Tracing Your Roots: Was Great-Grandma Part Creek Indian?

Relatives believed to be Lula (left) and Noah Craig  flanking a daughter, date unknown (courtesy of Sheila Craig Western)
Relatives believed to be Lula (left) and Noah Craig flanking a daughter, date unknown (courtesy of Sheila Craig Western)

Historic records point to a life of mixed heritage in the American West.

Dear Professor Gates:

My great-grandmother Lula Craig/Creg, born Jan. 26, 1870, appears on both the federal 1910 census in Depew City, Creek County, Okla., and the 1910 Indian-population census for that city and county. Lula and her children (including my grandfather Bobby) are listed as being part Creek, but her husband, Noah, is not. Lula’s death certificate and funeral records (date of death: Aug. 5, 1942), Wichita, Kan., Sedgwick County, list her mother as Precious Moore. The death certificate and funeral records for Lula list her mother’s birthplace as Muskogee, Okla., and her own birthplace as Okmulgee, Okla., which were both once Creek Indian territory. 

Can you help find out if my great-grandmother was part Creek? —Sheila Craig Western

Family lore about Native American ancestors is common in black families, but as Professor Gates has previously noted, citing research from Joanna Mountain and Kasia Bryc at 23andMe, the average African American is less than 1 percent Native American. Furthermore, if that DNA came from a single ancestor, he or she appeared as many as 10 generations (or roughly 300 years) ago. Many have no Native American ancestry at all.


The World in Which Your Great-Grandmother Lived 

However, there are exceptions involving the so-called Five Civilized Tribes—the Creek (aka Muscogee), the Choctaw, the Cherokee, the Chickasaw and the Seminole—who were removed by threat and force in the 1830s from the Southeast to Indian Territory, which included land in Kansas, the future Oklahoma and more. The dreadful Trail of Tears marches were a particularly brutal example of forced removal.

The five Native nations were known as “civilized,” in part because they owned black slaves. African Americans who descend from them or those they enslaved are more likely to have recent ancestry or affiliations from one of the nations. In his article “High Cheekbones and Straight Black Hair?” Professor Gates wrote:

[Barbara] Krauthamer (author of … Black Slaves, Indian Masters: Slavery, Emancipation, and Citizenship in the Native American South), told me, while “in the aggregate, most Native Americans and African Americans, free and enslaved, most likely had little contact over the 18th and 19th centuries,” the major exception to this involved these five tribes. “There is an abundance of evidence,” she said, “that documents ‘family’ and biological ties among African Americans and Native Americans in the Choctaw, Chickasaw, Cherokee, Creek and Seminole nations. The Choctaw, Chickasaw and Cherokees were slaveholding nations; so much that contact occurred in the context of slavery—i.e., sexual abuse of enslaved women.

“In the Creek nation, there is evidence of forced contact and childbearing but also of more consensual relationships, when Creeks incorporated runaway slave women from the southern states into their own communities and families.”


This complex history led to some confusion as to exactly who was considered Indian when the time came to enumerate the five nations’ membership. An 1890 federal census report on “The Five Civilized Tribes” (pdf) explains:

A serious difficulty was met in the answer to “Are you an Indian?” Under the laws of The Five Tribes or nations of the Indian territory a person, white in color and features, is frequently an Indian, being so by remote degree of blood or by adoption. … Negroes are frequently met who speak nothing but Indian languages, and are Indians by tribal law and custom, and others are met who call themselves Indians, who have not yet been acknowledged by the tribes.


This provides context for your great-grandmother’s documented heritage. As you mentioned, Lula appears on the Indian-population census residing with her husband, Noah “Creg,” in Depew City, Creek County, Okla., in 1910. This record states that she was one-quarter Creek (with a question mark) and three-quarters black. The Native nations of her parents are both recorded as unknown. This record indicates that her marriage to Noah was her first marriage and that they had been married 11 years. She had borne nine children, with eight still living in 1910. She and her mother were born in Oklahoma, while her father was born in Mississippi.

The following page of the record indicates that while most children of Noah and Lula Creg/Craig were born in Oklahoma, the two eldest were born in Kansas around 1894 and 1896. The census record for the family in Iowa Township, Lincoln, Oklahoma Territory, in 1900 records Lula’s birth in Texas and the fivebirths of the eldest five children in the household in Kansas. All of the records for Lula before Oklahoma becomes a state list her birthplace as Texas. Perhaps this is because she was actually born in Texas, or it could be their description of the location of her birth that was part of what was known as Indian Territory.


In 1895 Lula was 25 and living in Kansas with her husband, Noah Craig. Again, her birthplace is given as Texas, as is the birthplace of the eldest children in the household (who were likely from Noah Craig’s previous marriage, based on their ages). These records give you several family members you could also research in several locales. You may also want to investigate Lula’s parents and her siblings further to see if any of them were included on the Indian census or the Dawes Rolls, the latter of which recorded membership in the five nations between 1898 and 1906.

More Clues About Lula’s Parents 

The 1880 U.S. census includes Lula in the household of her father, Rankin Smith, in Mound Valley, Labette, Kan. The index records list the family as white, but if you view the original record, they are identified as “M” for mulatto. Her siblings included Charles Smith, born about 1868 in Texas. The 1885 Kansas state census (via; subscription required) includes these children but not the mother, and Rankin is recorded as a widower, which suggests that Lula’s mother died in Kansas between 1880 and 1885. The family was still residing in Labette County, and although Kansas death records in most counties don’t begin until 1885, you may be able to locate a burial for her in Labette County that could help you connect her to more of her relatives.


There may also be more to learn about Rankin Smith that may help you learn more about Lula’s mother. There was a Rankin Smith recorded in the 1867 Texas Voter Registration Lists. This is a good fit for your Rankin Smith because he was born in Mississippi and living in Texas, where many records for Lula claim that she was born. This record is just three years prior to Lula’s birth, and may suggest that she was born in Grimes County, Texas.

There are at least two marriage records for a Rankin Smith in Grimes County in the 1870s: one to Laura Johnson on April 6, 1871, and another to Precious Thompson on May 5, 1877. These are just indexes, but you could view digital copies of the originals at a Family History Library, or request copies from the county. It is possible that Precious Thompson is your Precious Moore and that perhaps she was married prior to her marriage to Rankin, with “Moore” being her maiden name and “Thompson” a surname by a first marriage.


Interestingly, when we searched for a Smith born about 1845 in Mississippi and living in Grimes County in 1870, we located a Clement Smith residing in Grimes County, who was born about 1844 in Mississippi, with a wife, Ida; a 2-year-old daughter named Ella; and a newborn named Ellen. If you examine the original record, just two doors away was a white man named James Rankin.

We found this record interesting because there is a marriage record in the Texas County Marriage Index for a Clem Smith and Ida Moore on Jan. 13, 1867. The death record of your Lula claimed that her mother was Precious Moore. There are many coincidences between these records and your own family. You will likely want to investigate them further to see if you can establish a connection.


Additional Records Uncover Evidence of Indian Ancestry 

Now that you have a variety of known and suspected relatives to your Lula (Smith) Craig, there are record collections specific to American Indians that you could check to see if any of these people are included. You could begin by searching for the family members you identified in the 1910 federal Indian census. The Family History Library also holds census records specific to the Creek Nation. Note any coincidences or things that stand out as you research, even if you can’t fully place them or prove a connection yet.


We noted some records that may be of interest. A Lulu Smith was recorded as a Creek in the “Arkansas Colored” in 1895, along with Andy Smith, Mellie Smith, Winnie Smith and Rose Smith.

Likewise, the Dawes Rolls index includes a Lula E. Smith roughly the right age who was a quarter Cherokee (remember, the 1910 census had a question mark next to “Creek,” opening up the possibility that her actual affiliation was with another Native nation). However, it is important to note that there are other Lula and Lucy Smiths listed. Once you have a card number, in this case Card No. 5885, check to see who else is listed on the card. We noted a Mary Thompson, which is a surname we recognized from one of Rankin Smith’s marriage records.


When working with the Dawes Rolls, do not just rely on a final-rolls search for known relatives in the Applications for Enrollment of the Commission to the Five Civilized Tribes, 1898-1914. If the family was rejected, the members would not be included on the final roll, but there may be a record of their application for recognition.

Finally, thank you for sending your DNA test results, but too little Native American ancestry was indicated for us to draw any conclusions about it.


Still, you have a wealth of information here to research about your great-grandmother. Good luck!

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

This answer was provided in consultation with Meaghan E.H. Siekman, Ph.D., a senior 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,, contains more than 1 billion 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.

Share This Story

Get our newsletter


I grew up believing the myth that my family was part Native American. My brother finally paid for and dug up family records to show we are nothing but black and white mixed going back to 1871. My grandfather was in WWI and we found his draft card showing he was a sharecropper and missing a finger even then. Not a single thing said Indian. Mind blown and we had to rethink our entire family history because it was mostly a lie.