Without wanting to rehash things this blog post and one more upcoming will provide a quick recap of my final AncestryDNA survey findings for continental Africans. These results were obtained during 2013-2018 but I had not fully processed all the data up till now. Arriving now at the first part of my West African section. Which contains results from Liberia, Ivory Coast, Burkina Faso, Ghana & Benin. One striking research outcome was the clear distinction between Akan & Ewe results.
I first published my preliminary West African survey findings on 24 February 2018 when I had only 41 AncestryDNA results available for my analysis (see this overview). As my survey has been ongoing I have managed to collect a sample group which has now doubled in size. Consisting of no less than 82 AncestryDNA results from Liberia, Ivory Coast, Burkina Faso, Ghana and Benin! Follow the link below for detailed analysis & new screenshots:
Table 1 (click to enlarge)
Aside from my Nigerian survey group (n=87) the sample size of my Ghanaian survey group (n=42) was the biggest within my entire African survey. Although still limited this already provides a rather robust basis. And also crucially I have managed to gather plausible ethnic details for almost all of my Ghanaian survey participants. Allowing for a finer detailed ethnic analysis of Ghanaian genetics. Which is why I think these “old” results may still be useful and are not obsolete yet. Even when I did discontinue this survey after Ancestry’s update in September 2018. Because in my opinion Ancestry’s new version of their ethnicity estimates regrettably has been a downgrade in regards to their African breakdown, overall speaking (see also this blogseries).
I originally singled out three main implications for Afro-Diasporans. All of which can be maintained and have been discussed already in previous blogs. In this blog post I will revisit the question if it is possible to distinguish Akan from Ewe lineage. Furthermore I will briefly touch upon 23andme’s new “Ghanaian, Liberian and Sierra Leonean” category and also how Ancestry might improve things in the near future. Especially in light of their upcoming update.
- “Benin/Togo” also describes DNA from Ghana (see this blog post as well)
- “Ivory Coast/Ghana” also describes Liberian DNA
- “South-Central Hunter-Gatherers” suggestive of remnant West African Pygmy DNA?
- Is it possible to determine the most likely ethnic source(s) of your Ghanaian lineage?
Figure 1 (click to enlarge)
Can the Akan be genetically distinguished from the Ewe?
Map 1 (click to enlarge)
To get straight to the point: my answer to the question posed above would be a qualified yes when it comes to Ghanaians themselves. I discuss this in greater detail on the main page if you scroll down to section 3. My survey findings might be counter-intuitive at first sight when overfocusing on country name labeling. Therefore I also take into account ancient migrations and the distribution of African language groups to make more sense of my research outcomes.
- West African results (part 1) (scroll down to section 3)
However for Afro-Diasporans this is an entirely different issue. Because of all the complexities involved.1 And generally speaking I do not think it is currently possible to obtain a “100% accurate” ethnic specification of your Ghanaian ancestry. At least not by relying solely on admixture analysis, such as performed by AncestryDNA and other commercial DNA testing companies. And actually also not by testing your direct maternal or paternal line (see this article). However by conducting careful follow-up research (DNA matches, historical plausibility, genetic genealogy etc.) I do think you can increase the chances of learning more about the likely ethnic backgrounds of your Ghanaian ancestors. See also :
- How to make more sense of “Ivory Coast/Ghana” scores (2013-2018)
- How to make more sense of “Benin/Togo” scores (2013-2018)
Returning to my Ghanaian AncestryDNA survey findings. The two ethnic groups I have selected for this research question the Ewe and the Akan2 are indeed fairly distinctive going by the preponderance of either primary “Ivory Coast” or “Benin/Togo” scores. This can be verified from Table 1 and Figure 1 above. And in even greater detail from the overview below. Which also features the actual range (min. & max.) within each subgroup. Obviously there are still limitations in sample size. Especially for the Ewe. Nonetheless it is quite telling that for all 18 of my Akan survey participants “Ivory Coast/Ghana” was convincingly reported as primary region (which actually also happened for my Ivorian Akan samples). The lowest score being “only” 71%. While for all 6 of my Ewe survey participants it was rather “Benin/Togo” which showed up with the biggest amount. Albeit with greater variation.The lowest amount being 46%.
