Chad Basin, lying within the bidirectional corridor of African Sahel, is one of the most populated places in Sub-Saharan Africa today. The origin of its settlement appears connected with Holocene climatic ameliorations (aquatic resources) that started ~10,000 years before present (YBP). Although both Nilo-Saharan and Niger-Congo language families are encountered here, the most diversified group is the Chadic branch belonging to the Afro-Asiatic language phylum. In this article, we investigate the proposed ancient migration of Chadic pastoralists from Eastern Africa based on linguistic data and test for genetic traces of this migration in extant Chadic speaking populations.
Results
We performed whole mitochondrial genome sequencing of 16 L3f haplotypes, focused on clade L3f3 that occurs almost exclusively in Chadic speaking people living in the Chad Basin. These data supported the reconstruction of a L3f phylogenetic tree and calculation of times to the most recent common ancestor for all internal clades. A date ~8,000 YBP was estimated for the L3f3 sub-haplogroup, which is in good agreement with the supposed migration of Chadic speaking pastoralists and their linguistic differentiation from other Afro-Asiatic groups of East Africa. As a whole, the Afro-Asiatic language family presents low population structure, as 92.4% of mtDNA variation is found within populations and only 3.4% of variation can be attributed to diversity among language branches. The Chadic speaking populations form a relatively homogenous cluster, exhibiting lower diversification than the other Afro-Asiatic branches (Berber, Semitic and Cushitic).Conclusions
The results of our study support an East African origin of mitochondrial L3f3 clade that is present almost exclusively within Chadic speaking people living in Chad Basin. Whole genome sequence-based dates show that the ancestral haplogroup L3f must have emerged soon after the Out-of-Africa migration (around 57,100 ± 9,400 YBP), but the “Chadic” L3f3 clade has much less internal variation, suggesting an expansion during the Holocene period about 8,000 ± 2,500 YBP. This time period in the Chad Basin is known to have been particularly favourable for the expansion of pastoralists coming from northeastern Africa, as suggested by archaeological, linguistic and climatic data.
Thank you for posting this Igbo. I do feel obliged to point out an impossibility in the text..
According to linguistic analyses of Afro-Asiatic branches, the common ancestors of extant Chadic and Cushitic peoples inhabited East or Northeast Africa ~7,000-8,000 years before present
Because for various reasons to do with goats and sheep there’s no way Cushitic is older than 5,500 BP. The evidence does speak to Chadic being a younger arrival in the area. Is the R1b associated with it from the North?
Another interesting observation (I’ve being trying to get hold of come Chadic Mt DNA) is the amount of L3 in the area. I suspect these are the surviving remnants of the North African L3 branches that got wiped out by the Eurasian back migrations across the Maghreb about 30-35k ago. I shall have another read of this later when my kids aren’t bugging me so much. Sigh.
Mathilda's Anthropology Blog. 
Because for various reasons to do with goats and sheep there’s no way Cushitic is older than 5,500 BP.
That can never be a conclusive evidence, as I have discused before: they can perfectly be loanwords adopted as pastoralism expanded (not necesarily by demic means).
Well according the them it just reinforces an African origin for these families. If Africans from the Sahara or East Africa migrated out of Africa (Which we know they did on many occasions) what languages were they speaking? Going from the assumption of “We dont know” they could POSSIBLY be speaking Proto-AfroAsiatic. Some think they were. IF these people in the Middle East adopted Proto-AfroAsiatic – you could easily see proto-AfroAsiatic words for Animals that are not native to Africa……..Such as sheep and goat.
Likewise WHEN or IF Semitic languages, thought to be one of the younger families was introduced INTO Ethiopia, Ethiopians adopting Semitic some Semitic word EVEN for their own Animals. Even when reading Ehret he statest that at the EARLIEST phase of Proto-Afroasiatic we have words for Donkey. That gives us 2 scenarios.
-People brought proto-Afraisan to Africa.
-Afraisan is native to Africa.
Because of the bulk of the families being IN Africa and also the Genetics of M35 which pretty much ties ALL Afro-Asiatic speakers together it makes more sense that it has an African origin.
