I’ve been having a rethink about Afro Asiatic origins

I’m having a rethink about Afro Asiatic’s origin after having a good look at the reconstructed nouns.

Particularly those dealing with with animals. I had a brief look through the nouns for PAA, and quite striking was the number of words for goats and sheep. Also included were horses  and camels. Since goats, horses and sheep and camels were not native to Holocene Africa prior to the neolithic, I’m reconsidering my support of an African origin for proto Afro Asiatic. Although, as has been kindly pointed out, the reconstructions are all pretty hazy for PAA, but still it’s suspicious.

Another factor making me reconsider is the dating suggested for the languages.  The presence of goats and sheep (many and varied terms) also gives an oldest possible date to the last node  (a languages TMRCA) for Cushitic, which is a pastoral language of sheep, goat and cattle herders. Since Cushitic is sub Saharan, very relevant is the oldest known date for the arrival of ovicaprines in the Sudan, which is about 5,500 years BP ( Esh Shaheinab, Sudan). This would suggest the proposed 10k date for proto Cushitic is off by about 45%- although this may just be it’s last node and the 10k date for it’s seperation may be correct. 

Relevant to this is the R1b Y chromosome present in the Ouldeme and the Hausa, both Chadic speaking groups, one in Cameroon and one in the Sudan. The Hausa have R1b ( R-P25* (R1b1*) at about 41%, and Ouldeme at 95%. This is quite a bizarre find for groups in the middle of Africa, as R1b is typically European and West Asian. It would be a logical suggestion that the Ouldeme and Hausa are quite closely related paternally, and may point to an East to West route for Chadic speakers- suggested by Blench in the ‘The Westward wanderings of Cushitic Pastoralists’- although there have been suggestions the Hausa moved from West to east recently, which would make the R1b in Cameroon possibly from  a north to south route across the Sahara.

This particular branch of R1b has been dated to an entry of about 4,000 years ago- but bearing in mind the older (2002) papers tend to seriously underestimate the date of the Y chromosomes – a pet peeve- the oldest entry date for it at 8,000 BP would be more reasonable, and a good match for the Neolithic sheep and goat pastoralists arriving in Africa from West Asia. It doesn’t do my older theory of M78/M1 being linked to the spread of Afro Asiatic any good though. Oh well.

The coalescence age of the African haplotype 117, which we estimated as 4,100 years (95% CI 2,400–8,060 years), could thus represent a date for such an expansion and a lower limit for the time of entry into Africa.

From this paper.

This all has some relevance to Ehrets dating of Proto Nilo Saharan (both families dated by glottochronology). He gives the same 15k date for Nilo Saharan as for proto Afro Asiatic.. so I’m thinking 10-9,000 bp for Nilo Saharan too. This also brings proto Northern Sudanic into the outer estimate for the Neolithic in Africa (7,000) although it’s unlikely as they have a dearth of terms for pastoralism and agriculture. His dates seem to vary from 35% to 45%  off the possible, which may be due to the difference in geographical points of origin in proto Cushitic and Proto Sahelian, so I’m assuming proto Sahelian is a little more Northerly in origin than proto Cushitic and have adjusted the dates for it  for a ‘best fit’. Even if it does give a close date for age of separation fro the sub groups, Ehret never seems to take into account there may have been more recent nodes to account for the pastoralist terms.

This doesn’t really support Omotic as an afro Asiatic language, as it shows no proto words for pastoralism before it’s split. But it has been pointed out by several linguists that it has no more in common with Afro Asiatic than it does with it’s other neighbouring language groups, so it’s AA status is pretty suspect to start with.

Edit:

A little more DNA evidence has come out showing a pre Neolithic population movement into North and East Africa  dating to 11-10k ago, involving J1 (Y) and H (mt DNA) which coincide with the IM/Capsian transition in North Africa.  This could be the reason for the odd structure of the tree; Cushitic languages are the result of an earlier AA population expansion into East Africa from the near East. This expansion (as far as I can tell) seems to start about 13,500 BP from southern Turkey?  I’ll need to dig into it a bit more. This cultural expansion may have been of a food ‘managing’ culture as opposed to food gathering or producing cultures, a proto Neolithic expansion wave of people that kept wild animals (a domestication step) and harvested and planted seeds from the wild. There are domesticated seeds from Syria at 12,500 BP so the people of the Turkey/near East area were definitely doing something along those lines at the right date.

13 responses to “I’ve been having a rethink about Afro Asiatic origins

  1. Well, as you know, I think that the words for horse and camel are loanwords like telephone or cat, whose expansion does not relate directly with the expansion of the linguistic families but rather with other cultural flows (sprachbund or whatever). When you borrow a new item from another cultural group it is very likely that you will borrow also the name.

    It doesn’t do my older theory of M78/M1 being linked to the spread of Afro Asiatic any good though. Oh well.

    In fact I think that it does indirectly support it, as it does suggest again a Sudanese or Egyptian origin for Afroasiatic. Egyptian/Sudanese R1b would seem to be at least partly R1b1b2-M269 (if it’s all like Egyptian R1b, about 50% is M269, while the other 50% would be R1b*, possibly the source of the Ouldeme and Hausa R1b* – ref, discussion at Dienekes).

    What you should not exclude is that different migrations may have had different founder effects, picked from that same genetic pool of the middle Nile. In many cases E-M78 was the main associated Y-DNA but in others was R1b or, in the case of Semitic, J1.

    What I wonder is how “European” R1b ended up in Sudan (and NW Africa). Along with other data (mtDNA U6, H and V especially) it does suggest me that Iberomaurusian honors its name (i.e. it’s derived from Iberian Gravetto-Solutrean). Of course, taken alone, it can’t be totally excluded that Egyptian-Sudanese R1b arrived from West Asia (but R1b is rare in West Asia if we except Anatolia), but when I look at the whole picture…

    • I had a (brief) nose throgh proto Berber. The theory is camel and horse really only date to its last ‘node’ but it’s been knocking about a lot longer diversifying internally then having one later dialect (with camel and horse) becoming dominant.

      Proto words and later loan words don’t show as much variety as each other, as I recall.

      well, the entry of the Ouldeme R1b is only at 8k, not massively old, and it does suggest a Neolithic Eurasian movement down the Nile that matches the spread of Chadic. You do get R1b in Asia, a samll R1b conntribution plus founder effect in a bottleneck I think explains this better.

  2. “places proto-Erythraic at 7,700 BP … not the vastly ancient hunter gatherer language proposed by Ehret”.

    There has always been an obvious problem for an ancient date. We can be fairly sure from looking at other language groups that if a connection between languages is discernable the languages are unlikely to have diversified anything like 10,000 years old. I guess there’s a wish by some people to give Hebrew an ancient origin, thus requiring its relations to be extremely ancient.

    • You know, I remember reading that thing about the 10k date before somewhere. Can you give me a source? It would be handy to read.

  3. Ahh here..

    After a thousand years, the original and new languages will not be mutually intelligible. After ten thousand years, the relationship will be essentially indistinguishable from chance relationships between historically unrelated languages.

    and here
    International handbook of bilingualism and bilingual education By Christina Brat

    the methods fpr establishing branching family relationships, like the method for reconstructing unattested stages in the history of a language family, break down at about 10,000 years.

    I wonder how they worked that out.

  4. I wonder how they worked that out.

    Basically working with language families, like Indoeuropean and Afroasiatic, and trying to link stuff.

    If my memory is correct is an aphorism from Starotsin, but very true it seems. At greater depths connections become ghostly and impossible to discern from the noise. Languages just change too fast and too unsystematically to make sense when the noise becomes too srong and the signal too weak.

    Of course there are the ones who try to overcome such barriers but their proposals are more and more getting cold receptions.

    • Just a suggestion- if linguistic relationships vaporize at 10k, is there any reason that Afro asaitic and the early PIE branch can’t have come from the same family? They are both showing at an acceptably close date (9k vs10k), with PAA right on the ede of being unreconstructable. They could both have come from the same early neolithic founder group about 11k ago; with that kind of sepergation in time you’d never know.

      I’d love to know how they worked it out though. Known migration dates vs level of similarity to parent culture in reasonably isolated island groups would be my choice to work that out. Got to say the 10k max date for language families isn’t doing what remains of my confidence in Ehrets work any favours at all.

  5. Luis – Great links. I was always under the impression that the R1* was from much older Ancient migration into Africa, Oman, Dead Sea and could be related [Paragroups are not always related.] while the other R1b with more downstream mutations was due to later migrations or Possible admixture from Turkish occupation?
    R-173 (R1*) – Was most frequent in the Fulani – They are actually a Niger-Congo speaking group.
    While the downstream makers R-P25(R1b) Was highest in Hausa……..and how do they correlate this with a sickle cell gene?……..Interesting history. Also interesting is that Fulani and Hausa are found all they way to the West of Africa too.

    I think one thing that we forget is the People that help formed Afro-Asiatic or Proto Afro-Asiatic may not be from the place where the language disbursed. Or they could simple be loan words but i always thought that People from the Middle East (Yemen) came into Ethiopia a long long time ago – like 8,000-10,000 years or something.

    mathilda37 – That is not a new theory but it can make sense. There is a presentation in which Shomarka Keita presents that idea and speaks upon it. In that case we should look to the Russians linguists and their idea of Nostratic:
    Indo-European, Afro-Asiatic, Altaic, and Indo-Dravidian are ALL related and they broke off and spread with Food Producing and went their own ways from Turkey.
    In that case i think that the roots of Proto-Afrasian being non-food producing as well as the Genetic Evidence just doesn’t work out for that theory. Some other people came about and said that Afro-Asiatic could possibly be a sister language family of Nostratic but not a daughter family. That takes us all the way back to OOA though.,

  6. “is there any reason that Afro asaitic and the early PIE branch can’t have come from the same family?”

    Possibly. But there’s a problem. Indo-European’s closest relation seems to be the northern group, Ugric. And that language is sometimes claimed as a relation of Yukagir in Northeast Eurasia. So Indo-European is probably closer to northern languages rather than Middle East ones. However the languges may all have been part of an originally widespread group, perhaps we could call it Nostratic.

    “That takes us all the way back to OOA though”.

    If it’s impossible to trace language connections back beyond 10,000 years we have no way of following language development back to OOA. Especially seeing that we have to include Australian Aboriginal and Papuan languages in any such reconstruction.

    • Possibly. But there’s a problem. Indo-European’s closest relation seems to be the northern group, Ugric. And that language is sometimes claimed as a relation of Yukagir in Northeast Eurasia. So Indo-European is probably closer to northern languages rather than Middle East ones. However the languges may all have been part of an originally widespread group, perhaps we could call it Nostratic.

      It depends.. if PIE’s affinities to Ugric come from the 5,500 ish era, it will still be okay if you accept that the original POO was in Anatolia 9k ago- it could have pick up it’s Ugric relationship in the later Steppes era.

  7. @ Mathilda:

    …if linguistic relationships vaporize at 10k, is there any reason that Afro asaitic and the early PIE branch can’t have come from the same family? They are both showing at an acceptably close date (9k vs10k), with PAA right on the ede of being unreconstructable. They could both have come from the same early neolithic founder group about 11k ago; with that kind of sepergation in time you’d never know.

    Actually the most commonly managed date for PIE is like 6000 years ago, not 10,000.

    But sure: there is no particular reason not to be related but the same that there is no particular reason for either not to be realted with Sinitic, Austroasiatic, Niger-Congo, Tupi-Guarani or NE Caucasian.