This type of clear-cut differentiation is quite astonishing given that the Akan and the Ewe are actually neighbouring people with a great deal of historical interaction. Possibly the Volta river may have acted as some kind of barrier for greater genetic overlap. But most likely it is rather the relatively recent migration history of the Ewe and fellow Gbe-speakers which lead to this outcome. Originating in southern Nigeria according to traditional accounts and ending up in Benin, Togo and eastern Ghana (see this map).
DNA testing is revealing that usually ethnicity cannot be distilled into one single genetic category. Unless you happen to be a perfect match to the samples your DNA is being compared to (see figure 1). Instead typically most ethnic groups will be described as a composite of neighbouring and interrelated ancestral categories. This might be counter-intuitive at first. But it does actually make sense if you take into consideration the genetic impact of shared origins, ancient migrations and inter-ethnic mixing throughout the ages.
The distribution of language groups is a good indicator of the broader patterns to be taken into consideration. As generally speaking the initial expansion of distinctive populations was accompanied by the spread of their language and often also their particular mode of livelihood. Various scientific papers have already demonstrated that broad-scale population structure within Africa reflects both geography, (pre-)history and language (see this section for references).
Unfortunately this highly fascinating topic of early West African (pre-)history is often absent or poorly represented in the curricula of primary and secondary education. Furthermore both scholars and laymen tend to pay more attention to other areas of Africa. In particular Northeast Africa because of early human evolution or Ancient Egypt. Regrettably this relative neglect of West African orientated research has lead to a severe lack of basic knowledge among the general public. Leaving them vulnerable to serious misinterpretation and even at risk of being mislead by agenda-driven fringe movements.
It is a true shame that West African (pre-)history too often is forcibly linked to other faraway places simply for prestige reasons or misguided ideology. This is not to deny that West Africa has indeed been in long standing mutual contact with neighbouring parts of Africa. And these connections are certainly also research worthy. Then again so much insight is to be gained when you consider West Africa to be a regional unit within its own right and on its own terms. With its own independent historical driving forces centered principally in modern-day Mali and modern-day Nigeria along the Niger river valley, the true bearer of civilization for West Africa in so many ways. Some suggestions for further reading:
- General History of Africa (Unesco) (all volumes online available)
- The Peopling of Africa: A Geographic Interpretation (Newman, 1995)
- Themes in West Africa’s History (E.K. Akyeampong, ed., 2006)
In conclusion my final Ghanaian survey findings are suggesting a noticeable degree of genetic differentiation can indeed be observed between the Akan and the Ewe. Not per se consistent on an individual basis but quite apparent when based on group averages and approximate tendencies. Highlighting that regional admixture DOES matter!3 Correct interpretation might not always be easy. But being aware of the relevant context does clarify things. More research is however needed to establish a firmer basis for determining the degree of genetic variation and how it may (roughly) correlate with ethnic background in Ghana.
Find your Akan & Ewe matches
Table 3 (click to enlarge)
African DNA matches (autosomal) provide one of the most reliable avenues to Trace African Roots. Even more so when combined with well-interpreted admixture analysis. Fortunately an ever increasing number of West Africans, or better yet West African migrants and their children, are taking DNA tests. All of them potential DNA matches for Afro-descendants to get in touch with and learn more about your specific Ghanaian lineage by finding your Ghanaian DNA cousins! See also my tutorial below as well as my recently started service which will provide you with a African DNA Matches Report.
Despite the great potential to learn more I would still advocate a cautious approach though! Make a careful assessment of your Ghanaian DNA matches, if you happen to have them. Taking into account shared segment size and the possibility of population matches (IBP) and false positives, so that only genuine IBD matches are left over. Ask yourself the following questions:
- Do you have any Ghanaian matches?
- How many of them are likely to be Akan, Ewe, Ga-Adangbe etc.?
