Great read Taranis, thanks.
|Forum||Europe Travel Guide||Facts & Trivia||Genetics||History||Linguistics|
|Eupedia Top > Eupedia Forum & Japan Forum|
Until relatively recently, I was quite convinced myself that Haplogroup E1b1b would be associated with the Proto-Afroasiatic peoples, and that the language family has it's origin in Africa (specifically, the Horn of Africa), around the time of the Last Glacial Maximum. But I'm beginning to have my doubts about that, and I would like to start a debate about these.
The first bit of evidence comes from R1b-V88 (taken from this paper by Cruciani et al.), which is found in the Levante, in the Maghreb, and most importantly, amongst the Chadic-speaking populations of central-western Africa. R1b-V88 is obviously a 'brother' clade to R1b-M297, which is found in Europe (mainly R1b-M269) and Central Asia (mainly R1b-M73). In a nutshell, we have here the first piece of evidence for a back-migration from Eurasia to Africa.
The second bit of evidence comes from lactase persistence amongst the Chadic populations. As pointed out in this paper by Lokki et al. from 2001, the same type of lactase persistence is found in the Chadic populations as it is in Europe, different from the types of lactase persistence as found in other parts of Africa, where lactase persistence probably evolved independently.
So, in a nutshell, we have two separate pieces of evidence for a back migration from Eurasia to Africa, and we have a strong correlation with the Chadic-speaking peoples.
The second question here is where the original homeland of the Proto-Afroasiatic speakers was located. If we look at the modern-day distribution:
If we disregard the dominance of Arabic as a result from the Islamic conquests of the 7th and 8th centuries (as well as the earlier Phoenician colonization in the Mediterranean), the Semitic languages were originally confined to the Middle East. This means four of the five main branches (Berber, Chadic, Egyptian and Kushitic - I'm excluding Omotic here due to it's disputed classification) are found in Africa and only one (Semitic) is found in the Middle East. It would seem more likely to assume due to the greater diversity of Afroasiatic in Africa that the original homeland of the Proto-Afroasiatic peoples was located in Africa. Does that however prove that Proto-Afroasiatic was spoken in Africa? Not necessarily in my opinion.
One critical aspect that connects to this is the question what kind of society the speakers of Proto-Afroasiatic were, and when Proto-Afroasiatic was spoken. If we assume that Proto-Afroasiatic was a hunter-gather language, then the language family could be undoubtably of African origin, as well as considerably older. However, if Proto-Afroasiatic was language of a farmer or pastoralist society, this narrows the setting in which the proto-language could have emerge quite a bit and suggests that instead that Proto-Afroasiatic was a Neolithic language that spread with the advance of agriculture/pastoralism. This scenario makes an African origin much more unlikely and favours an origin in the Middle East. Although I personally find that one must maintain some scepticism (notably due to his support for the Nostratic language family), Militarev (2009) makes a strong case in this paper for Proto-Afroasiatic as a language of a pastoralist society. Evidence also comes from individual branches of Afroasiatic, for instance Proto-Berber and Proto-Semitic were both Neolithic languages. The former is a particularly instructive example because the Mesolithic Capsian Culture of North Africa has often been suggested as speakers of Proto-Berber:
The problem with this idea however is that the Capsian Culture was a Mesolithic hunter-gatherer culture, and not a pastoralist society. It is thus impossible for the Capsian people to have been speakers of Proto-Berber. They may have been speakers of an earlier form of Proto-Afroasiatic if Proto-Afroasiatic was indeed a Mesolithic language. As presented above, I have my doubts about this, due to the fact that terms for domesticated animals that are found in Proto-Berber can also be reconstructed for Proto-Afroasiatic.
If we go with the Eurasian origin hypothesis for Proto-Afroasiatic, then this raises the question what Y-Haplogroups might have been spread with the spread of the Afroasiatic languages, since both the spread pattern and the age of E1b1b does not match such a scenario. If we take the evidence of the Chadic peoples, R1b-V88 was certainly one of the Haplogroups, but it would be highly doubtful that it was the only one. The next most obvious choice, in my opinion, that potentially fits an expansion pattern for the Afroasiatic languages is Haplogroup J1. This Haplogroup is usually associated with the spread of the Semitic languages, and considering the distribution (in particular high concentrations in the Caucasus) it is pretty clear that J1 as a whole cannot be Semitic, and that probably only the subclade J1c3 (aka J-P58) is genuinely associated with the spread of the Semitic languages. I'm not wholly convinced, however: there is this paper by Fregel et al. 2009 about the Y-Haplogroups of the Guanches (the aboriginals of the Canary Isles, which are thought to have spoken a relative of the Berber languages, hence Afroasiatic), and it would appear that Haplogroup J1 was present amongst the ancient Guanches. I'm not sure if this is genuine evidence that J1 is indeed Afroasiatic (especially, I think it remains to be discussed which subclades really are involved), it certainly supports the possibility of an out-of-the-Middle East migration, and is consistent with the general scenario I proposed here.