    Only that if you think of IE as coalescing in the Eurasian steppe, maybe with a south or central Asian background (recent analysis of R1a suggest a South Asian ultimate origin, long before its great expansion with IE migrations), and AA as coalsecing in the Nile valley, probably with a tropical African background (haplogroup E and derived)… then the likehood of such correlation appears much less likely.

    @Igbo:

    I was always under the impression that the R1* was from much older Ancient migration into Africa, Oman, Dead Sea and could be related [Paragroups are not always related.] while the other R1b with more downstream mutations was due to later migrations or Possible admixture from Turkish occupation?
    R-173 (R1*) – Was most frequent in the Fulani – They are actually a Niger-Congo speaking group.
    While the downstream makers R-P25(R1b) Was highest in Hausa……..and how do they correlate this with a sickle cell gene?……..Interesting history. Also interesting is that Fulani and Hausa are found all they way to the West of Africa too.

    Hard to say. R1 is generally believed to have coalesced either in Central or South Asia (R1a and R2, as well as some R1* are found in South Asia and appear old there), while R1b would be West Eurasian. Though indeed the African branches add many interrogants here.

    I imagine that the Fulani R1* is not the direct ancestor of R1b (or R1a) but either an ill studied sister sublcade most likely (“R1c” possibly) or just more R1b that has not been properly tested. Similarly it’d be nice to clarify what does Sahelian R1b* actually is within the context of R1b and R1. It’s of interest, I understand, not just to Africans but to West Eurasians as well.

    @Terry:

    Nostratic hypothesis seems quite dead and at best limited to the core group. Indo-Uralic instead survives but the question is wether it’s a genetic relationship (sister languages) or a sprachbund one (just good neighbours).

    Sometimes I wonder if Indoeuropean isn’t but Uralic spoken by non-Uralics (and hence deformed) but the inverse could also be true. If it’s confirmed that R1a oldest origins are in South Asia (even if most of he expansion would be Kurgan at later date) maybe Hittite and Tocharian (and who knows if even Albanian and Armenian) could have evolved from a proto-PIE instead of PIE-Kurgan with a more diffuse origin between Easternmost Europe and Southern Asia. And maybe the possible relatives of IE should be sought in Burusho and the like (Dravidian?)

    All very speculative anyhow. In any case, a nice site to look for possible unexpected connections in Numbers in over 5000 languages where you can try to find connections between any languages or language families on Earth by comparing their numbers. It was that way that I decided that Sumerian, Hurro-Urartean and NE Caucasic might be related and represent the remnants of an Eastern Gravettian language family, maybe with Basque (representing Western Gravettian?) as their closest yet very distant cousin. Very very tentative, I know. But it’s said that number names change only very slowly… who knows?

  8. New study on Chadic speakers and mtdna.

    Migration of Chadic speaking pastoralists within Africa based on population structure of Chad Basin and phylogeography of mitochondrial L3f haplogroup

    Background
    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.

  9. Marnie Dunsmore

    Bagpipe-R1b Theory

    The Bagpipe, strongly differentiated and of ancient origin, most likely originated in SW Asia. It is intimately linked with a pastoral sheep/goat/cattle herding people who also carry an R1b haplogroup signature.

    The inseparability of R1b pastoralism and the bagpipe suggest that the instrument fulfilled some essential herding function.

    The presence of the bagpipe might be used to narrow in on R1b populations for the purpose of haplogroup evolution testing.

    Significantly, the Hausa people of Tropical West Africa play a bagpipe, are a pastoral people and carry the R1b haplogroup in greater than 40% of men.

    Differentiation and innovation in bagpipe style, from primitive to finely developed, may also correlate with a westward migration from the cradle of R1b pastoralism.

    Copyright (c) 2010 by Marnie Dunsmore. All Rights Reserved.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s