I have been working on my family tree for years and cannot find anyone on my father's side earlier than my great-grandparents Texas Williams, 1871-1951, and his wife, Nettie Howard Williams, 1875-1912.
My father said that Texas always said that they were really “Dunns"—that the Dunn family sold them to the Williams family and they took the last name of Williams. All of this occurred, I think, in Talbot County, Ga. Yet I have been unable to find any information on slave owners in Georgia with the surname of Dunn. Can you help me find the Dunns and determine whether they may have owned my ancestors? —Eddie B. McCoy
As you have found in your own research, the matter of surnames is complicated when it comes to tracing enslaved ancestors. As noted in a previous Tracing Your Roots column, “How Do I Trace an Ancestor Through a Name Change?”:
Slaves, as chattel property, did not have legal surnames, even though they used first and last names among themselves. Masters and the law identified slaves by the first names they used (or names forced upon them by their owners) in order to distinguish them from other slaves in records such as wills and tax records. … But surnames were not recognized by law, even if a slave family insisted upon using them.
It was common for enslaved African Americans to adopt the surname of their slave owners for their own use. Moreover, if a slave was sold to a new owner, the slave often adopted the surname of the new master, but not always. In fact, on one branch of Professor Gates’ mother's family line, his fourth great-grandparents were Joe and Sarah Bruce, but they were owned by Abraham Van Meter, who freed them in his will in 1832 in Hardy County, Va. (now West Virginia). The family still has not determined where the surname "Bruce" originated, but it is constantly looking.
After the abolition of slavery, a freed African American might return to a prior surname—or adopt a totally new one. As the column referenced above described:
Some African Americans, following the Civil War, quite understandably decided to declare themselves free citizens through the selection of a new surname—names such as “Freeman,” for instance—or simply by a name that they found meaningful for their own private reasons. As Booker T. Washington pointed out in Up From Slavery, changing a surname was an attempt by a former slave to gain some psychological distance from a harsh master, specifically, or from the harsh realities of the nightmare of slavery, more generally.
Your great-grandfather’s assertion that your relatives were “really Dunns” implies that something made the Dunn name more meaningful than Williams. To confirm whether your ancestors were originally owned by a Dunn family and later sold to a Williams family, we tried to trace Texas Williams’ ancestry back one generation to his parents. Texas, born in 1871, was born after slavery had been abolished and was, therefore, likely born with the Williams surname. His parents, who were likely born between 1840 and 1850 (possibly earlier or later), may have been the ones eventually sold to a Williams family.
In order to trace Texas’ ancestry back to his parents, we first consulted U.S. census records. According to your records, Texas was born in 1871. The 1910, 1930 and 1940 U.S. censuses, moreover, show that he was born in Georgia. With this information, we proceeded to search the 1880 U.S. census for evidence of Texas and his family residing in Georgia. At the taking of the 1880 census, we would expect Texas to have been about 9 years of age. Unfortunately, our search of that census did not produce a strong match for a Texas Williams or a Texas Dunn residing in Georgia.
We continued by searching the 1870 census for Williams families residing in Talbot County, Ga. Since Texas was born in 1871, his parents were likely living together by 1870. Our search of the 1870 census produced a number of black individuals and families residing in Talbot County with the Williams surname, including Willis and Caroline Williams; Peter Williams and Dollie Williams; Jerry Williams; Lett and Emma Williams; and Louis Williams.
Of particular interest was the census record for James Williams, age 36, living in the household of a white man named William Dunn, age 26! Also in the household were a 45-year-old black woman named Rachal Hickey; a white female age “0,” identified only with the surname Dunn; and a white 4-year-old boy named Franklin Williams. If this particular James was the father of Texas (or related to him), one possibility is that James may have returned and sought employment with a family member of his original slave owners (the Dunns).
Unfortunately, records of births in Georgia were not maintained at the state level until 1919, making it difficult to locate a birth record for Texas Williams and confirm if any of the above listed individuals were his parents. A record of Texas’ birth might be found, however, in local church records. We suggest investigating available church records in the towns where the above Williams families were residing at the taking of the 1870 census.
