Ethnic Origins of South Carolina Runaway Slaves

South Carolina Runaway Slaves 1730-1790

Francisco Menendez, a Mandinga runaway slave from South Carolina, became leader of the free black militia at Fort Mose in the Spanish ruled Florida of the 1700’s.


Number of runaway slaves: 2,424
African origins specified: 508


“Angola” (mostly Bakongo) 149 – 29% of African specified
“Gambia” (incl. Mali/Senegal) 61 – 12% of African specified
“Ebo” (Igbo & related ethnic groups) 49 – 10% of African specified

Above summaries are based on the HIGHLY interesting collection of runaway slave advertisements published in American newspapers in the 1700’s:Runaway Slave Advertisements : a Documentary History from the 1730’s to 1790, Volume 3, South Carolina”. This very extensive work was put together by Lathan Windley in 1983 and has been used by many historians ever since.  Besides ethnic/regional origins you can discover many more fascinating details about these advertised runaway slaves if you read the texts closely. The advertisements can be consulted online (after registration) via this great website: The African American Experience.


In the remaining part of this blogpost i will provide a full overview of the slave ethnicities to be found in the advertisements and how they relate to known slave trade statistics. Then i will zoom into the most frequently named origin of these runaway slaves to find out how “Angolan” they really were. The linguistic connection between the Gullah and the Black Seminoles who fled to Florida originally from South Carolina is also explored. On the very end i will highlight/quote some intriguing clues about African ethnicity in South Carolina as found within these short but often very informative newspaper advertisements. But first let me emphasize the following:


By no means is this data presented as fully representative of the ancestral make up of the present day population of South Carolina! It’s more complex than that 😉 As always when it comes to sample based data you will have to take into consideration how representative these advertisements might possibly be. It should be clear that only a sketchy and partial snapshot of the ethnic composition of the South Carolina slave population can be hoped for when knowing that:

  • The total number of slaves being advertised was 2,424, of whom only a minority of 508 persons had their African origins specified. The total slave population of South Carolina is estimated to have been about 20,000 in 1730 and about 107,000 in 1790 (according to wikipedia). So obviously only a very small proportion of the entire slave population is being covered.
  • In fact the advertisements also greatly underestimate the frequency of socalled “maroonage” that is the act of slaves running away, as slave owners usually didn’t bother to report such cases in newspapers at all. Especially around the time of the American Revolution it is estimated many thousands of slaves ran away in just a few years (for an excellent online overview see: Runaway Journeys).
  • The restricted timeperiod being studied (1730’s-1790’s) is leaving out the crucial formative era of the late 1600’s/early 1700’s as well as the later period of the early 1800’s when a substantial number of Africans was still arriving in the South Carolina Lowcountry.
  • The advertisements were not done completely at random: a higher proportion of skilled (and therefore more “valuable”) slaves seems to be reported; most slaves ran away from places very near Charleston; there’s a peak of advertisements right before the American revolution of 1776 and also around 1739 when the Stono Rebellion took place (Polllitzer, 1999, p.59).

Despite all these limitations it’s still very informative to at least see what kind of ethnic groups were present in South Carolina according to the advertisements. However the “country names” being used for the slaves are often hinting more to their regional origins in a wider sense than specific ethnic groups! (see also “Ethnic identities of African-born slaves: valid or imposed?“)

It’s useful in this regard to compare these South Carolina runaway slave advertisements with those from other US states and especially with those from Jamaica (see also “Ethnic Origins of Jamaican Runaway Slaves“). As can be seen from the chart below there’s a great contrast within the frequency of ethnic details being supplied. Correlating with the share of locally born slaves in the total slave populationColonies with a majority of locally born slaves, such as Barbados and Virginia, logically reporting very few specified African origins in contrast with Jamaica which is known to have had a majority of African born slaves throughout the 1700’s. South Carolina being somewhat inbetween. These locally born slaves were known as “Creoles” in Jamaica and other parts of the West Indies but in the American newspaper ads the commonly used term was “country born”. There’s also some other striking differences in terminology between the US & the West Indies which i will discuss later on.

Africa in America (p.290)

From “Africa in America: Slave Acculturation and Resistance in the American South and the British Caribbean, 1736-1831” (Mullin, 1992, p.290)


African Origins

Only a minority of the runaway slaves in South Carolina is being described as African-born which has to do with the increasing share of socalled “country born” slaves (see also “From African to Creole“). These “country born” slaves (incl. also slaves brought over from other states and the West Indies) are interesting in themselves as they reveal a great deal about the demographics of the nascent African American population, more on them in a future follow-up blog post. But let’s first start with a full overview of the African ethnic/regional origins as specified in the advertisements.


African Origins of SC runaway slaves2

Pollitzer (1999, p.60)

Comparing the ethnic/regional names being used with those appearing in Jamaican advertisements (see “Ethnic Origins of Jamaican Runaway Slaves“) we see many familiar terms, however there’s also some notable differences:

  • In South Carolina “Angola” seems to have been the preferred term for designating Central African origins, even if ethnically speaking they could have been the exact same people as referred to as “Congo” in Jamaica (more details in next section).
  • A different type of composition may have existed for Biafran slaves, a clear majority of them named “Eboe” but also 9 “Calabar” and only 1 “Moco” in South Carolina, while in Jamaica the “Moco’ label was used much more frequently in addition to “Eboe”, and “Calabar” was rarely mentioned at all. The sample size for South Carolina is rather small though. The inclusion within the Bight of Biafra region of the socalled “Pawpaw”  – named after the slave port Popo located in presentday Benin – seems to have been a mistake by Pollitzer.
  • West African origins were very often lumped together by the imprecise term of “Guinea”  in South Carolina, while Jamaican slave owners tended to be more specific. From the context of the advertisements it seems “Guinea” was used most likely for the socalled Lower Guinea region (inbetween Liberia and Nigeria, see also the Maps Section). So that might include more origins of “Grain Coast” as well as “Gold Coast” than specified.
  • Instead of “Mandingo” the “Gambia” reference was much more popular in South Carolina than in Jamaica. It is known that actually most of the captives embarking from the Gambia were collected from interior locations in eastern Senegal and western Mali.
  • Upper Guinean ethnicities like Bambara, Fula, Kissi and Temne are surprisingly often being specified by South Carolina slave owners but mostly not so by Jamaican ones. Perhaps betraying a greater familiarity with these peoples based on a greater share of slaves with these origins in South Carolina. However these same terms do appear in the Anglo-Caribbean Slave registers.

In the above chart there’s also an attempt being made to correlate the regional origin proportions for the runaway slaves with those of documented slave trade records. As Pollitzer (1999) is using outdated slave trade estimates and regional definitions which are not always compatible it might be more useful to compare with this insightful graph:


Looking back at the Top 3 most frequently mentioned runaway slave origins it seems that the socalled “Angola” and “Gambia” highest rankings are justified as they are very much in line with the prominent documented slave trade records for “West Central Africa” and “Senegambia”. It also seems to corroborate the linguistic findings i blogged about earlier (Comparing the Gullah language with other English-based Creoles.)

However it seems that the number of “Igbo” runaway slaves might be inflated when compared with the slave trade records directly from Africa. Many historians have remarked about the socalled ethnic preferences of slave buyers in South Carolina and how these contrasted whith those of Virginia (e.g. Phillip Morgan in  “Slave Counterpoint”). Especially for the Bight of Biafra there was a clear difference, slaves from this area being predominant in Virginia (~40%) but in South Carolina the socalled “Eboes” had a reputation of being prone to resist enslavement by either committing suicide or fleeing (the folktale of the Igbo landing in neighbouring Georgia might be a remarkable testimony). These runaway advertisements seem to be in support of that assessment as the Igbo appear to have run away disproportionately to their numbers.

