration. The writings of John Ransom, John McElroy, and Robert Sneden are available in print, but other less-commonly known memoirs are now available to read online.

Review of Archaeology at the Camp Sumter

Congress established Andersonville National Historic Site (ANHS) under National Park Service (NPS) jurisdiction on June 30, 1971 (Bearss 1970). The first result of the NPS proposal ANDE-H-1 was a report on the history of the camp by NPS historian Edwin Bearss. He gathered all the documentary evidence available including maps, photographs, and primary accounts of prison life (Bearss 1970). The report mainly focused on loci that might leave archeological features behind, such as the stockade itself, other support structures, and prisoner huts (Bearss 1970). Bearss’s report also covers events that took place after the Civil War era occupation that might have left archaeological evidence behind, and even includes eyewitness descriptions of the condition of the site through the end of the nineteenth century.

Following Bearss’s report, University of West Georgia archeologists Lewis Larson and Ray Crook conducted the first archaeological investigations at the camp in 1973 and 1974. They determined that ANHS contained both historic and prehistoric remains with the historic remains dating primarily to the Civil War (Larson and Crook 1975). Larson and Crook recorded several potential archaeological features including portions of the outer and inner stockade walls and the north gate. In 1978, Southeast Archeological Center (SEAC) archaeologist Ellen Ehrenhard began investigations which uncovered additional stockade features in the southern portion of the prison along with the south gate, a hospital, and Captain Wirz’s office (Paglione 1984). A report was never completed for this fieldwork (Prentice, personal communication, 2012). Ehrenhard also tested one prehistoric site on ANHS property. The assemblage included artifacts ranging from the Paleoindian through the late Woodland periods (Paglione 1984).

In 1984, SEAC archeologist Teresa Paglione conducted an archeological survey on a tract of land adjacent to the prison site that NPS was considering for disposal. She recommended that NPS retain the parcel (Paglione 1984). In July of 1985, SEAC archaeologists conducted a soil resistivity survey of the potential hospital area delineated by Ehrenhard in 1978 (Marrinan and Wild 1985). Projections for the location of the hospital came from Bearss’s original report (1970) and from the recommendations of Ehrenhard (Marrinan and Wild 1985). They concluded that resistivity was not the ideal remote sensing technique for the sandy soils inherent to ANHS (Marrinan and Wild 1985).

The second phase of archaeological investigations at ANHS by SEAC archeologists began in 1989 (Prentice and Mathison 1989). This was required as a result of a 1987 amendment issued by NPS which proposed building reconstructions of certain stockade features, including the gates and walls (Prentice and Mathison 1989). The archaeologists successfully identified the North Gate during this survey and found that it was constructed with squared posts set in a trench that averaged five feet deep (Prentice and Mathison 1989). They also excavated the potential north gate area identified by Larson and Crook (1975), but this instead turned out to be the original wall feature (constructed before the prison was enlarged) along with some non-archaeological features (Prentice and Mathison 1989). The artifact assemblage included 2 iron ax heads and 19 cut nails along with food bone and a number of preserved wooden post samples (Prentice and Mathison 1989). Extremely well-preserved posts were found in situ by the excavations which delineated the stockade wall from the gate (Prentice and Mathison 1989).

SEAC archaeologists also conducted field investigations in 1990 to locate and investigate the southeastern corner of the inner stockade for reconstruction (Prentice and Prentice 1990). Post preservation for this area of the stockade was surprisingly poor, which stood in stark contrast with the excellent post preservation noted the previous season. Two units from Ehrenhard’s 1978 excavation were relocated during this field season (Prentice and Prentice 1990). Generally the posts here appeared to be squared off, as in the corresponding northwest corner and main gate area (Prentice and Prentice 1990). Visitors to the park today can enjoy the fruits of these archaeological labors as both the North Gate and the southeastern corner have been fully reconstructed for interpretation.

In June, 2005, the Georgia Department of Transportation (GDOT) conducted a ground penetrating radar survey at ANHS (GPR) as part of a joint effort with NPS (Pomfret 2005). The partnership between NPS and GDOT began in 2003 and a subsequent memorandum of understanding was signed in 2005 before the work at ANHS began (Pomfret 2005). Their goal was to locate the South Gate, the Dead House, and another hospital structure (Pomfret 2005). The survey precisely identified the location of the South Gate but the other test areas provided only inconclusive evidence (Pomfret 2005). The crew also conducted a GPR survey of Andersonville National Cemetery which showed that earlier POW graves at the site were dug individually and that it was only later in the occupation period that mass graves came into use (Pomfret 2005).


Bearss, E. C. (1970). Andersonville National Historic Site Historic Resource Study and Historical Base Map. Washington DC: Office of History and Historic Architecture,
National Park Service.

Larson, L. H. and Crook, M. R. (1975). An Archeological Investigation at Andersonville National Historic Site, Sumter and Macon Counties, Georgia. Tallahassee: Southeast Archeological Center, National Park Service.

Marrinan, R. A. and Wild Jr., K. S. (1985) Soil Resistivity Survey of the Hospital Site Andersonville National Historic Site. Tallahassee: Southeast Archeological Center, National Park Service.

Marvel, W. (1994) Andersonville: The Last Depot. Chapel Hill: University of North
Carolina Press.

Paglione, T. L. (1984). Archeological Survey and Testing of Tract 01-142: Andersonville National Historic Site, Georgia. Tallahassee: Southeast Archeological Center, National Park Service.

Pomfret, J. (2005). Ground Penetrating Radar Survey of Andersonville National Historic Site. Tallahassee: Southeast Archeological Center, National Park Service.

Prentice, G. and Mathison, M. (1989). Archeological Investigations of the North Gate at Andersonville National Historic Site. Tallahassee: Southeast Archeological Center,
National Park Service.

Prentice, G. (2012). Personal communication, email.

Prentice, M. C. and Prentice, G. (1990). Archeological Investigations of the Southeast Corner of the Inner Stockade at Andersonville National Historic Site, Georgia. Tallahassee: Southeast Archeological Center, National Park Service.