The mixed genetic origin of the first farmers of Europe

**Keneki20** · 11-27-2020, 01:48 AM

Originally Posted by Ramses

You didn't add much with that whole comment it seems you use too many words to tell very little.

Or, maybe you were committed to misunderstanding the text. Either way, I'll oblige and use fewer words.

1-your first assertion is not true ,(Not all afro-asiatic communities appear to carry african ancestry"
Chadic and cushitic speakers do , non-muslim Semitic speakers like Mandeans and Assyrians don't have any
and muslim speaking arabic in the Levant carry small african ancestry that is dated to the arab slave trade , purer arabic speaking groups like Bedouins don't show any significant african ancestry.
- most of the african ancestry found in Egyptians is dated to the slave trade period very small percentage is found old egyptians.
- the african ancestry in modern berbers is also mostly dated to the slave trade period while the old african ancestry in them is more related to the deep ancestry found in the Iberomaurosians and are irrelevant here.

One of my later comments clarifies this further for Berber and Semitic speakers. I can't say I agree with your other points, though. Additionally, with the logic you rely on, I can say that it's unreasonable to show an association of Natufian-like ancestry with Afro-Asiatic, because the vast majority of Chadic speakers show negligible West Eurasian autosomal ancestry. If it makes you feel better, though, I don't believe that Ancient Egyptians, be they northern or southern Egyptian, have ever had even close to a majority of Sub-Saharan African-related ancestry, and I believe it's quite clear that some of that Sub-Saharan African ancestry is due to Trans-Saharan movements. Nevertheless, I am unmoved in my opinions and unconvinced of the veracity of your stance.

First Natufians didn't carry E-M78 !

Ah, that appears to be true. Unfortunately, I misread the results of the study I was using. Thanks for pointing that out. I'll make the appropriate edit.

...the second point is not true Linguistic data does not do that most modern Linguists take Genetic data specially Ancient DNA very seriously , old linguistis were very restricted and there inferences are impossible to be as accurate or credible as Genetic data most of there hypotheses is controversial and not accurate also it depends on Choice and personal opinions of the researches very often biased by distribution and number of speakers etc.. while DNA is not its there actual blood.
anyway Afro-asiatic origin had two main hypotheses promoted by Linguistis either the Levant or East africa , current aDNA evidences support and "approves" the Levantine origin.
Section 3. Origins of Afro-Asiatic.

I never said linguists don't take genetic data seriously. I said that linguistic data take primacy. That is quite established. The Steppe Hypothesis of Indo-European was first posited without the wealth of genetic data geneticists now have on early Steppe pastoralists, for example. Also, indeed, personal opinions can yield confounding results. Although, I have yet to see you provide any meaningful counterargument to the information I provided, so I hope that you understand my lack of credulity.

Also, for examples of when linguistic and genetic data can be at odds regarding the identity of a language’s earliest speakers or a language family's homeland, I suggest you look at most Southern African Khoe-Kwadi speakers, Malagasy speakers, the various Austronesian peoples of Vanuatu and neighboring islands in Melanesia, Uralic speakers, Basque speakers, and modern Nubians amongst many other examples.

The Capsian culture territory extended of most of the ancient Sahara..and they ARE associated with with earliest afro-asiatic speakers in africa.

I would like to see unequivocal proof of that, especially in light of what I said, which was that they cannot be associated with the proto-Afro-Asiatic speakers due to dating and lifestyle. I'll wait.

mentioning the life style of sudanese hunter gatherers is meaningless , since they are not associated with Afro-asiatic.

So you claim, but have yet to earnestly demonstrate.

The rest of your comments is not very clear you seem confused. Its like a comments that doesn't make any meaningful point.

That's rather presumptuous of you. I wouldn't say I'm confusing anyone who reads my post. On the other hand, it's far easier for me to suppose you're externalizing.

you reversed the question, There's no evidences that Mota spoke any afro-asiatic languages. Afro-asiatic in horn africa is associated archeologically with Pastoral Neolithic.
Mota (pre-omotic) most likely spoke language similar to Hadza which is unidentified clicking language.

