Unfortunately, the Lara Cassidy paper on Ireland's genetic history is now embargoed. But it will interesting to read that paper with this linguistic question in mind:
"why are there typological similarities between Insular Celtic and Afro-Asiatic?"
Cassidy at Eurogenes. Hopefully someone will send me the PDF, but until then, the question...
The mainstream view is that the Celtic entered the Islands in the LBA with the IA representing its cultural apex. Population movements certainly spread this language, but genetic continuity would suggest that Bronze Age natives would have learned Celtic (as a new dialect or a new language)
Before the Celtic invasion, the Island of Britain probably spoke a number of languages related to those of the Low Countries (certainly some kind of LLPIE, possibly the speculative Nordwestblok). But Ireland may be another story as its ethnic enclaves were more choppy, with strange menages of new lifeways and ancient craft traditions. (food vessels as one example)
We think of the modern Irish as supremely North European with high frequencies of new traits associated with Bronze Age population movements. But we must remember that this was a long and cruel process on a little Island. A little Island that spent a lot of energy harassing and conquering the West Coast of the big Island. We have to be careful looking at Bronze Age Ireland though our modern prism of Celticity.
If Late Neolithic Irish tended toward the more Levantine variety of Farmer, and if they survived in any significant numbers to the MBA in Ireland, then it is well possible that portions of the LBA Irish population were learning (or mangling) Continental Celtic from the aspect of an Afro-Asiatic thinker.
I'll offer a speculative possibility for consideration...
1. Western European farmers descend from two major ancestral epochs/bodies
a. Danubian farmers originating in Anatolia and SE Europe that were variously mixed with Hunter-Gatherers (these are the I2 and G guys and are the majority throughout the continental heartland)
b. Maritime farmers originating in Syria that dot the littorals and river valleys across North Africa, parts of the Northern Mediterranean, and parts of the Atlantic facade to Ireland (these are possibly the carries of R1b-V88 and E1b among others)
2. Although Bell Beakers (probably IE speakers) pummeled most of the large Island, the Irish Sea and Ireland itself may have been more patchy in settlement where Neolithic cultural ways and ethnicities lingered and inter-married through the Early Bronze Age to Middle Bronze Age.
3. The bastard Celtic of the Irish Sea and Southern Ireland came to dominate or influence all of the Islands throughout the Iron Age, so what happened on the macro level doesn't matter anyway.
4. Insular Celtic was spread by Irish Bastards with daggers and swords.
See also: Insular Celtic Languages
**Update**
See also: Dispatches comments
CASSIDY, LARA, A Genomic Compendium of an Island: Documenting Continuity and Change across Irish Human Prehistory, Trinity College Dublin.School of Genetics & Microbiology. GENETICS, 2018
Do you think the thesis will have significanly more information than is in her paper in PNAS?
ReplyDeletehttps://www.pnas.org/content/113/2/368
I've only been able to skim the supplements and copied text bits in discussion forums. It looks like quite a few new genomes and a lot to chew on. Very non-boring material.
DeleteI've got bits of the new paper, hopefully I can get the full thing and then blog on it. The Irish Beaker period has a lot that needs to be written of. -BBB
It has a lot more information than her PNAS paper. 93 samples. Her PNAS paper had like 7 samples.
DeleteHere's a supplmenetary info of the Cassidy thesis.
https://docs.google.com/spreadsheets/d/1mk9pMMUbChzyW8CwVUYgokVL4iv83WBAKdIf3pWXJnw/edit#gid=833626537
I’m just utterly perplexed as to why it was embargoed (again) until 2023 after waiting since 2017 for it?
DeleteHer thesis won’t be any more relevant 3 years from now. Particularly in such a rapidly expanding field.
Her 2016 paper qpAdm models for WSH ancestry in the Bronze Age samples were also completely wrong. It seems she mistook EHG from Samara with the much later Yamnaya samples and came up with a gross underestimate of the steppe ancestry.
The embargo was probably to preserve the possibility of getting the content published in a leading journal which doesn't permit preprints.
