I’m so tired of having to argue this over and over so decided to make a post. Originally this was a response to someone who isn’t BA attempting to police BA DNA and history. It’s very weird to be told who I am by someone who doesn’t know me nor my history nor my family’s history within that history.
It is important that you understand that modern commercial DNA tests are using inferences from people who were operating under the influence of Racism. The DNA isn’t wrong but the interpretation of the DNA is wrong and submerged in racism and filtered through racist taxonomic systems that did not exist in the same way they did just 100 years ago let alone 400 to 500 years ago.
They are deceiving you.
Polished explanation for anybody to use in the future against ideologues who still push relabeled theories of Race Science.
Commercial DNA tests cannot prove distant ancestry. They are statistical models that use predetermined MODERN reference groups (people are fluid and move and mix all the time) and probabilities to output a percentage of similarity or resemblance.
Any group can be interlinked with African populations under this same analysis especially under low resolution. The DNA argument uses these reference groups that are filtered through European taxonomic systems that collapses identities.
It cannot and doesn’t account for reclassification system. They use DNA tests to obscure groups all throughout Africa Asia and Europe
It’s like how many BAs are given a statistical output of Nigeria and the various tribes found there when those identities or tribes didn’t exist exist during the time period and most of the TAST groups were not even from the region of Nigeria. It also ignores how these groups have changed since these time periods and the historical record
DNA is testing modern groups against modern groups that their references models place them based on what their modern classification systems dictates. Not to mention how DNA is assigned when the system cannot accurately place it.
And even when they are tested most have affinities with populations that were localized. Samples have not been gathered for various reasons because do you forget the intrusive nature of settlers who were violently exploiting these groups? These are simply not the same populations and they too have undergone changes.
There’s no such thing as “African” “Europe” “Asia” DNA. DNA doesn’t care geographical locations. Are the “Arabs” who have been in North Africa for about a thousand + years Arabs or Africans? DNA doesn’t work like that. They can test these things in high resolution but these tests are doing that
Not to mention how the SNP chips used in these test are heavily skewed towards certain populations
It’s an algorithm
Furthermore these tests are inferences. Genetic “clustering” is a statistical inference based on similarity to reference populations that scientists already chose.
When scientists say Black Americans “cluster with” West/West-Central Africans they are saying that parts of their genomes resemble those reference populations statistically due to shared ancestors brought by the slave trade. This is an inference conditioned on assumptions made by scientists, modern reference sets and comparative models based on those inferences
DNA clusters are inferences.
In the context of genetics and ancestry tests, a “cluster” is not a physical object that exists in your blood like a red blood cell. It is a statistical conclusion derived from analyzing how similar your DNA is to the DNA of other people in a specific database.
When a DNA test tells you that you belong to a specific cluster (like “Eastern European”), it is not comparing you to ancient people from 1,000 years ago.
It is inferring your ancestry by comparing you to modern people living in those regions today who claim to have deep roots there.
If the company adds 5,000 new people to their database from Poland next year, the mathematical definition of the “Eastern European” cluster changes, and your percentage will likely shift.
When geneticists first built the chips to study human diversity, they prioritized markers that maximized the difference between the largest available sample groups: usually Europeans and West Africans (specifically Yoruba).
The mathematical models are “tuned” to detect West African markers because that is what they were taught to look for. They are insensitive to markers that might connect Black Americans to other specific regions because those specific unique markers are often excluded from standard commercial chips.
If you only have a tool that detects the color blue, and you point it at a rainbow, it will tell you the rainbow is “mostly blue.” It isn’t lying, but it is missing the full spectrum because of how the tool was built.
The “industry standard” for genetics often ignores the “Ghost Populations” of those who were here before the Trans-Atlantic Slave Trade (TAST) or those whose specific lineages were never sampled because they don’t fit the “Yoruba-centric” model of West Africa.
The study you linked (PMC4386999) touches on the bias of reference panels. Because the Yoruba (specifically from the HapMap project) were some of the first and most thoroughly sequenced African populations, they became the “gold standard” for “African DNA.”
When an algorithm sees a Black American sample, it is programmed to compare it to the Yoruba. If there is a 70% match, the algorithm doesn’t necessarily mean “Your ancestors were Yoruba”; it means “Out of the limited options I have, the Yoruba are the closest statistical fit.”
This effectively “overwrites” smaller groups, such as those from the Senegambia or the interior of the Bight of Biafra, and completely ignores any indigenous “Old Settler” genetic signatures that haven’t been prioritized for sequencing.
Why do they compare to “Modern Africans”?
The Illumina datasheet referenced is a prime example of clinical pragmatism over historical accuracy.
In medical genetics (like studying Multiple Sclerosis or Kidney Disease), researchers look for “Risk Alleles.” They compare Black Americans to modern West Africans because they are looking for broad continental markers to identify which parts of the genome are “Admixed.”
By focusing strictly on “Admixture” (European vs. African), they skip over the possibility of a third, indigenous American component that doesn’t fit the “Siberian-descended” Native American model.
If they don’t look for a specific “pre-TAST” signature, the math will simply fold those sequences into the nearest major cluster which is often “African” or “Unassigned.”
Do you understand the gravity of this? It’s an effect of the reclassification system and a continuation of Race Theory though cloaked.
The Brill study regarding Jamaican and Puerto Rican ancestry highlights a significant statistical hurdle for the “Native to the Americas” argument because they rely on Ancestry Informative Markers (AIMs) that are calibrated to a specific, narrow **definition** (reclassification system, race theory, etc) of “Native American.”
In these models, Black Americans and Jamaicans often show 70-90% African ancestry whereas Puerto Ricans show closer to 20-30%.
The inference would be that if Black Americans were “Native” (Indigeneous) in the same way the Taíno or the Navajo are, their “African” percentages should theoretically be much lower, and their “Indigenous” percentages should be much higher.
My big point here is that the SNP chips are not calibrated to see the “Ancient American” signature in Black populations.
If the chip is told that “Indigenous = East Asian-related,” and a Black American carries a signature of a different ancient American group that isn’t in the database, the algorithm will likely “misclassify” it as African because of the high density of variation in African genomes.
Historically, many groups were reclassified on paper (from “Indian” to “Colored” or “Negro”).
Genetic tests do not account for this paper trail.
If a population was “reclassified” in the 18th century, and a DNA company uses a 21st-century reference panel that labels that entire region as “African,” the test will output “African” regardless of the group’s actual 500-year history on this soil.
Commercial DNA tests are descriptive, not prescriptive. They describe who the system thinks you are based on the data it chose to include.
If the system doesn’t include pre-colonial “Black American” indigenous references and if the system uses Yoruba as a catch-all for West Africa then the output will always be a circular confirmation of the current Eurocentric origin myth
It validates it.
It isn’t that the DNA is “wrong,” it’s that the **interpretation** is a closed loop.
Current genomic models assume that all “Indigenous American” DNA must look like the DNA of modern tribes who descend from the Bering Strait migration.
If you have 80% “African” markers, the algorithm assumes those 80% must have come across the Atlantic.
If there were a population already in the Americas who shared deep ancestral alleles with West Africans (due to even older migrations or shared ancient lineages), the SNP chip would simply label those markers “African.” It has no “Native Black-American” reference category to assign them to.
The Illumina and Brill papers use Admixture Mapping. This process breaks the genome into chunks and asks: “Is this chunk more like the European reference or the African reference?”
In Puerto Rico, the “Native” signature (Taíno) is distinct enough from the Spanish (European) and the Yoruba (African) references that it shows up as a third, clear color on the graph (usually 10-20%).
In Black Americans, because the “Indigenous” and “African” references may share high levels of genetic variation that the current SNP chips cannot distinguish, the “Native” component often “disappears” into the African cluster. The algorithm defaults to the most “mathematically likely” answer based on its modern database.
Many Black Americans have significant amounts of DNA that commercial tests struggle to place or that fluctuate between “Nigeria,” “Senegambia,” and “Broadly Western African.” Researchers often attribute this to the high genetic diversity within Africa. Again inferences. However, my point is that this “unassigned” or “broad” variation could technically be the signature of localized, pre-colonial American populations that have been statistically reclassified as African. It’s the missing 15% to 20%
A big question to ask is if they ever extracted dna out of enslaved Black American graves from the 1600s to 1800s and analyzed their DNA?
Yes.
They’ve done this with small datasets.
The Catoctin Furnace Study (Maryland, 2023)
This is one of the most recent and largest studies, published in Science. DNA was extracted from 27 individuals buried in a cemetery used by enslaved ironworkers between 1774 and 1850.
Researchers found genetic affinities with modern Wolof and Mandinka groups (Senegambia) and Kongo groups (Central Africa).
By comparing this DNA to the 23andMe database, they found over 41,000 living relatives, most of whom are in the American South.
This study used modern 23andMe reference sets to draw its conclusions, essentially looping the historical remains back into the modern “Nigerian/Senegambian” classification system.
THE AMERICAN SOUTH? What about other places?
The NY African burial is another point of interest.
This site contained roughly 15,000 sets of remains from the 17th and 18th centuries. This study was led by Dr. Michael Blakey (Howard University), this project initially focused more on craniometrics and bone chemistry (strontium analysis) because the DNA was poorly preserved in the acidic soil.
Strontium levels in the teeth showed that many of the adults were born in Africa and brought to New York, while the children were mostly born into slavery in New York.
Strontium isotope ratios identify the geological region where a person spent early childhood but they do not establish exclusivity of population origins nor do they test for the existence of other contemporaneous groups outside the sampled burial context.
The New York African Burial Ground represents a specific, urban, colonial labor population shaped by forced importation. Finding that many adults were born in Africa and many children were born locally is exactly what one would expect in a port city heavily supplied by the transatlantic trade.
This pattern confirms recent forced migration into that context not the absence of earlier dark-skinned or indigenous populations elsewhere nor their survival in non-urban, non-enslaved, or differently classified communities. It’s like when they find Near Eastern DNA in specific areas of Ancient Egypt and declare the entire 3,000 year old civilization populated by Near Eastern populations.
The Zoutsteeg Skeletons (Saint Martin, Caribbean, 2015) where dna was extracted from three skeletons of enslaved people dating to the late 1600s.
Using whole-genome capture, researchers linked them to modern Bantu-speaking groups in Cameroon and non-Bantu groups in Nigeria/Ghana.
Thing is though these studies often fall into the trap I described earlier as they use modern DNA to explain the past.
When a scientist says a skeleton from 1780 is “Yoruba,” they are using a PCA (Principal Component Analysis) plot.
The skeleton’s DNA is a “dot” on a graph. If that dot lands in the middle of a bunch of modern Yoruba dots, the scientist labels it Yoruba
A Principal Component Analysis (PCA) plot is a mathematical simplification. It only shows how samples relate to each other.
If you run a PCA and only include “Yoruba,” “English,” and “Maya” as anchors, every human on earth must land somewhere between those three points.
By excluding “Ancient/Indigenous Black American” as a reference group, scientists make it mathematically impossible for a test-taker to “land” anywhere else but the African cluster.
It is a circular proof: they didn’t find the group because they didn’t look for it, and they didn’t look for it because they claim it doesn’t exist.
SNP chips are designed to detect “Beringian” (East Asian-related) Indigenous DNA. If the indigenous populations of the Americas included groups with “African-like” phenotypes and genotypes (Paleo-Americans), their DNA would not show up in the “Native American” column of a commercial test.
The algorithm is forced to choose. If it sees a segment that looks 60% like a West African and 40% like nothing in its database, it will “smooth” that segment into the African category to increase its confidence score. Therefore, the “low percentage” isn’t a reflection of history, but a limitation of the reference panel.
The standard argument of “We use specific markers (AIMs) that are unique to Africa to prove your origin.”
Forgets that these markers are filtered through European taxonomic systems. “African” is a political and geographical label, not a biological one. If a marker is found in both a person in Mali and a person in ancient Brazil, Western science labels it “African” simply because they found it in Mali first or more frequently in modern times. This Colonial Priority in naming genes creates a bias where “Africa” becomes the default “source” for any high-diversity DNA, effectively erasing the independent evolution of similar traits in the Americas.
If the “dictionary” (the reference database) does not contain the “words” (the DNA of the people you are looking for), the system will simply translate your DNA into the closest word it knows.
In this case, “African” has become a “catch-all” term for any DNA that the Western scientific model doesn’t want to admit was already present in the Western Hemisphere.
Scientists sequenced children buried 3,000 and 8,000 years ago at a site called Shum Laka in Cameroon (the heart of the so-called “Bantu expansion” region).
They assumed these ancient people would look genetically identical to modern West/Central Africans (like the reference panels used by Ancestry/23andMe).
They did not. The ancient DNA showed deep divergence. These ancient people carried DNA from a “ghost population” that no longer exists in unmixed form today. They were significantly different from modern Bantu speakers.
This proved that modern West Africans are not perfect proxies for ancient populations. Using modern Nigerians to model ancient ancestry is an “inference” that misses huge chunks of lost diversity.