However we should not forget about the substantial intercolonial slave trade taking place between the West Indies and South Carolina and to some degree also with Virginia and other northern states. For example the also quite numerous mentioning of the “Coromantee” is more easily understood if we take into account the African born slaves, incl. many from the Gold Coast/Ghana, being imported via Jamaica, Barbados and other parts of the West Indies. Likewise it could be that more Igbo’s ended up in South Carolina via the West Indies or even via Virginia than by direct slavevoyages from Africa. Either way it seems that the direct slave trade between South Carolina and the Bight of Biafra peaked inbetween 1731-1760, and this timeperiod is covered extensively by these runaway advertisements from inbetween 1730-1790.

Given the widely publicized Sierra Leone connections of the Gullah it may seem surprising that any apparent Sierra Leone origins are not mentioned that often in the advertisements. (In fact in the chart featured in Pollitzer (1999) the “Bambara” should be excluded from the “Sierra Leone” region as this slave ethnonym is commonly thought to refer not only to the presentday ethnic group called Bambara living in Mali but more generally so also to any interior, upriver ethnic group from Senegambia.) A likely explanation for this outcome can be found in the slave trade numbers depicted above and their distribution according to timeperiod. For Sierra Leone the time period 1791-1820 is clearly the period when most slave trade occurred with South Carolina. As the runaway advertisements under study are limited to the period of 1730’s-1790’s the subdued Sierra Leone presence makes more sense. For Senegambia it’s instead 1761-1790 which saw the greatest intensity of slave trade with South Carolina and during 1700-1730 it was clearly the number one supplier of slaves. While for West Central Africa it’s also 1791-1820 when slave trade peaked however the earlier period of 1731-1760 also saw many Central Africans entering South Carolina and they were clearly predominant at that time.


‘Looking for Angola’, but what about Congo?


The northern Europeans knew the entire coastline of Central Africa south of Cape Lopez as “Angola,” and they designated slaves they purchased there as “Angolas,” employing the term in a sense entirely different from, and considerably less distinct than, Portuguese and Brazilian uses of it. (Miller, 2002, p.28)

However, the “Angolas”reaching the Caribbean and North America aboard the ships of the French, Dutch, and English from 1670s onward had begun their Middle Passages at any of the bays north of the Zaire – Mayumba nearest Cape Lopez, then Loango, Malimbo, Cabinda, and the “Congo” River (as the Zaire was known) mouth itself.(Miller, 2002, p.29)


“Looking for Angola” is the name of a very intriguing archeology projectaimed at discovering the location “Angola,” a maroon community that thrived on Florida’s southwest coast from 1812-1821.” Many of these maroons, also known as Black Seminoles nowadays, had originally fled from South Carolina/Georgia, and the naming of their settlement “Angola” seems to have been done in honour of the “Nation” they were thought to hail from. For more details see also this website and this paper. More discussion on the Black Seminoles will follow later on. But first i will try to provide more clarity on the socalled “Angolans” in South Carolina.

As already touched upon earlier there seems to have been a different usage of terms to designate similar Central African slave origins in South Carolina when compared to Jamaica. It’s striking that the more commonly used term (in the West Indies) “Congo” is only used 3 times while “Angola” is mentioned no less than 149 times in the South Carolina advertisements. Actually also in other contemporary documentation from South Carolina the term “Angola” seems to have been clearly preferred in the 1700’s. As mentioned in the quotes above it’s crucial to realize that the modernday country of Angola did not exist in the 1700’s and varying geographical interpretations were in use among the European colonial powers!

The below overview, taken from the Slavevoyages Database, makes it very clear that the slave ports being used by English/American slave traders were almost all located to the north of the Congo (a.k.a. Zaire) river. Only the presentday Angolan port of Ambriz seems to have been an exception but there’s no recorded slavevoyages from that place to South Carolina and it seems to have been minor overall and also located in Bakongo territory (to the north of Luanda) either way.





It seems the socalled “Angolan” runaway slaves from South Carolina were therefore actually mostly being shipped over from the socalled Loango kingdom, located along the Congolese coastline (for more details see also the West-Central Africa page in the Maps section). Instead of being embarked from the main Portuguese settlements in Angola located much more to the south: Luanda and Benguela. The third main slave port of Angola, Cabinda, was indeed frequently visited by English and other North European traders (aside from Portuguese as well). However Cabinda is actually not within Angola’s main boundaries but rather an enclave (or pene-exclave more properly 🙂 ) on the north side of the Congo river.

Making this finer distinction is very important if you want explore the ethnic origins hiding behind this label of “Angolan” slaves. Regrettably a commonly made mistake (also by historians!) is to just take this contextspecific terminology of the 1600’s/1700’s at face value and assume generic Angolan origins right away. In many charts detailing the African origins of South Carolina you will therefore often see Angola appearing and not Congo. Both Angolan and Congolese are mere post-colonial nationalities of course and both multi-ethnic countries show ethnic overlap in their border provinces. Especially the Bakongo are numerous both in northern Angola and western Congo (DRC & Brazzaville).

But the two main ethnic groups from Angola, the Ambundu and Ovimbundu, are only found in Angola, and given the slave trade patterns described above it seems that ancestral connections with these Mbundu people are much less likely for South Carolina and African Americans in general than for Brazilians for example. Instead it seems that the Bakongo are being referred to mostly (but not exclusively) by both the term “Angola” as used in South Carolina and “Congo” as used in the West Indies.  This also goes for the Africans slaves involved in the famous Stono Rebellion which took place in South Carolina in 1739.


“More recently John Thornton revisited the Stono slave rebellion that Wood so ably analyzed, and he concluded that despite contemporary accounts that described the uprisen slaves as Angolan, the rebels were more probably from Kongo. He argues that actions which to their pursuers (and chroniclers) seemed nonsensical – such as dancing, music making, and pauses along the road – actually relate to the maroons’ Kongolese military practice.” (Landers, 2002, p.228)


It is known from recent research that the very first Africans to arrive in Virginia were indeed from Angola proper. (see e.g. Virginia’s First Africans).The Portuguese slaveship carrying them from Luanda had been captured by Dutch pirates and subsequently they were sold in the USA instead of its intended destination in Mexico. Also in their own colony of “Nieuw Amsterdam” there were many Angolans present, as Dutch pirates continiously attacked Portuguese shipping in the early 1600’s and they even occupied Luanda for a few years in the 1640’s. For a brief period the Dutch were becoming the most important slave traders as they took over former Portuguese colonial possessions. And it seems very possible that more Angolan slaves were transported by them to the Thirteen Colonies in the first half of the 1600’s. This might have set in motion a historical precedent whereby the name “Angola” caught on and stuck also in the late 1600’s and 1700’s when the Central African slaves arriving in the US and especially South Carolina (transported by English/Americans ships this time) were no longer actually coming from Luanda or other southern ports in Angola but rather from the Kongolese coastline.


The Black Seminole-Gullah connection: a shared Bakongo legacy?