While I'm not convinced of the Mota individual speaking an Afro-Asiatic language (truthfully, how could one know with so little information at the moment?), I have to say, your hypothesis that Mota spoke a "language similar to Hadza" is quite novel and conspicuously lacking in any particular linguistic justification. Maybe you can enlighten the rest of us?

**TuaMan** · 11-27-2020, 02:18 AM

Originally Posted by Keneki20

Lastly, a 2014 study titled Ancient DNA Analysis of 8000 B.C. Near Eastern Farmers Supports an Early Neolithic Pioneer Maritime Colonization of Mainland Europe through Cyprus and the Aegean Islands noted that in one of its Pre-Pottery Neolithic B sampled individuals (labeled H8) from Tell Halula, Syria dated to between 6800 and 6000 B.C.E., researches detected mtDNA haplogroup L3 (not inclusive of more Eurasian-associated mtDNA haplogroups like N and H). It is identified specifically as L3, then that speaks to a far more recent African origin of the haplogroup identified.

Additionally, in some posts I have seen from Anthrogenica, haplogroup L2a1 has been identified in the same place site and some other slightly younger sites in addition to L3. I haven't been able to identify the study that specifically mentions L2a1, but if it indeed is readily verifiable, then that only buttresses the aforementioned information, while, interestingly, noting a link. The genetic results of the pastoralist individual from Luxmanda, Tanzania yieled L2a1 as well (maybe not under the same subbranch).

I'm not 100% sure, but I think the L2a1 sample you're referencing was in fact one in the same as the L3 - that sample was originally sequenced as L2a1 but then later refined to just L3 (and I don't know as much of about the nuances of mtDNA testing, if the sample was just poor quality and the L3 just represented the highest order of resolution they could get out of it, as opposed to actually being a proper L3 call). I believe Jean Manco (RIP) clarified this on here years ago.

**Helves** · 11-27-2020, 02:19 AM

Originally Posted by Keneki20

I hope this all makes sense.

It most certainly don't. You just wrote an essay of wishful thoughts. You link to studies that date the intrusion of SSA admixture into West Asia well after the period of proto-Semitic. And you fail to include this as an example from the 2011 study

Levantine groups harbor 4%–15% African ancestry with an average mixture date of about 32 generations ago, consistent with close political, economic, and cultural links with Egypt in the late middle ages

The 2020 Agranat-Tamir et al study modelles modern West Asians as a mixture of just Megiddo_MLBA, Europe_LNBA and Somalis. Nedless to say how this can't be taken any serious at all. Modelling Saudis with a BA Levantine sample as the only Natufian rich source is a joke. Yet we do see how the only non-Muslim sample, Iranian Jews not scoring any Somali at all, but in the 2011 study Iranian Jews are supposedly 4.6% SSA?
Even if the L3 is correct I doubt PPN samples from Northern Syria will show any SSA when contemporary samples from Jordan and Israel have not. Not that these samples have much to do with proto-Semitics.

The Slave trade had an impact on the Muslims, we know that. And the SSA admixture that was brought seems to be mainly East African.

When modelling Berbers with ancient samples + Dinka and Yoruba, the East African is non-existent or at noise levels. Natufian on the other hand is not at noise levels.

The ancient Egyptian samples show hardly any Dinka admixture either.

So what are we dealing with here? Was the Mota like admixture already watered down to zero prior to the Berber-Semitic-Cushitic-Egyptian split?

**drobbah** · 11-27-2020, 02:32 AM

Originally Posted by Helves

From what I've gathered from Aga, the Semitic-Berber branch split off around 5000BC and the pre-proto-Semitic speakers must've left(from the Nile valley) for the Levant shortly after that yet I don't remember the papers on the EBA samples from Jordan or Israel(certainly early Semitics) mentioning any Mota or Dinka-like admixture in these ancient samples and they had significantly less Anatolia/Iran/CHG than modern Levantines.
How much AEA ancestry to Copts and Berbers have today? Or the ancient Egyptian samples we have so far?