- Is any ethnic background predominant among them?
- Is the shared amount of DNA big (>10cM) or small (<7cM)?
- Are these matches also shared with your close family members?
- How do these matches fit in your overall African DNA matching patterns?
- How do your Ghanaian matches correspond with your admixture results indicating Ghanaian lineage?
Read the blog post below for much more details on how I performed such an analysis for 50 Cape Verdeans and their 437 African matches. I intend to do similar blog posts for other parts of the Afro-Diaspora eventually. And also among Ghanaians themselves (of both Akan and Ewe background) to establish which parts of the Afro-Diaspora appear to be most genetically linked to them, going by DNA matches.
Surprisingly also two Ghanaian matches were included among these findings. Uncommon and atypical given Cape Verde’s nearly exclusively Upper Guinean origins (confirmed also by admixture analysis). Both matches however had only a very minimal shared segment size (6 cM). A useful reminder therefore that smaller-sized matches might often be false positives (IBS) or also population matches (IBP) (see this useful chart). Genealogically not meaningful and definitely not for establishing if you could have substantial Akan or Ewe lineage. Although an increased frequency of such small matches might still be indicative I suppose. Especially when combined with bigger matches.
23andme better equipped now to distinguish West African lineage?
Map 2 (click to enlarge)
Ancestry’s West African regional framework (2013-2018) was pioneering and could be very useful. Given correct interpretation. However sadly after Ancestry’s update in 2018 it lost most of its informational value. Currently you are much more likely to get a realistic estimate of your West African lineage on 23andme. Even if of course not 100% accurate 😉 Still going by actual West African 23andme results the predictive accuracy of their 3 new West African regions (“Senegambian & Guinean”, “Ghanaian, Liberian & Sierra Leonean” and “Nigerian”) is quite impressive.
I have started a new survey based on the newly updated African breakdown on 23andme among not only Africans but also various parts of the Afro-Diaspora. It is ongoing but you can see a preliminary overview by following the links below:
- African breakdown for Africans according to 23andme
- African breakdown for Afro-descendants according to 23&me (scaled to 100%)
- 23andme results from West Africa (part 1) (after 2018/2019 update)
- 23andme results from West Africa (part 2) (after 2018/2019 update)
The number of my West African sample group on 23andme is still minimal however it is already quite telling how generally speaking a greater predictive accuracy can be observed. When going by the group averages of my West African survey participants. Even when actually the difference is not always that big when compared with Ancestry’s 2013-2018 version (see table 1). In particular for Ghanaians it seems that additional “Nigeria” scores on 23andme have substituted former “Benin/Togo” scores on Ancestry. Resulting in similar group averages of around 70% for either “Ghanaian, Liberian & Sierra Leonean” or “Ivory Coast/Ghana”.
It should be noted also that even when similarly labeled, ancestral categories on different DNA tests will not be perfect equivalents or measuring the exact same thing. The additional Sierra Leonean samples on 23andme for example are clearly causing a shift westwards. While there may also be some algorithm differences. Furthermore the current lack of similar categories on 23andme for “Benin/Togo” and “Mali” does impact its ability to attain the same level of often meaningful resolution on Ancestry’s 2013-2018 version. But regrettably Ancestry’s last update has only made things worse. The generally inflated “Benin/Togo” and sharply decreased “Ivory Coast/Ghana” scores epitomizing the loss of focus in AncestryDNA’s updated African breakdown.
Suggestions for improvement
“ Replacing the current “Ivory Coast/Ghana” region with three separate and properly labeled regions to describe and measure genetic affiliations with either Kru, Akan/Kwa or southwestern Mandé samples could increase its informational value tremendously.”
“The creation of a more narrowly focused region to describe and measure genetic affiliations with Gbe samples from not only Benin and Togo but also Ghanaian Ewe would be much more beneficial”
“Gur samples from Burkina Faso might be used to create a greatly needed intermediate region to cover the genetic legacy of people nowadays found in northern areas of Ivory Coast, Ghana, Benin and Togo as well as Burkina Faso itself. It may also result in a sharper delineation of especially the “Mali” region which will become more strictly suggestive of Upper Guinean roots.”