“Runaway slave advertisements and notices in South Carolina indicate that during the 1730’s, approximately 57 percent of the runaways came from the Kongo-Angola region. During the 1740’s, the percentage of Kongo-Angola runaways increased to 61 percent. Between 1735 and 1765, Kongo natives made up the majority of the former slaves listed in the Saint Augustine [Florida] records of black marriages.” (Dixon, 2007, pp.17-18)

 “In fact, the Stono rebels were headed for Spanish St. Augustine to avail themselves of the well-known religious sanctuary available in that colony – a sanctuary their  predecessors had secured in 1693. After that date the Spanish government in Florida freed runaways from the British colonies, granted them homesteads, and permitted them to elect their own political and military leaders in return for their religious conversion, military service, and fidelity. “(Landers, 2002, p.228)

 As archaeologists like Leland Ferguson have demonstrated, enslaved and escaped Africans spread Kongo traditions southward down the Gullah Coast from South Carolina to Florida.15 Some of these traditions survive to this day in the famous cemetery at Sunsbury, Georgia and in the Bosque Bello cemetery on Amelia Island, Florida where black graves are still decorated with white ceramic chickens and white conch shells–Kongo cultural markers described by Robert Farris Thompson and others.” […]

Kongo traditions also found their way into interior Florida as Central African slaves escaped bondage and became vassals of the Seminole nation.” (Landers, 2002, p.231)


The above quotes should make it sufficiently clear that the predominance of Kongolese runaway slaves in the South Carolina Lowcountry was continued in Spanish ruled Florida where many found refuge and some of them eventually formed the distinct Black Seminole group. It’s striking also that in the Spanish documentation the very same Central African slaves from South Carolina were mostly named  “Congo” rather than “Angola’s”, perhaps because the Spanish were more in tune with the ethnic composition of Angolans/Congolese. But perhaps also because the Spanish were more likely to document the way these runaway slaves would selfidentify themselves:


Charleston South Carolina Gazette and Country Journal, May 16, 1769.

from the subscriber’s plantation on Savannahriver, near Purrysburgh, about six weeks ago, a likely young new negro fellow, about five feet eight inches high, and about twenty five years of age, had on when he went away a white negro cloth jacket, breeches and cap, his jacket buttons are only notted cloth of the same, his breeches had no kneebands, he has a ringhole in each ear, calls himself POMPEY, speaks little or no English, calls his country name CONGO.”


A very fascinating and revealing connection can be made between the presentday Gullah population of the South Carolina Low Country and their “cousins” the Black Seminoles who originally resided in Florida where they found refuge fleeing from South Carolina/Georgia. Because they descend mostly from South Carolina runaway slaves in the 1700’s the Black Seminoles might have preserved cultural Gullah features which are more in line with African origins predominant in the 1700’s while later additions from the early 1800’s could be mostly missing for them.

Nowadays the Black Seminoles also live in Oklahoma, Texas, Mexico and the Bahamas as a result of their ever continuing pursuit of freedom. It’s most regrettable that their language, called “Afro-Seminole Creole” by linguists, is under threat of becoming extinct. Despite being distinct in its own right it’s considered to be very closely related to the Gullah language as it would have been spoken in the 1700’s. For more details see these two insightful online articles:

Below quotes clarify the fascinating links between the Gullah and the Black Seminole:


“The Afro-Seminole Creole language is an English-related Creole. It is a descendant or derivative of the Gullah language. Due primarily to the isolation of the Black Seminoles, the Afro-Seminole Creole language (ASC, hereafter) has, according to linguist Ian Hancock, “preserved far more of its original character than has Gullah.” […]

“The ASC is considered to be “almost identical to the conservative Gullah of a century ago . . . but it does not have the non-English sounds which Gullah has.” The African influence on the Gullah language is reflected in the incorporation of Sierra Leone Krio and Mende terms. However, the Mende did not have a strong presence in Sierra Leone until after 1800, therefore, Mende words are not found in the ASC. The ASC language formed prior to 1800  (Dixon, 2007, p.23)

“ASC [Afro Seminole Creole] lacks much of the Mende and other African-derived lexicon found in SIC [Sea Island Creole, a.k.a. Gullah]. Nevertheless, ASC [Afro Seminole Creole] contains about forty words of African provenance,  some half of which are traceable to KiKongo/KiMbundu” (From “Further Studies in the Lesser-Known Varieties of English” (Hancock, 2015, p.259)

“Known African cultural retentions, particularly in the area of religious expression, seem to confirm Rebecca Bateman’s supposition that the “Black Seminole naming system represents strong evidence for the existence of a naming system which bears striking similarities to the Kongo-Angola naming system.” The act of naming a child was considered a religious matter that warranted a ritual ceremony in many West African cultures. Historical evidence clearly suggests a strong Kongo-Angola influence in the Black Seminole culture, especially in the practice of naming.” (Dixon, 2007, p.22)



Bantu Place names in SC

Pollitzer (1999, p.123)


As a closing thought about the Congolese/Angolan, and more specifically Bakongo legacy in South Carolina, we should realize this heritage is not only measured in cultural/linguistic terms (e.g. the Kongo lexical predominance in the Gullah language: “Comparing the Gullah language with other English-based Creoles.“). Or the fascinating place names of Bantu origin as shown above. It’s even more profoundly in bloodlines and genetics that we might find substantial traces of Central African ancestry in South Carolina. Future DNA testing will undoubtedly uncover more of these inherited connections.


“West-Central Africans not only constructed the plantation complex; they also sowed the seeds for the uninterrupted growth of the African-descended population. Following the Stono Rebellion and the outbreak of King George’s War, both in 1739, South Carolina’s African trade foundered under the weight of prohibitive duties on importation of enslaved people and decreased trade generally. People from West-Central Africa thus constituted the last large influx of Africans for another decade. This brief respite coincided with the stabilization of self-reproducing communities by the end of the 1740s”  (Brown, 2002, p.300)

 “A glimpse of the importance of West-Central African men in this regard may be gleaned in the retention of the Bantu word tata in Lowcountry speech. Indeed, tata was the only African term for “father” maintained throughout the Lowcountry into the twentieth century” (Brown, 2002, p.302)

” As such, the perceptions and preferences of Carolina planters affected the much larger realm of Atlantic commerce far less than some scholars have supposed. While the wishes of rice growers and sellers certainly encouraged the importation of as many people from Senegambia, Sierra Leone, and the Windward Coast as possible” (Brown, 2002, p.303)

“Throughout the duration of the trade in enslaved Africans, people from West-Central Africa arrived in such large numbers that they formed one of the largest African communities in the Lowcountry. They lived in every district of the region since the first captive Africans landed in Charleston. They also represented the last victims of the Atlantic trade. From the first to last, West-Central Africans and their descendents inhabited the Lowcountry and made it their own land.” (Brown, 2002, p.305)


Clues of African Identity

Even though the advertisements are generally brief and to the point they sometimes contain surprising and valuable details about the runaway slaves, despite the obvious eurocentric bias. Below i will just quote some of the ones that seem to indicate retention of African identity in South Carolina. There are however many other interesting aspects to be discovered when browsing the advertisements, especially about family members helping to hide runaway slaves, the presence of freed blacks, racially mixed slaves etc. In a future blogpost i will follow up on the socalled “countryborn” or American origins of these South Carolina runaway slaves. But first let’s continue exploring their African origins!

 African “country names”

It’s interesting how often also the socalled “country names”, the original African names of the slaves, are being mentioned by the slave owners. Apparently the new names they had in mind for their slaves were not being maintained as they might have wished… It might be that the information about these original names was mostly obtained from fellow slaves btw, interrogated after the runaway slave had left already. Also noteworthy is how even locally born or racially mixed slaves would sometimes be known by their African names, that is second or perhaps even third generation South Carolina slaves, showcasing that this type of African retention was sometimes carried across the generations. Below examples just for illustration, there’s many more to be found.


Charleston SouthCarolina and American General Gazette, February 21, 1781.

“Two Guineas Reward.
RUN away a young negro fellow, named Quamina, well known in and about Charlestown by his impudent behaviour; he has told me to my face, “he can go when he pleases, and I can do nothing to him, nor shall I ever get a copper for him


Charleston SouthCarolina Gazette and Country Journal, May 7, 1771.

from the subscriber’s plantation at Edisto, on the 4th of April last, a Mustee fellow named CUDJO, near 40 years of age, […]”


Charleston SouthCarolina and American General Gazette, November 11 to November 18, 1774.