Unable to definitively identify Texas’ parents in census records, we next consulted the 1850 and 1860 U.S. Slave Schedules (available at Ancestry.com; subscription required). These schedules include important information such as the name of the slaveholder, along with the age, sex and color of the owner’s slaves. Sadly, because the enslaved were considered to be property, the majority of these schedules did not list slaves by name.
Given that Texas’ parents were likely born between 1840 and 1850, we searched the slave schedules for males and females born within this date range. Moreover, you suggest that your ancestors may have been sold to a Williams family in Talbot County, Ga. With this information, we proceeded to search the 1850 and 1860 slave schedules for evidence of slaveholders with the Dunn surname residing in Talbot County with slaves born between 1840 and 1850.
Our search of the 1850 Slave Schedule produced a record of a slave owner, Martin Dunn, residing in Talbot County. The only slave listed in his household was a female, who was 9 years of age, giving her an approximate date of birth in 1841. It is possible that this is Texas’ mother. Our search of the 1860 schedule did not produce any matching records. Our search of the 1860 Slave Schedule for other slave owners with the Dunn surname residing in Talbot County also did not produce any strong matches.
We were also able to locate a record of interest in nearby Harris County, Ga., which adjoins Talbot County: an entry for slave owner W.M. Dunn. At the enumeration of the 1860 Slave Schedule, W.M. Dunn was residing in District 679 of Talbot County, and owned five slaves. His only male slave was 8 years of age. His female slaves were ages 30, 10, 6 and 4. The ages of W.M. Dunn’s male and female slaves fall within the general range of Texas’ parents’ approximate dates of birth, lending them as potential candidates for Texas’ parents.
We also searched the 1850 and 1860 Slave Schedules under the Williams surname, in the event that your ancestors had already been sold to a Williams family. We were able to locate a slave owner with the Williams surname, James A. Williams, in the 1860 Slave Schedule residing in Talbot County. James’ only male slave is listed as 55 years of age, making him too old to have been Texas’ father. James also owned three female slaves, ages 55, 12 and 7, respectively. The 1850 Slave Schedule also lists a Josiah Williams of Talbot County, who owned 20 slaves. Several of his slaves are the right ages to have been Texas’ parents.
To investigate further whether the Martin Dunn of Talbot County, the W.M. Dunn of Harris County, the James A. Williams of Talbot County or the Josiah Williams of Talbot County was the owner of your slave ancestors, we recommend consulting available Talbot and Harris county probate and land records. Slave owners often bequeathed their slaves to relatives in their wills, and sometimes even listed their slaves by name. Consulting Talbot and Harris county wills for individuals with the Dunn and Williams surnames may allow you to identify possible parents of Texas Williams. Some Talbot and Harris county probate records have been digitized and can be browsed at Ancestry.com using its database Georgia, Wills and Probate Records, 1742-1992.
Talbot County land records may also hold clues that will allow you to confirm the story that your ancestors were sold from a Dunn family to a Williams family. The purchasing and selling of slaves was sometimes recorded in deeds. You could search available deed indexes and deed records for Talbot County for evidence of transactions between individuals with the Dunn and Williams surnames. Some of Talbot County’s land records have been microfilmed and can be ordered from the Family History Library in Salt Lake City and delivered to your nearest Family History Center.
Henry Louis Gates Jr. is the Alphonse Fletcher University Professor and founding director of the Hutchins Center for African and African American Research at Harvard University. He is also chairman of The Root. Follow him on Twitter and Facebook.
Send your questions about tracing your own roots to TracingYourRoots@theroot.com.
This answer was provided in consultation with Daniel Sousa, a researcher from the New England Historic Genealogical Society. Founded in 1845, NEHGS is the country’s leading nonprofit resource for family history research. Its website, AmericanAncestors.org, contains more than 300 million searchable records for research in New England, New York and beyond. With the leading experts in the field, NEHGS staff can provide assistance and guidance for questions in most research areas. They can also be hired to conduct research on your family. Learn more today about researching African-American roots.