“Create new African “migrations”, a.k.a. genetic communities. As far as I am aware currently there are only two “migrations” in place for Africans. […] Especially for Nigerians and Ghanaians I would imagine something could already be set up. Even more so when appropriate academic samples can be added. “
The quotes above are taken from my previous blog post from June 2018:
Unfortunately hardly any of these suggestions were taken up by Ancestry when they did their last update in September 2018… However it seems that a new update may be upcoming already (see this link). And from what I have seen in preview this will also affect the African breakdown. Either way Ancestry has announced they aim to expand their African Reference Panel (see this blog post). So I might as well give it another shot 😉 I still stand by my statements quoted above.
In particular I believe that when wanting to get more solid ground on the question of either Akan or Ewe origins for Afro-descendants it will be very helpful if Ancestry were to create Ghanaian “migrations”. However not ethnically labeled. As this will often be misleading! But rather something along geographical lines like distinguishing a western Ghanaian “migration” from an eastern Ghanaian “migration”. The latter grouping perhaps to be combined with southern Togo and southern Benin. After all if a country as small as Ireland has no less than 92 (!!!) distinctive sub-regions on Ancestry (see this article) why not further specify Ghana, which is among the principal countries of provenance for Afro-Diasporans!
Obviously AncestryDNA’s regional breakdown (2013-2018) was rather basic and had several flaws. I like to underline though that my West African AncestryDNA survey (2013-2018) did produce several potentially insightful findings. Not only for improving the interpretation of the results of Afro-Diasporans. But also for West Africans themselves it may lead to better understanding of the understudied migration history within the African continent. Both relatively recent (last 500 years or so) and more ancient. I have had many stimulating discussions with West Africans who did a DNA test over the years. Unlike what some people might assume many Africans themselves also take a great interest in these matters.
Then again my intention is not only to look backwards but also forwards! In particular I believe that my survey findings (see table 1) may provide a helpful baseline to compare with 23andme’s newly updated African breakdown (see this overview). As well as functioning as a benchmark for Ancestry’s upcoming update, which hopefully this time will lead to improvement!
Over all I would say that my West African survey findings are pretty much reinforcing the main message I have been trying to get across from day 1: the labeling of ancestral categories in DNA testing is often not be taken too literally. However this does not mean you cannot derive a great deal of informational value from your socalled ethnicity estimates! Proper interpretation is a precondition. And it has been my experience that being aware of the DNA results of native West Africans certainly helps in this regard. If only to know what to expect more or less. And also in order to make a proper inference of how you yourself fit in the scheme given any other ancestral clues you might have. Even more so when also taking into account your African DNA matches, relevant historical context, (advanced) genetic genealogy and other means of follow-up research.
I am aware that ethnicity can be a sensitive topic for West Africans. Aside from wanting to learn more about my own West African roots my motivation to research these so-called Ethnicity Estimates is purely scholarly. It stems from a deep fascination with West Africa’s ancient population migrations, its many connections with the Afro-Diaspora as well as a profound love for its vibrant & diverse culture. I do not condone the misuse of my research for identity politics! I am in full support of democratizing knowledge. However when parts of my work are being copied I do expect that a proper referral to my blog and the additional context it provides will be made.
Please also keep in mind that DNA testing can be very educational and may have many positive effects. However in some cases it may also be abused by people with bad intentions.
- Can DNA Ancestry Testing Make You More Racist/Tribalistic? (Scientific American)
1) Understandably many people desire to have the most specific degree of resolution when searching for their African roots. They want to be able to pinpoint their exact ethnic origins and preferably also know the exact location of their ancestral village. In a way following in the footsteps of the still very influential ROOTS author Alex Haley. Unfortunately these are rather unrealistic expectations to have in regards to DNA testing (at least in regards to admixture analysis.). Not only given current scientific possibilities. But also because such expectations rest on widely spread misconceptions about ethnicity, genetics, genealogy as well as Afro-Diasporan history. Too often people ignore how the melting pot concept is really nothing new but has always existed, also in Africa!