RUN away or stolen from the subscribers, the following NEW NEGRO MEN, viz. SHELALY a tall, well made fellow, of the Kissey country, whose country name is Bobodandy, speaks little or no English, but can tell his own name if he will,”


State Gazette of South Carolina (Timothy), July 31, 1786

“Five Guineas Reward. RUNAWAY From the Subscriber on John’s Island, about six weeks ago, A NEGRO FELLOW named SAMBO and a Wench FATIMA. Sambo is about 5 feet 9 inches high, very surly countenance, speaks bad English, and is of the Guinea country, but has been many years in this State. Fatima is about the same height, country born, a likely wench, but very artful.”


South Carolina Gazette (Timothy), August 27, 1753

“STROLLED away from my house, about midnight between thursday [sic] the 16th and friday the 17th instant, a tall well made new negro man (of the Mindinga country) of a yellowish complexion, can speak no English, […] His name is John, but he will more readily answer to the name of FOOTABEA, which he went by in his own country.”


SouthCarolina Gazette (Timothy), April 11 to April 18, 1748.

“RUNAWAY on the 17th Instant, and supposed to be gone in a Canow to the Southward, two Angola Negro Men, one named Moses, middle siz’d, his Country Name Monvigo, had on a Linnen Jacket, broad Cloth Breeches with silver’d thread Buttons,and a woollen mill’d Cap; the other a tall, lusty and elderly Fellow, named Sampson*, his Country Name Goma”


South Carolina Gazette (Timothy), October 23, 1751.

“Run away from the Subscriber, living on the Northside of Black River, a Gambia new negro fellow bought of Messrs Austin and Laurens in July last, answers to the name of Manso, is very black and middle sized, mark’d on the forehead with a cross, and three perpendicular strokes on each cheek”


South Carolina Gazette (Timothy), July 17 to July 24, 1755.

“RUN AWAY in January last, a middle sized, middle aged negro fellow, named London, but his country name is Appee: He came from St. Croix last year, speaks broken English,”


South Carolina Gazette Extraordinary (Powell & Co.), August 10, 1772.

From the Subscriber, in St. Matthew’s Parish, on the 19th Inst. Two new NEGRO FELLOWS, OF the Gold Coast Country, bought the 6th Instant, at Mr. GERVAIS’s Sale, neither of them can speak any English. One, a likely made Fellow, supposed to be about five Feet seven Inches high, appears about Twentyfive Years of Age, very black; his Country Name FODEE. The other, a slim made Fellow, about five Feet four Inches high, supposed to be about Twentyfour Years of Age, very black, his Country Name MAMBEE”


Charleston SouthCarolina and American General Gazette, July 15 to July 22, 1771.

“RUN away from the subscriber on Hilton Head in May last, a thin spare Negro Boy named DARBY, formerly called Quackoe, of the Guinea country,


Charleston South Carolina and American General Gazette, May 14, 1778.

“RUN away on the first day of May instant, a Negroe boy named Charlestown, his country name Tambo;


Cultural clues

Unlike for the Jamaican runaway slave advertisements i wasn’t able to find any clear references to African forms of music making (banjaw, gumba) or religion (Obeah). However below advertisements do seem to point towards an appreciation or at least recognition of African talent in these domains to some degree.


South Carolina Gazette (Whitmarsh), May 19 to May 26, 1733.

“Run away, from Mr. Alex. Vanderdussen’s Plantation at GooseCreek, a Negro Man named Thomas Butler, the famous Pushing and Dancing Master.”


South Carolina Gazette (Timothy), April 8, 1745.

“Run away from the Pelham Privateer, two Negro Men, viz.
Joseph Johnson, born in Bermuda, speaks very good English, had on when he went away a light colour’d Cloth Coat, a speckled Shirt, Oznabrug Frock and Trowsers. Thomas Ebsery, born in Jamaica, a tall slim Fellow, hard of Hearing, he beats a Drum very well and is well known amongst the Negroes in Charles Town, having been here before with Capt. Thrasher.”


South Carolina Gazette (Timothy), November 19, 1750.

“Run away from the Subscriber, at Bacon’s Bridge, a negro fellow named Newman, pretends to be a sort of Doctor,”


South Carolina Gazette (Timothy), May 1 to May 8, 1749.

“RUN AWAY from the Subscriber’s plantation at Stono, about a month ago, a negro fellow named Jack, lately belonging to Mr. Robert Thorpe (and was his driver) and formerly to Mr. Pendarvis deceased; he is a middle aged tall negro, very sensible, and speaks good English, is very expert in hunting, and is pretty well stock’d with cloaths.”


Charleston South Carolina Gazette and General Advertiser, January 15 to January 17,

“RAN AWAY A NEGRO MAN named DAVY, by trade a Carpenter, formerly the property of Mr. Hart, the chairmaker, and lived some time with Mr. Cooke the Carpenter. He is a stout fellow, about 5 feet 10 inches high, 35 or 40 years of age, stoops a good deal, wears a long beard, and is well known in Charlestown as a fiddler.”


Gazette of the State of South Carolina (Timothy & Boden), March 31, 1779.

“RUN away from the Subscriber, a negro Fellow named Quash, formerly a drummer, and well known in most parts of St. Bartholomew’s; he is about 5 feet 10 inches high, spare and well made, very likely and sensible; he carried off with him a drummer’s suit of clothes which he formerly wore when a drummer to the Borough company, with a green coatee and sundry other half worn clothes, as also a gun and a portmantua, containing a few books, papers, clothes, &c.”


 Ethnic markers/identity


Charleston South Carolina and American General Gazette, May 27 to June 3, 1774.

“RUN away the 21st of April last, TWO NEGRO MEN that have been in the country two years, and speak English so as to be understood, viz. Bucharah, a slim made fellow about five feet six or seven inches high, has one remarkable small thigh and leg, occasioned by an old wound; is about thirtythree years of age, and very black; Cupid, his country name Yonge, a small, low, set fellow about five feet three inches high, has an odd cast with his eyes; about twentyseven years of age; is branded on the left arm, above the elbow, BI, but very dull. They are both of a country which they call Sofo, and both speak the Bombra language and Cupid the Fulla language;”


Charleston South Carolina and American General Gazette, January 2, 1777.

“TEN POUNDS reward to any person who will apprehend Abram, a black fellow of the Moco country, having his country marks, particularly very long streaks on his arms, near six feet high, about 30 years old, had on when he went away a white negro cloth jacket and breeches in April last.”


Charleston South Carolina Gazette and General Advertiser, July 12, 1783.

“RUN away, a negro Boy named OTHELLO, about 5 feet high: had on an oznabrig jacket and trowsers, is very lusty, and his head bigger in proportion to his body than ordinary; also a negro Girl, answers to the name of MARIA, about 4 and a half feet high, both of the Gola country, and speaks very bad English”


Charleston Royal Gazette, January 9 to January 12, 1782.

“RUN away from the subscriber, on Christmas day last, a NEGRO FELLOW, named SAMBO, or SAM; he is about five feet high, slim made, and has a small limp or ketch in his walk; he is of a yellowish complexion, and is a little pitted with the smallpox; his upper foreteeth comes a good deal over his lower ones, and his hair is pretty long, being of the Fulla country,”


South Carolina Gazette (Timothy), January 26 to February 2, 1740.

“Run away from James Akin a Negro Fellow, lately belonging to Capt. James Wathen, named Basey, is a short well set Fellow of a tawny Complexion, speaks broken English, and is either an Eboe or Calebar Negro.”


Charleston South Carolina and American General Gazette, August 13 to August 20,1773.

“RUN away from the subscriber in St. Matthew’s Parish, SIX NEW NEGRO MEN; one of them went off in February last, viz. Dibbee, a well set fellow, about five feet six inches high, of the Timine country, his ears bored and his teeth filed, seems about twentyfive years of age and very black.”


Charleston South Carolina Gazette and Country Journal, December 5, 1769.

“RUNAWAY, or STOLEN, from the subscriber, in St. Stephen’s parish, on the 28th of October last, a short, likely, yellowish NEGRO WOMAN, about 26 years of age, named BECKEY, of the Ebo country, but looks more like a country born”


 South Carolina Gazette (Timothy), August 6 to August 13, 1737.

“RUN AWAY from the Plantation of Isaac Porcher on Wassamsaw, a new Angola Negro Man, named Clawss, […]
N.B. As there is abundance of Negroes in this Province of that Nation, he may chance to be harbour’d among some of them,”



– Brown, R.M. (2002). “Walk in the Feenda”: West-Central Africans and the Forest in the South Carolina–Georgia Lowcountry, in Linda M. Heywood, ed. Central Africans and Cultural Transformations in the American Diaspora, 289-318.
– Dixon, A.E. (2007). “Black Seminole involvement and leadership during the Second Seminole War, 1835–1842“, thesis. (link to online PDF)
– Landers, J. (2002). The Central African Presence in Spanish Maroon Communities, in Linda M. Heywood, ed. Central Africans and Cultural Transformations in the American Diaspora, 227-242.
Miller, J.C. (2002). Central Africa During the Era of the Slave Trade, c. 1490s–1850s, in Linda M. Heywood, ed. Central Africans and Cultural Transformations in the American Diaspora, 21-70.
Pollitzer, W.S. (1999). The Gullah People and Their African Heritage.

22 thoughts on “Ethnic Origins of South Carolina Runaway Slaves

  1. I love reading your blogs. They are not only highly informative but also accurate and unbiased. I’m most appreciative of your hard work. THANK YOU!

    Liked by 1 person

  2. You’re very welcome!. I truly appreciate your continued feedback on my blog! I always make a conscious attempt to be as unbiased as possible. However some bias might always be there even unintentionally. That’s why I also place all the main research findings featured in my blog posts separately in “Tables on Ethnic/Regional Origins” in the menubar. That way readers can also access the information without having to bother with my personal opinions or lenghty analysis 😉


  3. “Just across the border in North Carolina is Angola Swamp”. Someone brought this to my attention. Very fascinating, i suppose it might have been a place of refuge or sanctuary for runaway slaves just like the “Angola” settlement was for Black Seminoles in Florida. Couldn’t find any specifics on the net, except for this link:

    Which says: “the North Carolina swamps have seen military activity during the American Revolution and the Civil War; been havens for escaped slaves (some of whom gathered and formed colonies and redoubts, while others traveled in secrecy from swamp to swamp);”


    • Another report on how archaeologists are uncovering more details how escaped slaves managed to retain their freedom by moving into inhospitable swamps. This article is about a swamp located in the border area between southeast Virginia and northeast North Carolina. Some interesting quotes:

      “By downplaying American marronage, and valorizing white involvement in the Underground Railroad, historians have shown a racial bias, in Sayers’ opinion, a reluctance to acknowledge the strength of black resistance and initiative. They’ve also revealed the shortcomings of their methods: “Historians are limited to source documents. When it comes to maroons, there isn’t that much on paper. But that doesn’t mean their story should be ignored or overlooked. As archaeologists, we can read it in the ground.”

      “The largest community of American maroons was in the Great Dismal Swamp, but there were others in the swamps outside New Orleans, in Alabama and elsewhere in the Carolinas, and in Florida. All these sites are being investigated by archaeologists.

      “The other maroon societies had more fluidity,” says Bercaw. “People would slip off down the waterways, but usually maintain some contact. The Dismal Swamp maroons found a way to remove themselves completely from the United States, in the recesses of its geography.”

      “What became of the Dismal Swamp maroons? Olmsted thought that very few were left by the 1850s, but he stayed near the canals and didn’t venture into the interior. Sayers has evidence of a thriving community at the nameless site all the way up to the Civil War. “That’s when they came out,” he says. “We’ve found almost nothing after the Civil War. They probably worked themselves back into society as free people.”

      Early in his research, he started interviewing African-Americans in communities near the swamp, hoping to hear family stories about maroons. But he abandoned the side project”

      Would be awesome to discover any lingering oral history about these heroic people!

      Deep in the Swamps, Archaeologists Are Finding How Fugitive Slaves Kept Their Freedom


  4. Hi, I read your blog lately and I want to thank you for the job you’ve done. I’m of Congolese origin and I knew some slaves were taken from Congo but I didn’t know the number was so huge. About African Americans, I read somewhere most Angolan slaves were likely shipped to Brazil, Argentina and other countries in South America while Congolese slaves were shipped in USA, Cuba and Haiti. The fact that the largest port where slaves were taken is Cabinda intrigues me because it doesn’t even touch Angola and I think those Bakongo slaves are more likely to be Congoleses and not Angolans.

    Do you think it is possible to know what proportion of the slaves came from Congo (any of the two) or Angola?

    Liked by 1 person

    • Hi Geoffrey, thanks for your comment and glad you like the blog! Good question! I’ve asked myself the very same thing as well actually. It’s a complex matter because you’re dealing with modernday nationalities and ethnic groups which are often closely related. I intend to blog about it in more detail in the future. However just quickly these are my thoughts:

      1) Angolan & Congolese borders are manmade and ethnic groups can sometimes be found on both sides, especially the Bakongo, see also the Central African Map Section
      2) Slaveports had trading connections which reached deep into the interior, crossing modernday borders. Interior Congolese slaves could also have been transported via Angolan ports in this way. See for example this map or this one.
      3) Still like you said there’s a pattern of (southern) Angolan ports like Luanda and Benguela being used almost exclusively for slave exports to Brazil & Hispanic America while slaveports near the river Congo, such as Cabinda, were much more frequently used by French, Dutch and English traders.
      4) I think the current state of DNA analysis (autosomal) is not able yet to make any firm distinction between Angolan and Congolese ancestry, because afterall they are nationalities first and even different ethnic groups unique to either Angola (like the Mbundu) or Congo (like the Luba) might show considerable genetic similarity.
      5) However i suspect in the near future when largescale & representative sampling has been done in both Angola and Congo socalled IBD segment matching will provide more clarity.


  5. After reading your article, I am wondering what your thought are concerning Dr. Kittles of African Ancestry who stated that Angola introduced “non-African DNA into SC”?


  6. Hi Raymond, i would have to read the full text of that quotation to be able to put it in its proper context. Frankly i’m not familiar with such a statement by Dr. Kittles. But at any rate this blog is focussed on Tracing African Roots, so non-African DNA, even when interesting in itself, is not really a topic of discussion.

    Just to hark back on what i’ve said in this blogpost: according to up-to-date research the Central African captives brought to the US were usually disembarking from slaveports near the Congo river, technically this includes parts of northern Angola, Cabinda as well as both Congo’s. This area is dominated by Bakongo speakers who live across borders. And it might be assumed they were heavily represented among the people labeled “Angolans” in 18th century South Carolina even when elsewhere these very same people might have been identified as “Congo’s”.

    Slaveports located further south in Angola, like Luanda and Benguela, were controlled by the Portuguese and rarely visited by the English or other northern European slave traders. The captives leaving from these ports would have diverse ethnic backgrounds just like the ones from near the Congo river, but it can be expected that the mix would be different despite some overlap. Especially the Mbundu people, who only live in Angola, would be predominant. It’s probably safe to state that for African Americans as a group Bakongo origins are (far) more important than Mbundu origins.


  7. Sorry if im late. but I have been looking for someone who could help me find my possible african lineages,and its very hard to find those people because of me being Afro-american. You seem very well knowledgable and trustworthy . And everyone i ask just basically have more negative responses like “all these results are just fake”or “there no where close to being accurate its just something “fun” to do” or “you’re asking to much out of this” im left with nothing..I know you cant take things TOO LITERALLY, but i should be able to take SOMETHING right? possibilities?? Basically its like everyone is just telling me to “give up ,you have no bases ,and dna test are all bull and just random guesses” .My Mother Side is not from South Carolina but North Carolina instead. and i haven’t found much information on african origins of NC. Same for my Fathers side, as far as we know everyone is from Maryland yet i research and research and find very little of the possible origins of blacks in MD. I ran into your african breakdown forum on ancestry, but by the time i took my dna test the forum had closed.
    So if you don’t mind, i’d like to post my results for your insight. and i’ll post the rest of my unafrican dna just for the sake of it.

    Africa 83%

    Benin/togo 31%
    Cameroon/congo 15%
    Ivorycoast/Ghana 10%
    Nigeria 9%
    Mali 6%

    Trace Regions all together 28%
    Senegal 5%
    Africa Southeastern Bantu 4%
    Africa South-Central Hunter-Gatherers 2%
    North africa 1%

    European trace regions
    Scandinavia 6%
    Greta britian 4%
    Italy/greece 2%
    Ireland 2%
    Finland/Northwest Russia 1%
    Europe West 1%

    West Asia trace region

    Middle East 1%

    Liked by 1 person

    • Hi Vanessa, thank you for your message and congratulations on your results! They are certainly not worthless or “just for fun”! I find that people who are overly dismissive of DNA testing tend to have ulterior motives beyond any basic grasp of how DNA testing actually works… IMO it all boils down to correct interpretation and not having unrealistic expectations. A very quick way to find out how these AncestryDNA results can indeed be predictive (but within given restrictions!) is by looking at the DNA results of actual Africans who already know their background:

      African AncestryDNA results

      Also my own survey into the African origins of Afro-Diasporans as reported by AncestryDNA is largely in support of these DNA results being quite reliable as they are broadly in line with historical information and cultural retention:

      Diaspora Comparison

      As i have mentioned throughout my blog AncestryDNA’s socalled “Ethnicty Estimates” can provide very valuable insight indeed but only within a (sketchy) regional framework. You will need additional context/info to pinpoint any specific ethnic details or also combine with other DNA results, especially any African matches you might have.

      So my first advise to you would be to regularly do a search among your DNA matches. You can either carefully browse through each of your pages and look for any matches which seem “African” because of their profile names or the preview of their ethnicity estmates being 100% African. Another way is to search for birth location and type in countries like Nigeria, Ghana, Benin, Cameroon, Liberia, Democratic Republic of Congo, Sierra Leone, Senegal etc. etc. The number of Africans who are taking AncestryDNA tests is small but steadily growing. So the chances of eventually finding an African match will increase with time. Keep in mind though that finding an African match doesn’t per se mean you or your MRCA (most recent common ancestor) share the same ethnic origin as your match. Without any paper trail there will always be several ancestral scenarios which may be valid.

      In fact as your results already demonstrate you are firstmost a fusion of several ethnic/regional origins within Africa. (and also non-African ones). As is the case for practically all African Americans. Who are all individually descended from most likely dozens if not hundreds of African born ancestors who were forcibly relocated to the US. Which makes perfect sense given the history of African Americans as a population group within its own right. What is striking especially is the great deal of mobility as well as blending of various African ethnic groups which must have taken place across the generations within the USA. This tends to complicate getting a clearer picture however it is also fundamental part of the African American experience. For more details see also:

      African American AncestryDNA results
      Fictional Family Tree incl. African Born Ancestors

      Having said all that i think your biggest regions will probably be most informative for you to find out about your predominant African lineages. Although in fact also the smaller regions may be providing valuable clues. About your 31% socalled “Benin/Togo” it is important to realize that you may have inherited this sizeable chunk of your DNA from most likely more than a dozen individual African born ancestors who had offspring in the USA. Unless you happen to have an African grand parent or greatgrandparent ofcourse. Otherwise it’s not like this is to be traced back per se to just one single ancestral location or just one single ethnic group. As in fact there may have been several ones combined in that 31% score. Although in all likelyhood still closely interrelated and geographically nearby.

      Secondly it also crucial to be aware that infact also southern Nigerians (Yoruba but also Edo’s, Urhobo’s, Igbo’s etc.) tend to get substantial Benin/Togo scores when they test with AncestryDNA. And this also goes for the Ewe people from Ghana, who also live in bordering Togo. So don’t take the country name labeling too literally. There is always going to be overlap. It can still be informative though for getting a rough idea if you also include neighbouring countries (as also advised by itself). For more info see also:

      Is “Benin/Togo” really pinpointing origins from within Benin’s borders?
      Nigerian AncestryDNA results

      The socalled “Cameroon/Congo” region is again ambigious because it may suggest a wide array of possible ethnic origins. This again also includes Nigeria in fact, but more so restricted to southeastern Nigeria (Igbo’s, Ijaw, Efik etc.). Otherwise based on slave trade statistics i’m assuming it will firstmost signal Central African origins and in many cases from the Bakongo people specifically. As i mentioned already you will need to have additional research findings and/or DNA results though to zoom in to such a level.


      • Thank you FonteFelipe!
        I would have to agree on the overly dismissive people when it comes to DNA testing. besides i didn’t spend 100$ just for “fun”. It was the same way with using GEDMATCH calculators, it took me a while before i could understand the basics of how the calculators work. And why some of the calculators gave me ridiculous results. Once i did a little research i figured out why one calculator kept matching me with North africans. It
        was because i was using a calculator made for full blooded Africans, and it seemed like my European Dna was being mistaken for North African or some sort of Fulani Dna.
        i literally watched one all African calculator take my 16% euro DNA and turned it into 16% North African, so i stopped using them. I noticed the same Fulani like/North African results for ALOT of AA’s and i knew something wasn’t right. I think Alot of AA’s actually believed it though sadly and didn’t even catch that this was a calculator for pure bloods, and that so called fulani-morroccan-egyptain-Alegrian match is getting mixed up with your European Gene’s, But i think it most likely being accepted because we want to be abit more “Exotic” than just a generic simple west african, Even if it makes no sense History wise. Its so easy to misinterpret these things. Do you have any advise on how GEDmatch could be useful DNA wise?

        And thanks for the Benin Togo comment. That had been puzzling me for the longest. i was like “why is my benin togo score so high??” i actually was expecting a high nigerian score since most AA’s have that. but mine was only 9%, so it threw me off. And based off my family tree research i have done so far, there seems to be not recent african immigrations on either side of my family. Plus added , the 52% of the rest of my african dna got divided into a bunch of smaller percentages all over the place. To be honest, i think i was bit dissapointed at first with my results, The biggest dissapointment was seeing almost 30% of dna turned into trace regions while being told “dont pay attention to trace regions they”re most likely just noise” and im saying”soooooo..basically 30% of my dna is just NOISE?….i get if its just like 2% but almost 30???” i thought something went wrong with the testing. I felt like they needed to test me again.

        The Cameroon Congo region i noticed had alot of neighboring countries included Nigeria, Chad, equitorial Guiniea ,Gabon, DRC,republic of congo, Central african republic, and Angola. I though “wow way to round up Everyone into one category….talk about over simplification, this i going to be alot harder than i thought. Doing some research on ancestry and on google. I have gotten pretty much the same thing “Majority of these people in all of these regions are of Bantu origins”. If thats the case then why does everyone with cameroon/congo results (including me) have such small bantu results?. shouldn’t the bantu result match the cameroon/congo score?. If all these people are Bantu Ethnic groups, why isnt it showing? and by it not showing that much does that leave and increase likely hood that the Cameroon/congo gene is also mostly from the Southern parts of nigeria? rather than the actual places that its being labeled as?


        • Hey Vanessa,

          You raise very valid points about Gedmatch! Without proper understanding it tends to be really confusing and easily misleading indeed. I always compare the wild array of calculators to a box of candies whereby you get to pick out whichever ethnic flavour you happen to like or wish for 😉 Basically as with all DNA testing the input (a particular constellation of sample groups they use to compare your DNA with) determines the output (admixture results and ethnicity predictions).

          To be frank i’m not at all impressed sofar by any of the Gedmatch calculators when it comes to describing the within-Africa origins of Afro-Diasporans. I believe their socalled Oracle analysis (which only goes as far as finding a best fit to 4 populations) is best suited for people without any complex background. Almost by definition this leaves out people from the Diaspora who will be heavily mixed in their various African lineages from allover West & Central Africa. Infact also people with only 1 ethnic background sometimes receive very surprising and inaccurate predictions. I have been meaning to put up a separate page featuring the GEDmatch results of Africans with verifiable background, as i believe this will be very insightful, even when Afro-Diasporans are as i mentioned a much more tricky case to unravel.

          The regional admixture analysis being provided by GED Match seems to be mostly reflecting VERY ancient migrations, taking place more so thousands of years ago rather than hundreds of years. In itself this can provide interesting information if you want to learn more about ancient population migrations. However i do not find it useful at all for the purpose of tracing back African lineage from within a genealogical timepriod of say the last 500 years. Again without proper understanding this can be very misleading, especially when the socalled East African scores seem to hold a great appeal for many people…

          Having said all that i do think GED Match can be useful for all the other services it provides, especially connecting with matches from a potentially greater pool. Also zooming into non-African ancestral components is made easier, eventhough you do need to pick out a suitable calculator and have a basic idea of what you’re looking for. If you google for it there are many good tutorials out there on the web.

          I feel you about the initial disappointment. I think most people get into this DNA testing thing wishing to be told to be from just a minimal number of places. Preferably being from one single “tribe”. I suspect the sheer number of ancestors going back to the 1700’s when most African arrived in the US is often underestimated. As well as the heavy intermingling of African bloodlines over the many generations. But history cannot be undone. Personally i think a very colourful breakdown featuring many different African origins is also something beautiful in itself as a true expression of the (West & Central) African roots of the Diaspora.

          About the Trace regions, i find that they are least useful when it comes to the European breakdown of AA’s, generally i would only go by the West European regions with the highest amounts and which also make historically sense unless you have compelling evidence in support of the more “exotic” regions. Keeping in mind also that Europeans themselves are described by AncestryDNA as being regionally mixed because of ancient population migrations. Also the tiny Asian amounts (usually <1%) are very close to what i would consider noise level, even when in fact they could still be suggestive of (very distant) Native American ancestry being misread. I intend to do a blogpost where i will delve deeper into the non-African scores for AA’s.

          The African Trace regions are still undeniably part of your African DNA, but it's just too difficult at this stage for AncestryDNA to be absolutely sure to assign them regionally speaking. Frankly i applaud such an honest approach above pretending that a 100% flawless ethnic prediction could somehow be managed. Despite the reduced confidence level i find that some genetically distinctive regions such as Senegal, SE Bantu and North Africa can still be expected to be reasonably accurate even at trace level.

          There is indeed a great overlap between the socalled Cameroon/Congo” and “Southeast Bantu” regions. Both of which actually detect Bantu speaking origins. I would say Cameroon/Congo more so than SE Bantu, despite the labeling. As in fact for Northeast Africans this socalled SE Bantu region is indicative of Nilotic or Cushitic origins. This all has to do with the rather minimal and unfortunate combining of samples to underpin this socalled SE Bantu category. These samples are taken from Kenya, South Africa and Namibia (which borders Angola on the Atlantic coast!). I will actually make a separate section for Central & Southern African AncestryDNA results soon, where i will try to clarify this further. So keep an eye out on that!


          • Hello Fonte Felipe

            Thank you again for the info!
            I Have one question though about what you said about the African trace regions. “Despite the reduced confidence level i find that some distinct regions such as Senegal, SE Bantu and North Africa can still be expected to be reasonably accurate even at trace level” I can understand it being abit more believable as a trace region Like Senegal or Mali (even though Mali for me escaped the trace regions,though not very high)as they are still on the west coast of Africa and were involved in the trade. Also taking into account that neighboring countries have Mali and Senegal DNA also like Ghana(So Mali DNA could actually be just more Ghana Lineage or something else), And Malians have Senegal DNA and Senegal has Malian DNA. But Isn’t it more likely that SE Bantu is NOT actually South eastern from Africa, but more on the lines of batu ethic groups on the west side of Africa? Because i know they said slaves were taken from Mozambique. but it was something small like 2% or something like that which should be washed away by the Majority Of other ethnic groups. So wouldn’t it make more sense that it still most likely includes West Africa?. And North Africa is always 1% it seems for people, I doubt that should be taken literally as a North African, and more on the lines of Fulani IF however it isn’t just noise.


            • I always advise to take the country name labeling with a grain of salt and take into consideration crossbordering overlap and ancient migrations/admixture. This applies even more when regions are being reported at trace level. So i agree with all your remarks. However the mere appearance of genetically distinctive regions such as Senegal, SE Bantu or North Africa – even below trace level – does suggest a genuine finding which could be useful.

              The point i was trying to make is that I would not consider these trace amounts to be mere random outcomes or just noise to be discarded. Let’s take your 5% Senegal, i personally find it hard to believe that somehow AncestryDNA misread this portion of your DNA and instead it would really be 5% Cameroon/Congo or any other region which is geographically far removed from Senegal. The same goes for North Africa and SE Bantu which are on the extreme opposite sides of the African regional range and therfore more distinctive than neighbouring regions like Iv. Coast/Ghana, Benin./Togo and Nigeria.

              Although again the interpretation should be done very carefully and not just by taking the country name labeling as gospel. So to continue with your 5% socalled “Senegal” this actually also could have been inherited by way of ancestors from Gambia, Guinea Bissau, Mali, Sierra Leone and possibly even Liberia (Mande speaking groups).


  8. @FonteFelipe

    The Senegambia and windward coast region has also puzzled me. Doing hours of research i commonly notice that Senegambia and Winward coast region is ranked high for origins of slaves in North America. like #1 or #2. yet it also(Senegal in specific) comes out one of the smallest percentages for african americans. I know it can be a representation for other regions there, but its so very low. Is this because so many countries in that region have not been tested and haven’t ACTUALLY been added to the database yet? so they put it all into the Nigeria and Ghana category? because It doesn’t seem like very much of them has been put into the Senegal region( or other northern parts of west africa) if its percentage is so low, but its deportation of slaves Reportedly so high (the Gambia region) because that would mean 1 or two things. The reports of senegambia and winward coast are somehow wrong, or its is right and Ancestry still has way to many missing west African Countries, that they have to make a wild guess where the remaining Dna should go. Because AA’s have some missing dna that has not been added to the data base yet. It would seem that Gambia,Guinea Bissau,seirra Leone,and liberian Dna hasn’t truley been identified yet. It seems if they added them to the database. Every person that hailed from african slaves results would do a complete 180 and change DRAMATICALLY . And most likely people like me who have a high score in ANY region. would suffer a mass loss in percentage and the percentage would come out more even across the board.


    Africa: 81%

    Cameroon congo: 10%
    Guinea Bissau: 10%
    Senegal: 9%
    Benin/togo: 7%
    Liberia: 6%
    Mali: 6%
    SEB: 4%
    SSHG: 2%

    And id like to Thank you so much for your incite and patience, you have most likely been the most helpful in making sense of everything for me. It is very much appreciated.