We have evidence of L Sub-Saharan lineages in both Levant Neolithic and a L0f sample in the bronze age Levant aswell.Also those EBA sample don't represent the earliest Semitic speakers from Africa and those egyptian samples are too young.Modern Berbers are an highly admixed population and don't represent imo the earliest Berber speakers who came from NE Africa.

I'm only speculating like those who claim proto-AA speakers were pristine West Eurasians with Basal admixture, we can't rule out the possibility of possibly AEA admixed populations being some of the earliest proto-AA speakers.We have examples like Jebel Sahaba where these two very different populations lived side by side & sometimes went to war.If we were to take the Early Pastoralist cushitic samples and take away their high Mota ancestry which they got when they entered the Horn.They would be 70% (Levant_N + IBM) & 30% Dinka like.

**davit** · 11-27-2020, 02:34 AM

Originally Posted by Helves

It most certainly don't. You just wrote an essay of wishful thoughts. You link to studies that date the intrusion of SSA admixture into West Asia well after the period of proto-Semitic. And you fail to include this as an example from the 2011 study

The 2020 Agranat-Tamir et al study modelles modern West Asians as a mixture of just Megiddo_MLBA, Europe_LNBA and Somalis. Nedless to say how this can't be taken any serious at all. Modelling Saudis with a BA Levantine sample as the only Natufian rich source is a joke. Yet we do see how the only non-Muslim sample, Iranian Jews not scoring any Somali at all, but in the 2011 study Iranian Jews are supposedly 4.6% SSA?
Even if the L3 is correct I doubt PPN samples from Northern Syria will show any SSA when contemporary samples from Jordan and Israel have not. Not that these samples have much to do with proto-Semitics.

The Slave trade had an impact on the Muslims, we know that. And the SSA admixture that was brought seems to be mainly East African.

When modelling Berbers with ancient samples + Dinka and Yoruba, the East African is non-existent or at noise levels. Natufian on the other hand is not at noise levels.

The ancient Egyptian samples show hardly any Dinka admixture either.

So what are we dealing with here? Was the Mota like admixture already watered down to zero prior to the Berber-Semitic-Cushitic-Egyptian split?

Aren't Somalis also a bad proxy for AEA ancestry given their West Eurasian admixture?

**Helves** · 11-27-2020, 02:53 AM

Originally Posted by drobbah

We have evidence of L Sub-Saharan lineages in both Levant Neolithic and a L0f sample in the bronze age Levant aswell.Also those EBA sample don't represent the earliest Semitic speakers from Africa and those egyptian samples are too young.Modern Berbers are an highly admixed population and don't represent imo the earliest Berber speakers who came from NE Africa.

I'm only speculating like those who claim proto-AA speakers were pristine West Eurasians with Basal admixture, we can't rule out the possibility of possibly AEA admixed populations being some of the earliest proto-AA speakers.We have examples like Jebel Sahaba where these two very different populations lived side by side & sometimes went to war.If we were to take the Early Pastoralist cushitic samples and take away their high Mota ancestry which they got when they entered the Horn.They would be 70% (Levant_N + IBM) & 30% Dinka like.

There's also SSA haplogroups in Copper Age and Bronze Age Iberia, mtdna is gonna trickle down between neighbouring regions. Again no Mota-like admixture in the PPN samples published so far. But what is this L0f BA sample? Has it been dated?

The Egyptian samples are too late, as are the EBA samples from Jordan. Modern day Berbers are too admixed, I suppose that's the case for Copts aswell. Okay then, I guess we are just full on speculating and ignoring what's in front of us.

**Granary** · 11-27-2020, 04:56 AM

Originally Posted by Riverman

Stuff

My understanding is that the Nile was not really occupied throughout the African Humid period and this in of itself couldn't have played a big role for millennia, my understanding is that during the pre humid period the Nubian Nile valley was densely populated, during the humid period the Nile itself was abandoned or underpopulated given it was of less use and was harder to access to hunter-gatherers and in the post humid period the Nile became a refugium, how can the linguistic development fit into this?

In my view the beginning of the Humid period probably saw an explosion in linguistic diversity as people from populated areas on the Nile and elsewhere populated the grassland, I'm not sure exactly what archeology says about the origin of the people dwelling on the grasslands, but I imagine Afro-Asiatic speakers were at least widespread in the Eastern half of the new grasslands.
Also if Chadic-Egyptian is a valid node I imagine that the end of the humid period could have been the event that separated the 2 branches(whose speakers would have dwelled in the Western desert in Egypt and Lybia).

**Keneki20** · 11-27-2020, 05:08 AM

Originally Posted by dmana

Grouping some subgroup of Omotic with Cushitic makes no sense, but north and south Omotic could still be two different language families. Maybe both of them belong to AA, or none of them, or just one of them. If "Omotic" really was a valid node within AA then it would be easy to list the common innovations that exist exclusively in all branches of Omotic and no other AA group. It would be even more convincing if they could reconstruct proto-Omotic and "proto-core-AA" (AA minus Omotic) and show how proto-AA could've evolved into both branches through regular changes. But they can't even do that, which is all more the reason to be skeptic of statements about how the inclusion of Omotic into AA is a consensus, specially when there still is skepticism on the matter among some AA linguists.

Those are completely understandable reservations to me. That's partly why I said at the end that I personally see Omotic as promising, and that I need more information on South Omotic languages, which have less clear Afro-Asiatic parallels. I have seen the proposition of regarding Omotic as a genealogical pool rather than a distinct lineage, which I believe is also reasonable. However, there are some features that both, for example, North Omotic and South Omotic share that, I believe, can make one hesitate somewhat at that proposition.

For instance, on the phonological end, a shared isogloss is sibilant harmony, which Hayward suggests can be reconstructed for his idea of Proto-Omotic. Then, for Omotic languages, there is the formal marking of imperative and declarative moods, which is a comparatively unusual feature. Of course, features can be cross-linguistic, such as how ATR vowel harmony, prenasalized consonants, and labial-velar consonants cut across some lineages in Sub-Saharan Africa that can be rather different. However, that would be more significant if these languages were more different than they generally are. For instance, for Wolof and Fula, I'm not convinced of their being related yet, but they share initial consonant mutations, which is an ostensibly significant phonological isogloss.

However, reliance on the construction of a proto-language is not necessarily a limiting factor. Specifically Proto-Chadic reconstructions are apparently no more than about 130, and not much is known about the Proto-Chadic language at large. The situation for Proto-Cushitic and even Proto-Berber is rather similar, with some even calling into question if Beja, for example, should be considered Cushitic. Outside the Afro-Asiatic domain, the Proto-Sino-Tibetan language is also not satisfactorily reconstructed, even though the unity of Sino-Tibetan is generally not called into question.

Of course, Omotic's unity with Afro-Asiatic is not exactly straightforward, but neither is the opinion that it is an independent phylum. For example, in the book "The Afroasiatic Languages," the author states the following:

Theil (forth- coming) criticizes the methods used in research on the historical investigation of Omotic languages. According to him the correspondences reported between Omotic and Cushitic/Afroasiatic do not account for more than mere chance-correspondences, e.g. between Omotic and (Proto)Indo European. He thus claims that Omotic is an independent language phylum. Theil correctly pointed out that systematic and rigorous historical comparative evidence solidly based on sound correspondences is essential for any claim on classification. However, such evidence to support his claim that Omotic is an isolate is (yet) to be presented.

Establishing rigorous sound correspondences to try to establish Afro-Asiatic unity will not be fruitful, since the branches are too dissimilar lexically. So, really, what most strongly unifies them are things like shared pronouns (including pronominal affixes on verbs and possessive pronominal clitics), some idiosyncratic grammatical isoglosses, cognate verbal derivatives, and a few more. If using that as the yardstick, while recognizing that Afro-Asiatic is really an exception here (this process really can't be replicated with other families), then some of the limitations of the historical comparative method become less stringent, but only within appropriate reason. Nevertheless, who knows, some languages conventionally classified under Omotic might eventually turn out to be independent lineages (most are understudied), much like how Bangime has been comfortably excised from the Dogon languages. Time will tell.

There are many examples of linguists grouping together poorly understood languages and then backtracking later. There's Hadza which was grouped with Khoisan but is treated as an isolate today. Then there's Nilo-Saharan which no one knows if it's really a valid family, specially Songhay which was grouped into Nilo-Saharan randomly.

Yes, but I would say that's due to Africanists tending to abide too much by Greenberg's initial classification of Africa. Indeed, Hadza's unity with "Khoisan" was the conventional approach in the past, but I've seen Khoisanists for many, many years saying that Hadza is not related to other languages/families classified under "Khoisan." They've demonstrated the lack of relationship without difficulty.

Also, I wouldn't say the assigning of Songhay to "Nilo-Saharan" was necessarily random. Rather, to me, it's the overreliance on only some core lexical data and using that to justify its membership in a proposed family that is itself lacking in diagnostic justification.

For example, for years, it's been clear to me that Songhay is an independent family, but I've seen, for instance, Roger Blench, who, while still giving lots of exemplary and truly indispensable linguistic information, treat some of the Songhay pronominal similarities to other "Nilo-Saharan languages" as strongly suggestive of a link with, for example, Saharan languages and more. He uses some other lexical data to try to buttress that opinion. To others, these similarities are spurious, and I agree. As for "Nilo-Saharan," I don't think anyone could ever prove its unity at large.

I don't know where you get the information about shared wild grass vocab between AA and Omotic. Afaik, only a few words can non-controversially be reconstructed for AA (and that's excluding Omotic) which relate to pronouns, some numbers, kinship words and very basic words such as blood/water. Even if there was such a thing, it wouldn't prove anything conclusively because you can't rule out coincidence or borrowing of only wild grass related vocabulary.

[/QUOTE]

I sourced some of the information about the wild grass collection terms from Linguistic Stratigraphies and Holocene History in Northeast Africa. That is written by Christopher Ehret. I believe that he sometimes makes not well-substantiated claims, which are often grating to see him clutch to, but I still find his work on wild grass collection and Afro-Asiatic languages convincing enough. Additionally, wild grass collection doesn't generally constitute basic lexicon, which, I believe, is likely why you hadn't come across it. The lexicon you did provide, on the other hand, such as kinship terms, are generally considered basic lexicon.

Yes, borrowing is a possibility, but then that would be a very particular set of words to borrow, especially given that, for example, the agricultural vocabulary of Omotic languages does not overlap with that of Cushitic. Likewise, as I said before, the contact between Cushitic and Omotic languages cannot account for all the similarities between them. Additionally, these similarities in wild grass collection terminology do not appear to be shared with other neighboring phyla. Also, for these to be chance resemblances, they shouldn't be as many as they are. The English word "hand" and the Hausa word "hannu ("hand")" are false cognates, but then there's no appreciable number of anatomical words shared between the two. Then, the origin of the /h/ in both languages is not the same. The /h/ in "hand" comes from an earlier /k/, whereas there's no evidence for this Hausa. So, chance similarities mustn't be overstated.

To return to borrowing, Northern Songhay languages, for example, while borrowing extensively from Mande languages like its relatives under other Songhay branches (the 3rd-person singular pronoun was borrowed), has also borrowed very extensively from Berber languages such that only a few hundred words are of Songhay origin. Then, sometimes, even some words borrowed from Berber languages take on Berber inflections (e.g., Berber nouns taking on Berber plurals and Songhay nouns taking a suffix in languages like Tadaksahak). Other Songhay branches also show Berber influence, but it's less prolific. In spite of that, Northern Songhay is still clearly distinguishable from and unrelated to Berber languages, let alone any Afro-Asiatic language. The extent of the borrowing, which has gone on for a while, mind you, did not result in the transference of conspicuously core Berber traits or very core vocabulary, such as pronouns. Now, some agricultural vocabulary was transferred into Northern Songhay from neighboring languages (not only Berber languages), but there is a still a core of agriculturally-linked vocabulary that is unique to Songhay at large.

Likewise, the information on Songhay's relationship relative to Mande languages, which is similarly prolific, doesn't demonstrate a genetic relationship between these two phyla. The relationship is much like what's seen between Japonic and Koreanic, which are themselves not proven to be related, and which, in spite of some obvious overlap, also show significant differences in agricultural vocabulary and crop names, even for terms that cannot otherwise be tied strictly to agriculture (e.g., seed, buckwheat, rice plant, etc.).

So, when once again placing Omotic languages in this context, the relevant information doesn't speak to all these similarities necessarily being down to borrowings or chance. Being noncommittal toward the linking of Omotic to Afro-Asiatic is still, to me, understandable, but I don't think people should be too taken by the statements on the questionability of Omotic languages.

**Granary** · 11-27-2020, 05:17 AM

Originally Posted by Keneki20

Linguistic stuff

Do you have a good source that shows a definite linguistic tree for Afro-Asiatic that is generally agreed upon(or at least with which you or Agamemnon agree upon)? Also do you have anything on the chronological estimation on when various branches split from one another based on reconstructed lexicon or other methods?

**Ryukendo** · 11-27-2020, 05:21 AM

The ancient Cushites from the Pastoral Neolithic had quite a lot of Dinka and Mota ancestry, the ancient Egyptians and Levantines had levels so low as to be hard to recognize. Modern middle easterners have more SSA ancestry than ancients, but this all apears recent; Berbers have low to no Dinka ancestry, and Chadics have very little recognizable West Eurasian ancestry at all save for uniparental R1b-v88. I think these are all facts that we can aknowledge, and to be honest I don't see how one can conclude from this whether or not proto-AA speakers from thousands of years before even the oldest samples we have currently were some combination of natufian, Dinka and Mota, all natufian, or some other mixture. Even if we accept an urheimat it doesn't need to imply a certain ancestry (who would have predicted the Baltics being straight WHG and Scandinavia being quite EHG-shifted SHGs, with Norway being the most?) Clearly the picture right now is already contradictory and the story is not going to be as clean as it was for IE. I'm not even sure we would recognize a sample from the proto-AA speaking population if we had one.

It would probably take a lot more sequencing, and an attempt to trace each subgroup of AA separately, before we get a clear answer to proto-AA as a whole. Can we just note the uncertainty at this stage and move the thread back on track?

Thread: The mixed genetic origin of the first farmers of Europe

Thread Tools

Display

The Following 7 Users Say Thank You to Keneki20 For This Useful Post:

The Following User Says Thank You to TuaMan For This Useful Post:

The Following 8 Users Say Thank You to Helves For This Useful Post:

The Following 11 Users Say Thank You to drobbah For This Useful Post:

The Following 2 Users Say Thank You to Helves For This Useful Post:

The Following 8 Users Say Thank You to Granary For This Useful Post:

The Following 5 Users Say Thank You to Keneki20 For This Useful Post:

The Following 27 Users Say Thank You to Ryukendo For This Useful Post:

Similar Threads

Late Pleistocene DNA suggests a local origin for the first farmers of centr. Anatolia

The First Farmers if Europe Prof Stephan Shennan

Europe's first farmers carried significant Iranian-related ancestry

Early farmers from across Europe were direct descendants of Aegeans

Europe's first farmers came from Turkey: DNA from Anatolian skeletons 8,000 years ago

Posting Permissions