Generally ALL populations that speak Afro-Asiatic have a downstream mutation of an East African gene i previously noted as Ethiopic. But not all populations speaking Afro-Asiatic have genes from the Near/Middle East which would most likely be the home of AfroAsiatic IF it was brought from those migrants.
There is a study on Socotra that you should read. I would like to see others comments on it. Socotrans have very high numbers 70% of J*. And some of them have physical affinities similar to Sudanese, Egyptians, Horn Africans and of course Yemenis.
The problem is that there’s only so far back in time you can relate a language family, and the movements of the Y chr are well outside that. There mutation is fromo M78, not its ‘Ethiopian’ ancestor. I’d like to comment here that its’ about 50% possible that the E Y chr also represents the backmigration from Africa. There’s no evidence of DE’s place of origin, it’s possible E itself isn’t African in origin.
the fact the most AA langugaes are in Africa doesn’t really apply. Most IE languages are nowhere near PIE’s point of origin- languages disperse and so far a ‘dispersal with farming’ is offerening a good scenario. I’m not the onbly person thats come to this conclusion after looking at the DNA and archaeology.
The PAA words for goat and a few other very early agricultural terms are common in Indo European, and proto Semitic in particular shows a long proximity to proto Indo European from the start of farming even. An African orrigin for AA languages doesn’t explain how this happened. There’s also no evidence that any African population from before about 24k ago expanded into Asia, or of any kind of cultural expansion either, but you see a lot of traffic from Asia into Africa in the relevant time frame.
Thanks for the study BTW. Please feel free to keep leaving comment- I enjoy constructive arguing.
Please note that Chadic is in fact an exception amongst Afroasiatic speaking populations in how unimportant that E-M35 is amongst them. Indeed they have a very high level of surprising haplotypes like R-M173 and RP25. So if we combine this with this new data should we suggest different ancestries for the dominant male and female lines?
You can get very assymetrical ancestry in colonist groups. The R1b has an entry age compatible with the Neolithic in North Africa, and the expansion of the L3f is close to it.
The other AA speaking (not inc Omotic) groups show a big chunk of Asian ancestry from other sources (J mostly).
Another example of a possible maternal/paternal ancestry contrast is the Berbers.
Omotic has not been much tested, only the Wolayta in Cruciani et al.’s data, but I don’t think they tested them for J. But they show E-M34 and appear to be related to their Amharic neighbours. Also see the Sudan data in Hassan 2008. (Omotic is relatively close to Sudan.)
E-M34 and most J clades can possibly be later introductions which came with Semitic languages into the various African Afro Asiatic areas. Or at least I am not sure this possibility has yet been clearly tested and rejected. One thing which makes me think we might some day be able to prove that Middle Eastern clades dispersed widely very early is the pattern in Iberia.
Omotic- no solid evidence it’s an AA language. I’m sure Ehret claims it’s one to boost the African origin theory of AA languages.
^
I wouldn’t really say that for sure.
The cushtic groups are very very little Haplogroup J.
Oromo in Ethiopia have only 3% J.
Oromo in Kenya – Even less.
Somalis in Somali have also only 3%J
Even North African Berbers that DO have haplogroup J is Characteristic of the Arab expansion.
When you REALLY look at genetic studies of Africa only the Amhara show large genetic input from Neolithic J and a lot of it – 33%. By no coincidence they speak Semitic languages.
And then you have Ongota, they live in Southern Ethiopian and have NO haplogroup J.
People only name the largest groups in Ethiopia but their a LOT of Ethnic groups probably like 85-90 and a LOT of them speak AfroAsiatic.
Even North African Berbers that DO have haplogroup J is Characteristic of the Arab expansion
Not really, J is also characteristic of the Neolithic and Capsian expansions, only some is of Arab ancestry. Depends on the exact clade. One goes back about 10k.