DeleteBBB here,
ReplyDeleteTo add a little clarity, let me point something out with regards to Theo Venneman.
First, it is unlikely that the Danubian Farmers spoke any sort of Afro-Asiatic language. In fact, I doubt Venneman's theory suggests this as he is concerned with a ruling elite of Atlantic megalithans. In all likelihood, most of Neolithic Europe spoke agglutinative Near Eastern "mystery languages", such as those that survived in Southern Europe.
In fact, these languages probably dominated most of the continent and Britain. But the question here is what affinity Ireland had to the continent. Was Ireland uniquely influenced by these Southern Farmers over Danubian farmers in the Neolithic, and does it's isolation increase the possibility that Afro-Asiatic languages were spoken her at one time?
A Vasconic substrate might make more sense to explain Venneman's substrate pattern in place names than an Afro-Asiatic one.
DeleteIndeed, likely originating in the Atlantic Bronze Age; and I believe that Basque itself has been mooted to have an Afro-Asiatic substrate.
Delete1. I don't see traces of Afro-Asiatic/Semitic DNA in Irish Neolithic or Bronze Age samples, although (i) it does appear sporadically in some other NW European Neolithics (e.g. Funnel Beaker), and (ii) Levantine J-M319 yDNA is diverse, distinctive and therefore likely ancient along the Atlantic fringe.
ReplyDelete2. A component of Iberian Chalcolithic (Bell Beaker ancestral) aDNA appears shared with modern NW African populations, so it looks to me like the same people left their genetic mark from NW African Berbers to Bell Beakers. Even if this brought a negligible amount of NW African aDNA into Europe, it might well have left a minor linguistic or cultural legacy.
3. If we accept the idea that Celtic language, culture and a little DNA was brought to the Irish during the Iron Age from South Central Europe, then some Orientalising Period (Levantine) cultural and linguistic traits might have accompanied it.
Good comments but let me continue to add a little precision to my comments above.
Delete1. The genetic component I'm speaking of is something akin to Cardial folk (if there is such a thing), being very similar genetically to the Danubian Farmer minus continent specific HGs. So basically zero Red Sea components or J-guys, and probably very sparing levels of anything uniquely North African.
2. The Western coast of Europe seems to have hosted both of these farmer migrations but the continental variety with its various HG mixtures dominates everywhere, including Britain and Iberia. But I remember our first Irish Neolithic lady was surprisingly Southern and I wonder what these new Neolithic folks look like.
I'll concede that it is highly unlikely that Ireland is some kind of bastion of Cardium derived farmers, the Irish Neolithic was probably highly mixed. But it is theoretically possible that its isolation allowed it to follow a more conservative linguistic path, even if that meant following the language tradition of a minority of its ancestry.
One other issue concerns the geographic and linguistic origin of Afro-Asiatic, and what we sometimes loosely refer to as an Afro-Asiatic ancestral components (like hypothetically Natufian, CHG or J-guys or whatever)
Quite honestly, I have no idea as I've bounced between NE Africa and the Peninsular Near East quite a few times.
Whatever the story here, I think Langfocus has pointed out something non-random mainstream linguists have been puzzled by for a long time. Could be totally random, but a deep historical connection seems to be in order. -BBB
The origin of Afro-Asiatic could be too far in the past to be relevant to the Bell Beaker period.
DeleteWhat I do trace though is a movement of DNA from Greece/Bulgaria through Italy and the Mahgreb to Iberia and ultimately the Baltic. It looks too late to be Cardial, as the samples only pop up in Spain 3,400 BC, Germany 3,200 BC and Sweden 2,950 BC. There was some Natufian in it, and I suspect some yDNA J and R1b as well.
"I'll concede that it is highly unlikely that Ireland is some kind of bastion of Cardium derived farmers, the Irish Neolithic was probably highly mixed. But it is theoretically possible that its isolation allowed it to follow a more conservative linguistic path, even if that meant following the language tradition of a minority of its ancestry."
DeleteThis seems very plausible. A couple of examples come to mind. First, Icelandic is closer to its Old Norse parent language than any of the other Germanic languages, all of which had more language content. Language frontiers are frequently most conservative because the conventional wisdom tends to underestimate how much language change is contact driven (not just borrowing but also conscious differentiation) v. how much is random drift.
Another example would be the Appalachian dialect of English which is the most conservative of the dialects of the English language extant today (relative to, e.g., Shakespearian English) again due to relative isolation for other dialects and foreign languages. Similarly, but not to the same extreme, the English dialect spoken in New Zealand is more conservative relative to the 19th century English dialect due to greater isolation, than that of England, Scotland, Ireland, Canada, the U.S., Australia, all of which had more outside contact and less of a conscious social desire to differentiate from the source language.
"all of which had more language **contact**.
DeleteSea based language contact seems more plausible than a substrate influence. See, e.g., https://dispatchesfromturtleisland.blogspot.com/2016/12/phoenicianpunic-names-in-britain-and.html and item 7 in the summary chronology at https://dispatchesfromturtleisland.blogspot.com/2018/02/a-short-history-of-demographic-change.html and it is also notable that the Celtic Galatian community apparently left little or not population genetic traces in Central Anatolia despite centuries of existence in the Iron Age there, suggesting that Celtic language shift and expansion may have been elite driven rather than demic.
ReplyDeleteThat's pretty interesting. I remember that.
DeleteThere's no indication from DNA that Maritime European farmers came from Syria. They have no sign of Levantie ancestry. Daube farmers on other hand do have small amount of Levantie ancestry.
ReplyDeleteOverall, I'd it is unlikely Atlantic farmers spoke anything Semetic influence.
That's based on similarities in pottery and we're talking something quite old like going all the way back to the pottery Neolithic. I wouldn't expect any of the modern Levantine ancestries to be present. Also, Syria is just a rough approximation, it's possible that this tradition expanded from a secondary location in Western Greece.
DeleteI agree with your second point. There is no evidence of Semitidic substrate in Western Europe. Most Atlantic Farmers would have spoken agglutinative languages however related to the ancient Iberian Languages. The hydronomy seems to suggest this in the Western Continent.
However, Ireland and Western Britain may be its own weird thing. It does appear the Neolithic Irish are more Southern, and that may be an indication of their language contacts or origin. BBB
As they say, "a million ways to skin a cat".
ReplyDeleteThere's another hypothesis on the correspondence between the Insular Celtic languages and Hamo-Semitic or Afro-Asiatic.
One solution was by John Koch who classified Tartessian as a mixed Celtic with South Iberian Phoenician. This "Milesian" being imported to Ireland in the Iron Age, which then spread to Britain. A critical review of that hypothesis is presented by Steve Hewitt, University of Cambridge:
https://www.academia.edu/35795700/Tartessian_as_Celtic_and_Phoenician_as_a_possible_substrate
See also:
For a more comprehensive list of similarities between these groups and an overview of the hypothesis/criticism
https://www.academia.edu/40827562/The_Question_of_a_Hamito-Semitic_Substratum_in_Insular_Celtic_and_Celtic_from_the_West
I think this hypothesis becomes unnecessary if SW Ireland, and perhaps most of the Irish Sea, had substantial pockets of genetically very southern farmers lingering to the MBA. If these people continued speaking their primary tongue in opposition to neighboring Beakers, it's conceivable that the Celtic language was the death knell of their old pre-IE tongue. Meaning, a new, pan-Island Celtic was the first opportunity for substantial numbers of people to butcher Celtic.
But it is also not impossible that Southwest Iberia supported additional linguistic updating in the Iron Age through trade, etc.
Some possible lexical similarities
https://independent.academia.edu/LughaisMacAoidhBanbridge
And to add one more point...
DeleteWhile the Beaker invasion of Britain appears to have been catastrophic, we also know that there is a substantial decline in steppe ancestry over time and punctuated in the LBA to IA time period. So while much of Britain may have already been speaking an IE language prior to Celtic, it's quite possible that either side of the Irish Sea was still speaking something else, or maybe pockets that would later become influential.