Generally speaking the African regions on AncestryDNA have become more generic and less specific after the update in 2018. As a consequence the country name labeling has become even more misleading when taken at face value. On the other hand ancestral categories referring to ethnic groups might be just as deceptive or even more so! As many people will again tend to take them too literally. Underestimating not only the sheer number of ethnic groups existing in Africa (thousands!) but also the complexity of interplay between fluid ethnicity, overlapping genetics and shifting political borders. The same goes for precolonial African kingdoms which again were not static entities. But instead very often ended up being multi-ethnic after expansion and assimilation of neighbouring peoples.
Either way for such an endeavour to succeed one should preferably use ancient samples from relevant time periods and locations. Rather than modernday African samples whose pedigree may very well include many inter-ethnic unions within the last 200-300 years or so. Of course the intermingling of African ethnic lineage continued even more so within the Afro-Diaspora. Again during several centuries but this time also involving ethnic groups geographically far removed from each other. All of which resulting in a very intricate mix which remains tricky to disentangle.
Personally I prefer ancestral regions which are referring to either non-political geography or meta-ethnic/linguistic groups. Such as Atlantic, Mande, Kru, Akan, Gbe etc. (see this page). But I fear that inherently there will always be some degree of blurriness involved and exact delineation might be impossible to achieve in many cases. Instead of generating false hope it might be a more honest approach to go by the motto of “don’t be more specific than your data supports”. Previous blog posts of mine dealing with this topic:
- “What tribe am I?”
- Fictional Family Tree incl. African Born Ancestors
- Is it possible to pinpoint a plausible ethnic origin for one’s African bloodline?
- What can be learnt from AncestryDNA when trying to trace African ancestry?
2) Technically speaking the Akan are more so a so-called meta-ethnicity: a grouping of closely related ethnicities based on linguistic similarity. Consisting of well known subgroups such as the Ashanti, Fante and also the Baoule from the Ivory Coast.
3) According to some people only continental admixture is to be taken seriously in DNA testing. Sub-continental admixture, a.k.a. ethnicity estimates, a.k.a. regional admixture only being fit for entertainment purposes. I myself have never taken this stance. Preferring to judge each case on its own merits. Attempting to maximize informational value despite imperfections and avoiding source snobbery. Which is why I have conducted my AncestryDNA surveys among Africans and Afro-descendants in the past. Applying an additional macro-regional framework for extra insight (Upper Guinea, Lower Guinea, Central/Southeast Africa).
Basically combining interrelated & neighbouring genetic regions in order to allow certain regional patterns to show up more clearly. This turned out to be particularly helpful when wanting to explore any rough correlation between aggregated AncestryDNA results for Afro-Diasporans and slave trade patterns. Which I indeed established for the most part. See this post below for a summary of how my Afro-Diasporan findings (2013-2018) more or less fall in line with historical plausibility.
In the context of my Ghanaian survey it seems that useful clues about underlying ethnicity can be revealed (for Ghanaians themselves!), given correct interpretation and awareness of relevant (pre) history. Even when in fact from Ancestry’s information (see this plot) it can be gleaned that the “Benin/Togo” and “Ivory Coast/Ghana” regions actually cluster the closest from all other African regions on AncestryDNA (2013-2018 version). And furthermore the samples from Benin/Togo and Ivory Coast/Ghana are the only ones that seem to overlap to some degree.
4) To be fair Ancestry’s African breakdown during 2013-2018 was far better in my opinion than 23andme’s African breakdown during that same period (before their update in 2019). Unlike any of its commercial competitors AncestryDNA was able to distinguish no less than 5 different West African regions. In addition it also did a reasonable job at separating West African DNA from Central/Southern African DNA (“Cameroon/Congo” and “Southeastern Bantu”). Which is essential for getting a full scope perspective on the origins of the Afro-Diaspora. See also: