I8904 is the PIA outlier.Originally Posted by mpatsibihugu89
I8904 is the PIA outlier.
What do you think is the most probable explanation? Perhaps there wasn't much hunter-gatherers in modern SL/Djibouti when the ancestors of the Somalis entered modern Somaliweyn
Interesting your graph (post #910) has 7% CentralAfrica1 ancestry going into the Gravettians...
Basal Eurasian signal?
If that was the case Somalis would be much more Anatolia_N shifted like all the other pastoralists, a possible explanation is admixture from a more Dinka + Iran rich pastoralist population, or the simpler model of Somalis as Kenya_PastoralIA + Levant_BA/Baqah..etc.
Last edited by Mnemonics; 01-12-2021 at 03:02 AM.
drobbah (01-12-2021)
That would mean the ancestors of Somalis and other related peoples were still in NE Africa by the time Iranian ancestry entered modern Egypt/Sudan.If we were to look at E-Z813 the most basal samples were an Egyptian & Sudanese on ftdna and an Upper Egyptian is the most basal sample for the mostly Somali dominated E-Y17859.Southern Horner E-Z813 & Northern Horner E-FGC14382 probably represent the later more Levant_BA rich populations and this would make sense with their younger tmrca.I would assume that the E-V32 PN sample belonged to the most divergent E-BY8114 branch that the Luos also belong too which isn't surprising as they live not too far from his burial.
By what date did Iranian ancestry enter NE Africa specifically Egypt? And do isolated modern South Cushites like the Iraqw completely lack the Iranian/CHG signal found in Somalis?
Justnotyou (01-12-2021), maroco (01-13-2021), Mnemonics (01-12-2021)
I haven't run the Iraqw because they aren't in my dataset but the Iran_N/CHG affinity is missing from the Masai which I think indicates that they probably don't have it.That would mean the ancestors of Somalis and other related peoples were still in NE Africa by the time Iranian ancestry entered modern Egypt/Sudan.If we were to look at E-Z813 the most basal samples were an Egyptian & Sudanese on ftdna and an Upper Egyptian is the most basal sample for the mostly Somali dominated E-Y17859.Southern Horner E-Z813 & Northern Horner E-FGC14382 probably represent the later more Levant_BA rich populations and this would make sense with their younger tmrca.I would assume that the E-V32 PN sample belonged to the most divergent E-BY8114 branch that the Luos also belong too which isn't surprising as they live not too far from his burial.
By what date did Iranian ancestry enter NE Africa specifically Egypt? And do isolated modern South Cushites like the Iraqw completely lack the Iranian/CHG signal found in Somalis?
I have seen mota as having a "deep" component that variously is south africa-related, central africa-related, basal human-related or something of its own.
And the relations of basal/central/south-african doesn't seem to be "stable"(varies a lot in different qpgraph models: the shumlaka paper had basal/central-african forming a clade to the exclusion of south/"neo"-african; some have south-african, some have central-african being closest to non-africans etc).
Same with the eastafrica(the mota ancestry that is closer to eurasians), basal-eurasian/ANA, eurasian itself and the different west africa ancestries(IBM, yoruba, dinka).
If I had to guess the most that can be said at the moment is that human populations in africa seperates into central-african, south-african, basal-african, neo-african and probably others(perhaps a middle stone age north african and east african contributing to IBM(probably not a high amount) and mota respectively) splitting around 200k-300k years ago(?) with the specific phylogeny unknown (until we have further ancient dna).
The "neo-african" then, likely in (north?-)east africa, splits into eurasian, basal-eurasian, mota-related, AWA(yoruba-related), AEA(dinka-related) and ANA(IBM-related) perhaps around 100kya (where again the specific phylogeny is unknown).
Basically this(colors, except orange which is a clade, are common paraphyletic groups):
africaphylogeny0.png
The following is then what I would guess is the most likely phylogeny(could change 'BasalEurasian' and 'NorthEastAfrica' to orange as well):
africaphylogeny1.png
An alternative would be to move ANA to between EastAfrica and NorthEastAfrica and you get the ANA in the dzudzuana paper basically.
Actually, thinking about it this is basically just what the shumlaka paper(https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/Shum_Laka_published.pdf") says...
shumlaka1.png
About that centralafrica-related ancestry into european hgs I have a hard time believing it. Shouldn't that be something obvious someone would have found already?
IDK much about this but wouldn't F4(chimp, mbuti/shum_laka/biaka;ust-ishim/east-eurasian/goyet/ma1, kostenki14/sunghirIV/vestonice) be (significantly) positive in that case?
which it was(already had these populations f2stats computed): F4(chimp, shum_laka; ust-ishim, kostenki14): est=.0023, Z=2.41
I assume it is more likely that it's the other way around: west-eurasian admixture in central africans. Or I guess it could be that some ancient north african population(before IBM), related to central africans, contributed to some european hgs?
Alfa (01-16-2021), Angoliga (01-16-2021), drobbah (01-13-2021), gihanga.rwanda (01-13-2021), Mnemonics (01-13-2021), pgbk87 (01-14-2021), ThaYamamoto (01-13-2021), traject (01-16-2021)
You’re very welcome.
I think your hypothesis is reasonable, but I would say we have a ways to go with understanding the full nature of Ubangi. I brought up Proto-Central Gur to show that it shows a kind of heightened lexical relationship with Proto-Gbaya, and that this would help point toward an origin in the west rather than in the east. However, there’s still more that needs to be done to establish that it has a closer relationship to Central Gur than to anything else.