I suspect that the heaviest Steppe ancestry only arrived in Ireland with a Bronze Age influx from NE Central Europe after the height of Bell Beaker; and paradoxically, that this really only declined with the arrival of other people from S Central Europe with a different kind of Steppe ancestry (e.g. Celts).
ReplyDeleteCarlos Quiles has Afro-Asiatic originating in Europe with hunter-gatherers, then migrating into North Africa:
ReplyDeletehttps://indo-european.eu/wp-content/uploads/2019/12/4-mesolithic-early-languages.jpg
https://indo-european.eu/maps/prehistoric-languages/
BTW, she did a podcast on her work with Razib Khan on Instome (including an characteristic Nixon moment on "where the f*ck" we Irish lived back in the day when we were building stone non-residences.
ReplyDeleteSo, we would have an "Irish" populations speaking something like Berber adopting (and adapting) the language from invaders who entered the island from the north, speaking something akin to Polish? This is fun. Old European Language will surely like it
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteBBB.
ReplyDeleteThe genomic ancestry of British/ Atlantic Neolithics is a mixture of Cardial, LBK & WHGs. If anything, i would gather that (given the large amount of HG-introgression), they spoke 'HG langauges'.
The Cardial culture probably sprang out of NW Greece. Sure, there are some cardial-type pottery in East Med, however they are not the lineaer progenitors of the west Med. Cardial groups.
What I do wonder is, *if* the AA stratum actually exists, if its somehow an indirect adstrate effect from Phoenicians
I think an adstrate works if Celtic and Semitic co-existed for several hundred years in the same location, but why not Celtiberian and Gallacian as well? BBB
DeleteThe oldest DNA we have so far from Egypt and Carthage is of western European hunter-gatherer origin..
Delete@ BBB
DeleteCOuld that be due to the fact that insular Celtic is well attested, whilst Celtiberian & esp. Galatian are fregmentery. If a large corpus of the latter 2 were available, perhaps theyd sure this 'stratum' too
@ A
Sorry - can you elaborate ? I thought the Dynastic DNA from Egypt was E and J, not particularly 'European' .
Are you sure Carlos thinks Afro-Asiatic is from WHGs ? That would be a very peculiar theory ...
Djehutynakht, Middle Kingdom ruler, 2000 BC, mtDNA U5b2b5.
DeleteThe oldest DNA samples with U5b2b5 are two samples from eastern Spain, c.3500 BC.
U5b2b5 has also been found in two samples from Sardinia (c.2300 BC).
The R1b-V88 found in Africa appears to have come from Spain and/or Sardinia: https://static-content.springer.com/esm/art%3A10.1038%2Fs41467-020-14523-6/MediaObjects/41467_2020_14523_MOESM1_ESM.pdf
Interestingly, mtDNA H4a1 was recently found in the mummy of an Upper Egyptian noblewoman (Takabuti, 600 BC), and the oldest H4a1 sample we have (c.5300 BC) is from exactly the same area in Spain as the oldest U5b2b5. It has also been found in Sardinia (c.3800 BC). https://i.imgur.com/5gNqR5V.png
This suggests an ancient migration from western Europe to Egypt.
R1b-V88 is associated with the Afro-Asiatic Chadic language, and is found at a high frequency among Berbers in Egypt.
Carlos Quiles:
Delete"One of the theories I proposed in the Indo-European demic diffusion model since the first edition – based mainly on phylogeography – is that R1b-V88 lineages had probably crossed the Mediterranean through southern Italy into a Green Sahara region, and distributed from there through important green corridors, humid areas between megalakes. ...