    • Hi Vanessa,

      sorry for taking a while to respond, your comment sort of got swamped. Plus you ask very pertinent questions so i also wanted to take the time to give you a proper answer. Even if as always it will still be only my 2 cents 😉

      You are right to point out the lower than expected “Senegal” scores for African Americans. This was actually one of my main survey findings for the 350 AA results i analyzed. The group average i calculated (scaled) was around 8% while only very few people (5/350) had this region as biggest region in the African breakdown. Combining with socalled “Mali” i did find a higher Upper Guinean contribution of around 17% but still somewhat subdued compared with Central African and Lower Guinean contributions.

      Also during my Diaspora comparison i again reached the same conclusion and actually found that on average Hispanic Americans tend to have higher “Senegal” scores than African Americans (and actually also higher “Mali”). This contradicts the often made claim that North America proportionally received the most Senegambian captives in the Americas.

      It’s likely that there are several reasons at play which caused this outcome. If you read those two blogs of mine i go into it in greater detail. But in short this is what i came up with:

      – The estimated share of Senegambia in direct Trans Atlantic Slave trade to the US is about 20-25%. However this doesn’t take into account the role of Inter-Colonial (with the West Indies) and Domestic Slave Trade (often by way of Virginia). So quite likely that estimated share is overstating the actual proportion of Senegambian captives among the US slave population.

      – Possibly the gender ratio among Senegambians was less favourable (many more males than females) for having offspring when compared with other groups. And consequenty their bloodlines became relatively more diluted and dispersed.

      – Also the more specific origins of Senegambians within the wider area of Upper Guinea might have been misinterpreted sofar. The socalled “Mali” region scores (on average about 9% during my survey) seem to be suggestive of these origins often being located moreso into the interior or to the south of Senegambia proper.

      You also mention the Wind Ward Coast. The ancestral contribution from this area is however more trickier to distinguish (atleast when going by AncestryDNA results). Because people from Sierra Leone and especially Liberia will usually have predominant socalled “Ivory Coast/Ghana” scores! This is something which i have been saying from the start. Due to the lack of a separate region for these countries, their origins are going to be described by surrounding regions, in particular “Ivory Coast/Ghana” but also in addition socalled “Senegal” and “Mali”. You can verify this for yourself by looking at the Sierra Leonean and Liberian results i have posted here. Also i have calculated a group average for 6 Liberian AncestryDNA results i have seen sofar and their socalled “Ivory Coast/Ghana” average of around 80% is nearly the same as for 8 Ghanaians! (see this link).

      This is understandably due to a lack of samples but it does mean indeed that for example your 10% socalled “Ivory Coast/Ghana” could (in part) be traced back to the Wind Ward coast rather than to the Gold Coast. I am positive that in the near future new African sample groups will be added to the socalled Reference Panel of AncestryDNA and this will surely improve their analysis (which will however still ultimately be based on estimates!). So yes you can expect that with an upcoming update your breakdown will change and the percentages for some regions might increase while for others it will decrease. However i do not foresee a drastic 180 turnaround but rather a finetuning. For one the continental breakdown should remain pretty much nearly the same. But i suspect that also the 3-way subcontinental breakdown i have been using in my survey (Upper Guinea versus Lower Guinea versus Central AFrica) will remain roughly the same, at least on average. I expect the subregional composition to change the most, but combining neighbouring regions (incl. the new ones) should still add up pretty much the same. I would be highly surprised if in your case for example your combined Senegal + Mali scores (together 11% Upper Guinean) would all of a sudden become a combined Upper Guinean score of 35% (adding Gambia, Guinea Bissau, Senegal and Mali from your example).

      Adding more samples from Guinea Bissau or Liberia will undoubtedly help detecting ancestral connections to these areas. But genetic overlap between countries and ethnic groups will still be an issue (inherently so). Also i believe that the current “Senegal” and “Mali” regions despite their limitations are already quite predictive as they have proven very valuable in confirming the Upper Guinean origins of Cape Verdeans.


  9. Excellent blog! I’m happy I found your information about South Carolina. I’m one of the high Mali outliers at 47%. I also only have 3% European and 1% Native. My breakdown is:

    Mali – 47%
    Benin/Togo – 19%
    Ghana/IC – 14%
    Nigeria – 11%
    Cameroon/Congo – 5%

    This is my African composition.

    Reviewing your blog, it seems the largest chunk of my ancestry is from western Mali? This makes since because my family is from coastal South Carolina and traders brought lots of people from that region.

    Liked by 1 person

    • Thanks for your comment, that’s an exceptionally high Mali score! Sofar in my AncestryDNA survey the highest Mali score i have seen was 42%, this includes over a 1000 AA results but also hundreds of results from the West Indies, Latin Americans and actual Africans!
      Quite extraordinary as according to Ancestry’s own info the “typical native” (based on their own 16 Malian samples) would only score 39% “Mali”.

      I have been doing a separate survey on results from South Carolina and i would love to include your complete results! Here’s the link to my online spreadsheet which has all the results sofar:

      South Carolina AncestryDNA results

      If you’re okay with it please send me an invite to view your results by following these steps:

      – Sign in to
      – Click the DNA tab and select Your DNA Results Summary
      – Click the Settings button on the right side of the page
      – Scroll down to the sharing DNA results section
      – Enter my Ancestry username which is FonteFelipe
      – Select role of guest
      – Click the SEND INVITATION button

      With your permission i can then also have a look to see if you have any African matches according to this method i blogged about:

      How to find those elusive African DNA matches on Ancestry
      Quite possibly you might already have some matches which connect to your 47% Mali score and might then also give additional insight!

      Thxs in advance!


  10. Thanks! I sent the invite. This reminds me the results I posted are the new update results from Ancestry in early 2016. I screenshot those results. I talked to AncestryDNA a few weeks ago and they did the regions first and then they will post the updated results. My old results are still in my account which is good because I can get an opinion on the old and updated results. Please feel free to run the analysis of ethnicity. I’m curious to know what you discover.


    • That’s peculiar! The results which are viewable now only show Mali as a trace region with an amount of 3%. While your primary region is socalled “Ivory Coast/Ghana” with 29% and also Senegal is showing up quite strongly with 13%. As i always mention on my blog the country name labeling by AncestryDNA should not be taken too literally. From what i have seen from actual African results it appears that people from Sierra Leone especially often receive results which are a combination of socalled “Ivory Coast/Ghana” and “Senegal”. This is due to the linherent limitations of the current AncestryDNA setup.

      While looking through your 3,261 DNA matches i was able to filter out 5 matches who are most likely African. One of them from Nigeria, one possibly from Central Africa and most intriguingly 3 matches who could be from either Liberia or Sierra Leone! As these countries are geographically in between Mali, Senegal and Ivory Coast it’s no surprise that their DNA is described as a combination of these regions. I suppose this might also apply to you and that’s why Ancestry had to update your results.


  11. Thanks for the info! My family is all from South Carolina and your research lines up with what I have researched on ethnic origins for South Carolina. The 3% Mali and rest of the ethnic breakdown you saw on Ancestry was the original breakdown from four years ago. However, in Feb 2016, Ancestry was updating there DNA estimates and expanding ethnic categories. That’s where my 47% Mali comes from and Senegal was less than 1%. So I suspect the biggest chunk of my ancestry is from people of the Mande language group. Also Ivory Coast/Ghana and Benin/Togo had Mande people intermixing with the ethnic groups there like the Mossi, Dagomba, Diola peoples. Those were the people captured by Ashanti and Dahomey empires and shipped to slavery. That explains the high Mali percentage. This also explains the original results.

    Thanks again!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s