For instance, even though there is a lexical relationship between the two, no efforts have been made to indicate sound correspondences between them yet. That’s not to say they cannot be established, but they just haven’t been done yet. I do think many of the the lexical similarities between them are significant in terms of highlighting not only a Niger-Congo genealogical classification, but also possibly a closer relationship to Central Gur.
However, since we see morphological differences between Central Gur and Gbaya (the former has a typical Niger-Congo noun class system, but the latter doesn’t), there’s still a lot more we have to traverse to be certain about Gbaya’s exact placement.
As for other Ubangi groups, though, I haven’t seen them compared to Central Gur or other Niger-Congo groups farther afield. However, the genealogical links between Ubangi groups at large are not well understood right now. Since they appear to be independent branches from one another right now, and since Ubangi groups have also suffered from a lack of research and documentation, we also have quite a ways to go with them as well.
For example, the Zandic group of Ubangi has been considered rather divergent compared to even other Ubangi groups. Some Ubangi groups could indeed be closer to one another eventually and even closer to other Niger-Congo groups, kind of like how Italic and Celtic have been treated as independent Indo-European branches even though there’s a growing consensus that they are closer to one another than to other extant Indo-European branches. But, the studies of Ubangi languages are just not advanced enough yet to be totally firm in opinions concerning closer genealogical affiliations of its constituents.
Now, that’s not to say we can’t treat the various Ubangi groups as having occupied a roughly contiguous area before finally dispersing eastward into their current locations. Their shared features seem to be the result of long contact that, I believe, couldn’t have been the result of being only in their current locations for a long time, but rather even being in contact before arriving in their current locations. That’s what I think anyway.
However, regardless, in line with what you wrote, there was likely the incorporation of peoples speaking unrelated languages into the fray for Ubangi groups, and I believe that had may have had an effect on them. When looking at many Benue-Kwa languages, for instance, a popular hypothesis right now is that what initially kickstarted the loss or reduced importance of many of the features typical of Niger-Congo was language contact with pre-Niger-Congo languages in the region, particularly if these incoming Niger-Congo languages came to occupy a more heavily forested zone.
I think you’re right about there being many lakes and rivers, and that Lake Chad doesn’t have to be the only place to look at. I suppose it keeps showing up in these kinds of proposals because of its size and how it would have been a barrier to movement, but I can understand how its importance can be overstated.Not to discredit any peer-reviewed article but Lake Megachad has been milked long enough, the humid period had enough lakes in the Sahara to support most proro-languages/language isolates, as seen in the second attachment, that were subjected to far more adverse consequences from the desertification, there disappearance would realistically act as better incentives/motivational catalyst for the refugees that moved to rivers, than the slowly growing scarcity from Megachad.
Pygmies and neighboring non-pygmy peoples have tended to live symbiotically in a patron-client relationship. So, they’ve exchanged goods and services quite often. Pygmies have been far more adept at living off the forest, since they’ve lived intimately with it for so long, and have been able to procure forest produce for non-pygmy farmers. In exchange, they would receive agricultural produce from farmers. Oftentimes, pygmy peoples have worked on land tended to by farmers and would lend a hand in exchange for the produce that farmers can provide as well.How do you suppose the Ngondi–Ngiri culprits managed to introduce the Niger-Congo language to the nomadic Biaka pygmies?
Since they can receive valuable agricultural produce in exchange, the links to non-pygmy peoples had become strengthened, and the need to communicate increased with these very different peoples. Eventually, the close contact appears to have resulted in a language shift, though I think this shift may have taken some time. There appears to be substrate influence from the now-extinct and unidentified “pygmy languages,” which are very often seen in botanical terms. Unfortunately, I haven’t been able to identify clear examples, but the existence of substrates in their languages is what I’ve always come across throughout the relevant literature.
Otherwise, there has also been rather extensive admixture from non-pygmy farmers amongst many, though not all, pygmy peoples, such that a handful of pygmy peoples actually have a slight majority of farmer-related ancestry, such as the Babongo in Gabon, who have between 42.8% and 52.9% farmer-related ancestry. Otherwise, other pygmy peoples have less of this ancestry, but the percentage of West-African-related ancestry is still rather noteworthy. Biakas in this case have about 5.5% farmer-related ancestry, though the range is large (0% - 10.1%). The group so far that has mixed the least appear to be the Mbuti, who apparently show almost no farmer-related ancestry (recent official statistics estimate 0% farmer-related introgression).
Last edited by Keneki20; 01-14-2021 at 03:34 PM.
I'm pretty confident that there existed an admixed East African-South African population cline that also had some sort of paleo-Eurasian admixture that seems equally related to ancient East and West Eurasians.
Hora, Mota, and Kenya_LSA all show peaks of this paleo-Eurasian ancestry in Africa at around (8-12%), and the other Malawi samples show significant less but still have this admixture.
A population from this cline likely contributed to Shum_laka (which is mostly from a population related to western Central-African branch like the Biaka) which explains both its minor eurasian admixture and its increased affinity to East Africans.
This paleo-Eurasian population also seems to have significantly contributed to certain Eurasians, particularly the "Basal rich" ones.
You can successfully model these populations with both qpAdm and the G25 nmonte (with relatively low distances) as a combination of East African, South African, and Ust-ishim, which make me feel much more certain about the existence of this cline.
I wonder if this paleo-Eurasian ancestry is what gave the Hofmeyr skull its strange paleo-European features, the initial explanation was that this similarity was due to shared OOA morphology but that never made much sense due to the fact that modern South Africans seem to be a very distinct and isolated population and the sample was dated to ~36K Ybp, which is very late for any significant paleo-Eurasian affinities to be present in South Africa.
Last edited by Mnemonics; 01-13-2021 at 08:51 PM.
Which East African population?
Posting Permissions
Reply With Quote