Genealogists have brick walls, too

 Genealogists are often asked how they got into the profession. I got into genealogy simply because I was born after much of my close family had passed, and wanted to be able to better visualize the relationships between everyone my parents and other living relatives spoke of. I quickly found the project very fun, and began learning names and finding places of ancestors whose memory hadn’t survived to my time.

But what ‘hooked’ me was my first brick wall. Over a decade later, that brick wall is still there, though I’ve made a little progress on knocking it down.

That brick wall is my third great-grandmother, Emma Thompson, who lived in Robeson, North Carolina around the turn of the 20th century.

All I know about Emma’s life comes from exactly one document: the 1900 census¹, where she lived in Sterling, Robeson County with her husband John Arnett and six children. Her eldest child is my 2nd great-grandmother, Easter Blanche Arnett. Easter is 13 years old, while her mother Emma is listed as 24 years old. If this is correct, Emma was only 11 years old when Easter was born, while husband John was 20. The census further reported that Emma and John have been married for 16 years, which would mean she was only about 8 years old at marriage! It’s easy to assume (and hope) that Emma’s age was simply misreported by a decade, but when I consider the fact that John Arnett’s own mother was only 13 when her relationship with his father, 22 years her senior, began, I am not so certain (more on this in a future post, perhaps).

By 1910², Emma had vanished without a trace, and John Arnett was remarried to Lonnie Magnolia ‘Maggie’ Connor. He was head of a household with Lonnie, her son from a previous relationship, and 2 children born to him and Lonnie (the first arriving in 1906), as well as a mystery 24-year-old ‘Ava J Arnett’, reported as a sister of John, though I’ve never been able to place her as both of John’s parents were apparently dead before her alleged birth year. It’s possible she’s the 12-year-old daughter Catherine of John and Emma from the 1900 census, a person who, like Emma, I’ve never been able to confirm the existence of after that record. The youngest two children of Emma and John lived with neighbors next door, while the elder siblings were already married.

So what happened to Emma? All of this led me to assume that Emma must have died between 1900 and 1906, and probably closer to 1900, since no children of hers born after the 1900 census have been identified. No obituary for Emma was ever printed, no birth record filed, no headstone preserved (John and Lonnie both have wooden cemetery markers, so if Emma had anything in a local cemetery, it’s probably gone). Then I located the marriage record for John and Emma’s youngest daughter, Willie, filed in July of 1915 at Lumberton³. Willie is described as “the daughter of Jno Arnett and _____[left blank] the father living, the mother living.” More errors, or did Emma abandon her family, or was she otherwise removed from it?

Thankfully, there was just a bit more to learn about Emma from her family’s vital records. Her maiden name is provided by the 1970 death certificate of son Wade Arnette⁴, where she is ‘Eunice Thompson’; and by the Social Security Application of daughter Mary Della Arnett-Scott, where she is ‘Emmie Thompson’⁵.

After finding these documents, I spent a very long time mapping out the families of every single Thompson in Robeson and surrounding counties, to no avail. That is, to no avail until I had access to DNA test results, and found a very strong and large cluster of matches sharing the appropriate amount of DNA to my father to suggest that Emma’s parent was one of the six children of well-to-do Robeson family William Berry Thompson (1818-1857) and Dorothy ‘Dolly’ Regan (1821-1858). Which of the children it is, I’m still not sure, though my father has no DNA matches among the many descendants of eldest son Stephen Archibald Thompson, so it’s probably not him. Two of the Thompson children, daughter Elizabeth Martimer Thompson, born 1846, and son William Berry Thompson, born 1853, disappear after the 1870 census, so perhaps one of them is a good candidate. But I have no evidence to support either hypothesis at this time, it’s just a guess.

Emma is never enumerated as ‘Emma Thompson’ or ‘Eunice Thompson’ in any census, as I likewise mapped out every Emma and Eunice born from 1860-1880 in the vicinity. But at least one of her children passed down this name to his daughter, who would serve as her father’s death certificate informant. Emma never lives with any member of the Thompson family as far as I’ve found. Given that her husband and children were of a different socioeconomic status from the Thompsons, I’ve assumed she was an illegitimate child, but no bastardy bond was ever filed for her.

I’ve tried very hard to work out the identity of Emma’s other parent. Unfortunately, pretty much every person in Robeson County is very related to every other person, and DNA leads that looked good at first collapsed, or ended up at their own brick walls of illegitimate children.

Emma’s identity was the first mystery to ‘hook’ me in genealogy, and her parentage and what became of her still elude me. It frustrates me to no end, as a genealogist, that I know so little about a woman who lived in the 20th century — far less than I know about many ancestors who lived in the 18th century! I still hope that one day I will find Emma, and go back to the case every few months.

There is some boilerplate in every letter of agreement I make with a client: “You understand that I can make no guarantees regarding what information and documentation will be found. I will, however, do everything within my power to find the information that you have requested.” So far, I’ve never found nothing about a client’s ancestor. In fact, I’ve never found anything close to nothing! But at some point, the available evidence to examine wears thin, as it has for Emma, after the hundreds of hours I’ve spent chasing her. 

So, I can’t guarantee that I can find your Emma, but I can certainly look carefully at every source I could possibly imagine finding her in, and I can probably make some educated guesses about the story of her life. Hopefully, one day someone will find a lost records collection in an attic (I’m rooting for the 1910 marriages of Marlboro County, South Carolina, for Emma’s sake — that’s when and where I expect one of her daughters was married) or basement; or some person with the magic DNA inheritance will test, and suddenly, all the good work done in the past will fit together and reveal the truth. Fingers crossed. Until then, some brick walls will have to wait to come down. 

0. North and South Carolina. Mitchell, Samuel Augustus Jr. 1868. (Philadelphia, 1868); digital image, David Rumsey Map Collection (http://www.davidrumsey.com : accessed 6 Sep 2022)

1, 1900 U.S. Census, District 118, Sterling, Robeson, North Carolina, p. 17B, line 93, John Arnett; image, Ancestry.com ; citing National Archives and Records Administration microfilm T623, roll 1214.

2. 1910 U.S. Census, District 120, Sterlings Mill, Robeson, North Carolina p. 3A, line 42, John Arnett; image, Ancestry.com ; citing National Archives and Records Administration microfilm T624, roll 1129.

3. Robeson County Office of Register of Deeds, marriage register p. 511 (1915), Robert Hendrix and Willie Arnett; image, Ancestry.com (http://www.Ancestry.com : accessed 06 Sep 2022); citing North Carolina County Registers of Deeds, North Carolina State Archives, Raleigh; North Carolina, U.S., Marriage Records, 1741-2011; Marriage Register (1800 – 1983); RG 048.

4. North Carolina State Board of Health and Bureau of Vital Statistics, death certificate, 36315 (1970), Wade Arnette image, Ancestry.com (http://www.Ancestry.com : accessed 06 Sep 2022); citing North Carolina State Archives, Raleigh, North Carolina, North Carolina Death Certificates, Microfilm S.123; Month: 10, County: Robeson.

5. Social Security Administration, “Application (SS-5) Files, 1936 – 2007 (Last Names S through T),” database, Archives.gov (aad.archives.gov : accessed 06 Sep 2022), entry for Mary DellaArnette Scott, 1943, SS no. 246288782, Reference no. 65286429315.