You don’t need vast amounts of genetic input for a language to become dominant. Indo European a prime example-there’s virtually no trace at all. An African origin for PAA just doesn’t explain how proto Semitic was present so close to Indo European languages at the very start of farming, and why the words for so many early agricultural terms in IE are AA based. This couldn’t happen if AA was African- loanwords appear when a concept is new, they don’t roll in after the object is in common use. It places PS in Asia about 9k ago- something no African origin scenario can explain unless you accept a ludicrously old date for it and have the near east as a secondary expansion point in the neolithic. Which I’m not averse too and had in an older blog entry- it’s just theres a real limit to the age of language groups and the African AA origin just needs a way older date than linguistics allows. After 10k you can’t group languages into a family. As an example, PIE and PAA could both come from the same source if the split date was 11k ago- they wouldn’t show more than chance association at that date of separation.
When you look at the dates for Cushitic- they can’t be older than 5500bp because it’s a pastoralists language, and that’s the date that the animals appeared in East Africa. When you use this to ‘calibrate’ the other GC datesfrom Ehret for AA languages they really neatly match the neolthic expansion in Africa.
Not really, J is also characteristic of the Neolithic and Capsian expansions, only some is of Arab ancestry. Depends on the exact clade. One goes back about 10k.
I get a brutally much older date for the expansion/diversification of J, I2 and G in fact (42-32 kya) and I am starting to wonder if J2b non-West Asian spread (found mostly in Europe and South Asia) is not actually Paleolithic (even if it may have also participated in Neolithic expansion as probably did I2a as well). The same I am considering re. G2a. I in general seems an Arignacian remnant but Nordic I1 shows only a very recent expansion after an extremely long coalescence in an unknown location.
These three Y-DNA clades appear as the oldest West Eurasian layer, while R1b (and maybe R1a too) would represent a second wave, maybe associated with Gravettian (many doubts about this last) with an ultimate origin (P, R, R1) further east than West Asia.
My opinion anyhow.
Dating Y chr are not my speciality. I can see logical impossibilities, but the greater detail is beyond me. I’ll take your word for it.
“Omotic- no solid evidence it’s an AA language. I’m sure Ehret claims it’s one to boost the African origin theory of AA languages.”
I know there are serious doubts, but apart from those doubts there is no doubt of strong connections even if “sideways”. My understanding is that even Orel and Stolbova accept that the “Cushomotic” Sprachbund does have common Afroasiatic roots, just hard ones to tease out. See Blench R (2006) Archaeology, Language, and the African Past, Rowman Altamira, ISBN 0759104662, 9780759104662, http://books.google.be/books?id=esFy3Po57A8C
“An African origin for PAA just doesn’t explain how proto Semitic was present so close to Indo European languages at the very start of farming, and why the words for so many early agricultural terms in IE are AA based.”
This remark depends on a circular assumption that Indo European can not be too much younger than PAA. If this assumption may be questioned, then you could consider whether there might have been time for PAA to disperse into the Fertile Crescent and contribute to the local vocab for pastoralism, before IE started to disperse.
Semitic was quite likely just a language from the corner of the Fertile Crescent near Africa which hitch-hiked on a new way of living which was very mobile – pastoralism. For areas where have very early written records we know that it entered the area as a newcomer associated with pastoralists from the less fertile areas around the Fertile Crescent. See Zarins J (1990) Early Pastoral Nomadism and the Settlement of Lower Mesopotamia, Bulletin of the American Schools of Oriental Research, No. 280, pp. 31-65
Depends on your assumed age for PIE. The last node was about 5k ago, but the older family members could have been Anatolian in origin.
“I’d like to comment here that its’ about 50% possible that the E Y chr also represents the backmigration from Africa. There’s no evidence of DE’s place of origin, it’s possible E itself isn’t African in origin.”
I think 50% is an exagerration. Yes, E has East Asian cousins, but ALL Y clades have their eventual ancestry in Africa and ALL of them also have cousins outside Africa. So how can we use this to build a convincing argument about E being non-African?
More importantly, the other hand, only a few quite recent twigs of one particular branch of E-M35 are not found ONLY in Africa, and all these are found in patterns dispersing out of Africa. In other words, there is no controversy I am aware of about E-M35 being African, so it can not logically be used to argue that it’s ancestral clades were non African?
I think I read somewhere that one E* was found in India, and to me that one person represents the entire evidence of E being non African. What are the chances we will one day find at least one D* in Africa and make that evidence more obviously inconsequential?
“Depends on your assumed age for PIE. The last node was about 5k ago, but the older family members could have been Anatolian in origin.”
Sounds about right, but for P-Afroasiatic people estimate ages are out to well above double this. A very wide range of estimates exist. Ehret thinks older than 15kya, and even Militarev is happy to say 10kya. I’d say consensus is certainly older than 5kya.
I’d agree with Militarevs estimates, which aren’t that far of Greenberg for PAA. PAA is on the threshold for reconstructability, so if the ’10k max’ is true for a language family this would put an upper limit on it.
After reading through some of his work- I’m not a big fan of Ehret’s. He’s claiming a pastoralists language with goats etc that predates that arrival of patsoralism and those animials by about a couple of thousand years, which doesn’t make sense. Militarevs recons actually make more sense- the plethora of words for sheep and goats and some other words that are very non hunter gatherer make me very much doubt an ancient age and therefore African origin for AA. To fit in with population movements from Africa, PAA would need to be over 20k old, and that’s just beyond the bounds of reason. Once you exclude a date that ancient, the possibility fo an African origin just goes right dow. You have to explain how PIE and P Semitic were neighbours so early on, and an African origin is hard to shoe horn into that scenario.
I think I’ve already said, I’m an ex-African origin supporter for PAA. I understand the populaiton movements and ancient cultures around North Africa pretty well, and after the Mushabian/Kebaran population came out of Lower Nubia, it’s all been back into Africa. Without a seriously ancient date for PAA, an African origin won’t work, and it seems that language groupings disintegrate at about 10k.
“Dating Y chr are not my speciality. I can see logical impossibilities, but the greater detail is beyond me. I’ll take your word for it.”
Nothing much to point to in the literature yet, but I predict that for the post LGM period age estimates are going down in a fairly big way. But perhaps the bigger message is that the tolerances are enormous sometimes.
Militarev’s views about a near Eastern origin are really not widely held by linguists, and I think it is not a very strong argument to say that having a word for a goat makes a language a “pastoralist language”. Clearly North Africans hunted many animals similar to the ones eventually associated with pastoralism. (It is still not 100% clear that pastoralism is only a introduction from Asia, but let’s put aside the uncertain.)
On the other hand I would also say that Ehret’s ideas about the great age are not the consensus either. I do not see how you can state “To fit in with population movements from Africa, PAA would need to be over 20k old”. Why can’t PAA have entered Asia with the Mushabian or even some later period? There is no reason to believe Semitic was dominant until quite well into proto historical times, much later than the period we are talking about. Languages can be changed by populations with very little change in genetic make-up.
Ehret was claiming that Cushitic was pastoralists because it had words for ‘drive’, cattle, goats etc. I was just pointing out that if he’s using GC to date pastoralism he has to accept the reverse is true, you can’t have goats until they arrive from Asia and we know the dates for that. He’s dated the words for goat and sheep to the same age as cattle, so on his head be it.
I’m pretty familiar with the DNA movements between NE Africa and Asia; after the arrival of the Kebaran in the Levant(ultimately from Nubia) there’s no hint of movement from Africa into Asia. The Mushabians are looking more likely to be either of Negev or Jordanian origin as they’ve found older sites in Jordan now, the Nizzanian culture has been proposed as a parent.
Languages can be changed by populations with very little change in genetic make-up
I think I typed that in myself a while ago. But there’s no evidence of a cultural expansion OOA in the relevant time, but a lot going into Africa.
I think I typed that in myself a while ago. But there’s no evidence of a cultural expansion OOA in the relevant time, but a lot going into Africa.
Not so much in fact: only the concept of agriculture/pastoralism. We don’t see clear migrations, not even major cultural flows from West Asia into Africa. Not enough in my opinion to justify such a massive and diversified hypothetical expansion of AA languages in Africa.
Instead in West Asia you only have Semitic (now and historically – all other known West Asian languages were non-AA) and Semitic could well have originated in the southern Palestinian pastoralist area, in direct contact with Egypt.
Additionally, what do we know from West Arabian prehistory? Not much I think. But West Arabia (the Hedjaz), which is not as extremely desertic as Central/East Arabia could also have acted as coalescence area for proto-Semitic and was necesarily in most direct contact with the eastern Egyptian and Sinai-Negev pastoralists.
I would seriously explore a northern Red Sea area of genesis for Semitic, prior or parallel to the post-PPNB pastoralist complex of the semidesertic areas that is, without almost doubt, at the origin of historical Semitic expansion.
Placing the origin of AA in West Asia is clearly way too forced, so an alternative explanation must exist and, IMO, an “Egyptization” of West Asian pastoralists makes all sense.
Still not buying it Luis- mainly due to the semitic origin of too many PIE agricultural terms.
1. Ehret’s claims about proto Cushitic obviously apply to a much later phase than PAA, so it is not relevant to this discussion. His claims are also more complex. He notes loanwords between early Cushitic and early Nilo-Saharan.
2. I am pretty familiar with the modern data and interpretative literature which people use to speculate about “DNA movements between NE Africa and Asia” “after the arrival of the Kebaran in the Levant (ultimately from Nubia)” and I do not detect anything like the confidence you seem to have. The data simply is not strong enough, and methods for calculating the ages of clades are quite controversial, allowing big variations in estimates.
3. A similar thing can be said about the origins of the Mushabian within archaeology. There are disagreements, and this is partly because the data is not decisive. There are certainly microlithic similarities with nearby cultures along the Mediterranean coast of Africa, and technology such as pottery seems to have an African origin. There is the introduction of Sycamore figs from Africa also. I’d say that the neutral hypothesis is that the Levant has never NOT had strong links with Egypt. Of course that does not mean there was a constant flow of people, but it should put us on guard from getting over-confident with the data we have so far?
Are they agricultural for sure? For example could they have been spread by pastoralists moving around amongst early Farmers after the Neolithic was already well under way? Some of the words I guess you are thinking of also look very much like “wander words” which are found also in language groups very far from any proposed homelands for PIE or PAA? Thanks for the discussion by the way.
… due to the semitic origin of too many PIE agricultural terms.
1. Semitic or pre-Semitic that you think are Semitic only becase you (nor anyone in fact) know nothing about the older layer?
2. Why Semitic and not PAA? Isn’t that totally contradictory? To me it’s like building a theory of IE based on technical neologisms like telephone, astronaut and xilphone, what would place the origin of IE obviously in Greece.
3. I see zero Semitic agricultural neologisms in Basque, even if I may be able to relate one or two words with Greek (ahari – aries for instance) and, if we dig enough, maybe one or two with Berber too (unsure). If agriculture and civilization was so originally “Semitic” (and again why not PAA?) why we do not find such loanwords in other languages? What kind of privileged relationship had PIE and Semitic to make one so dependent on the other (apparently) in terms of Neolithic words?
4. Why do we see the very un-Semitic word “iri” (and similar like uri, uru, ili) all along the Eastern and even Western Mediterranean to mean city/town? From Jerico (the oldest known city anywhere), Jerusalem (where “salem” is the only semitic word in that toponymic) to Ilion, Elis and the very roman term for city “urbs” most likely; from Ur, Uruk and Eridu (and we know that “uru” meant city in Sumerian) to the sooo many Iberian and Basque toponyms in SW Europe? That word is not Semitic at all. It might be Sumerian but Jerico and other East Mediterranean towns like Catal Hoyuk are older than any Sumerian urbanization process. In Basque “hiri” still means city or town.
So, sorry, but I’m not buying it either. Semitic is IMO a secondary layer even in the Levant and must therefore have arrived from somewhere else (i.e. near Egypt/Red Sea). The words that you and others take as “Semitic” are with all likehood pre-Semitic Levantine in fact.
The word uru,iri can be related to the Semitic ‘arya meaning city.