So, to summarize, I (carefully) would propose that both J1 (perhaps J1c3?) and R1b-V88 (the latter, I would argue, certainly holds true for the Proto-Chadic peoples) may have been associated with the spread of the Afroasiatic languages (if anybody has a better explanation, I would love to hear it), and that the Proto-Afroasiatic homeland lies in the Middle East, and not in Africa.
Last edited by Taranis; 06-03-12 at 15:05.
Great read Taranis, thanks.
It seems that there is more to R1b-V88, indeed Cruciani et al's reply to Andrew Lancaster proves to be a quite entertaining read.
We thus read:
"The second observation was regarding a genetic contiguity between the Chadic-speaking peoples from the Central Sahel and several other Afro-Asiatic-speaking groups from North Africa, including Ouarzazate Berbers from Morocco, Mozabite Berbers from Algeria, Siwa Berbers and several Semitic groups from Egypt, and, possibly, different groups from Algeria, Tunisia and Egypt, with R1b1a frequencies ranging from 1 to 3% in Algeria to about 4% in Tunisia, to 26.9% in the Siwa. We interpreted these data by suggesting that they are more compatible with Ehret's hypothesis, which proposes that Chadic peoples arrived from the North through the Sahara (the ‘trans-Saharan’ hypothesis), rather than with Blench's theory, which states that Chadic-speaking pastoralists reached the Chad Basin through the Sahel from an eastern Sudanic Cushitic-Chadic motherland (the ‘inter-Saharan’ hypothesis)."
I find the 26.9% frequency amongst Siwa Berbers to be quite telling, contextually-speaking, as this scheme tends to agree with Ehret's subgrouping fitting neatly with his "North Erythrean" subgroup (the same counts for Militarev's "North Afrasian" subgroup).
By itself, this fact undermines Keita's (and, to a lesser extent, Ehret's) unilateral approach since it implies that this subgroup is not solely related to M35 and its subclades. V88 also scores up to 40% of Shuwa Arab samples.
I also recall Hassan et al finding J1 amongst Copts and Beja (39 and 36% respectively), also worthy of mention is Chiaroni et al where we find that 2.6% of Oromo samples along with 8.3% of Amharic samples are J1*.
Wether these have more to do with the spread of Cushitic than that of Semitic is purely out of reach for now.
Only one thing is certain, Afroasiatic J1 will be interesting to study.
If anything, data suggest that M35 wasn't the only marker involved in the spread of "North Erythrean"/"North Afrasian" yet alone that of Afroasiatic as a whole (if J1* really turns out to be involved with Cushitic, then the whole subgrouping has to be revised unless the urheimat itself is faulty).
Time will tell.
I've been giving some thought on this:
1) You pointed out that the DNA evidence is compatible with several scenarios on the relationship of the Afroasiatic languages. It should be pointed out that a recurring scenario (most notably Ehret's "North Erythrean" and Militarev's "North Afrasian") is a 'northern' grouping that includes Berber, Chadic, Egyptian and Semitic. Even Ehret concedes that a common farming vocabulary can be constructed for sub-groupings of Afroasiatic, and if we combine this with the genetic evidence, it would make sense to suggest that the spread of the "northern" Afroasiatic languages was indeed the result of a Neolithic spread, which is reflected today by the distribution of Y-Haplorgoup R1b-V88 in Africa.
2) If there is genuine Cushitic J1 (which remains to be seen), we must likewise assume that the ancestors of Cushitic arrived from the Middle East, and in my opinion a wholly Middle Eastern homeland may be warranted then.
The Chadic project, which is no longer viewable was about 43/57 R1b1c/E1b1(E1b1a, E-V22) (0% J1e), and notable is the appreciable presence of R1b among all Jewish groups, and other AA speakers such as Assyrian and other isolate regions in the Levant. Consider that language could be spread by elite dominance or mass migration, or even through women only. Nothing can be ruled out for certain.
EDIT: According to Sumerians scripts, the language changed suddenly via a elite group rather than mass re-population. This is more tangible evidence than any Indo-European linguistic evidence in Europe.
2) In a sense and in relation with what was said above, we are better off going by lineages such as J1 and T(1a) for the moment, since these are two Eurasian, not to say West Asian, lineages which are found in considerable frequency amongst AA speaking groups. At least, as far as J1 is concerned, we do have genuine archeogenetic evidence for the presence of this marker among antique Guanches.
If anything, this issue has not received enough attention while extensive archeogenetic work is needed in order to answer the most basic questions.
Anything else is mere speculation.