If we accept that the migration of R1b-V88 lineages is the last great expansion through a Green Sahara, then this expansion is a potential candidate for the initial Afroasiatic expansion – whereas older haplogroup expansions would represent languages different than Afroasiatic, and more recent haplogroup expansions would represent subsequent expansions of Afroasiatic dialects, like Semitic, Hamitic, Cushitic, or Chadic."
https://indo-european.eu/2018/02/r1b-v88-migration-through-sardinia-into-green-sahara-corridor-and-the-afroasiatic-connection/
Carlos comment:
"a) EITHER Afroasiatic is related to Eurasiatic, then may have spread from somewhere close to the Black Sea, e.g. initially with hg. R1b-V88, hence the potential links (also, mostly EITHER/OR) to Mesolithic Italy and/or the Neolithic expansions through the Mediterranean into the Green Sahara.
b) OR Afroasiatic is native to Africa (or the Levant) and spread southwards within Africa, probably during the Neolithic."
https://www.youtube.com/watch?v=f1KqoWfCbRI&t=0s
I think that R1b-V88 is much more likely to arise from the Balkans than Iberia. See my analysis at https://dispatchesfromturtleisland.blogspot.com/2017/09/the-source-of-proto-chadic-y-dna-r1b.html R1b-V88 is primarily Chadic with some minor introgression into neighboring populations with similar cultures. And, other linguistic and genetic elements of the Chadic people favor a Nile route for the bearers of R1b-V88 men as does a quite tight chronological boundary since the archaeological association with the likely proto-Chadic people is quite convincing. (The R1b-V88 people elsewhere in Europe could have the same source population as the pre-Chadic men, but non-Chadic).
DeleteBalkan/Greek Chalcolithic DNA profiles can be tracked along the following route > Italy > Tunisia > Western Mahgreb. This presents a prime candidate, in my view, for the Steppe-infused and Levantine or Natufian-infused DNA that turns up in Iberia during the mid 4th millennium BC, and in NW Europe a little later. Given such a route, North African language features could well have been acquired on the way by the people bearing this DNA.
DeleteR1b-V88 could, I suppose, have spread in the opposite direction at that point towards the Nile, although it looks more likely to be older. It is, of course, difficult to ascertain whether Western fringe Europeans and/or V88-bearing newcomers to Africa might have brought any Chadic lingusitics with them or picked them up as they passed through.
"We detected R1b-V88 equivalent markers in 11 out of 30 ancient Sardinian males from the Middle Neolithic to the Nuragic with Y haplogroup calls. At present, R1b-V88 is prevalent in central Africans, at low frequency in present-day Sardinians, and extremely rare in the rest of Europe. By inspecting our reference panel of western Eurasian ancient individuals, we identified R1b-V88 markers in 10 mainland European ancient samples (Fig. 8), all dating to before the Steppe expansion (at 3k years BCE). Two very basal R1b-V88 (with several markers still in the ancestral state) appear in Serbian HGs as old as 9,000 BCE (Fig. 9), which supports a Mesolithic origin of the R1b-V88 clade in or near this broad region. The haplotype appears to have become associated with the Mediterranean Neolithic expansion - as it is absent in early and middle Neolithic central Europe, but found in an individual buried at the Els Trocs site in the Pyrenees (modern Aragon, Spain), dated 5,178-5,066 BCE and in eleven ancient Sardinians of our sample. Interestingly, markers of the R1b-V88 subclade R1b-V2197, which is at present day found in Sardinians and most African R1b-V88 carriers, are derived only in the Els Trocs individual and two ancient Sardinian individuals (MA89, 3,370-3,110 BCE, MA110 1,220-1,050 BCE) (Fig. 9). MA110 additionally carries derived markers of the R1b-V2197 subclade R1b-V35, which is at present-day almost exclusively found in Sardinians.
DeleteThis configuration suggests that the V88 branch first appeared in eastern Europe, mixed into Early European farmer individuals (after putatively sex-biased admixture), and then spread with EEF to the western Mediterranean. Individuals carrying an apparently basal V88 haplotype in Mesolithic Balkans and across Neolithic Europe provide evidence against a previously suggested central-west African origin of V88. A west Eurasian R1b-V88 origin is further supported by a recent phylogenetic analysis that puts modern Sardinian carrier haplotypes basal to the African R1b-V88 haplotypes. The putative coalescence times between the Sardinian and African branches inferred there fall into the Neolithic Subpluvial (“green Sahara”, about 7,000 to 3,000 years BCE). Previous observations of autosomal traces of Holocene admixture with Eurasians for several Chadic populations provide further support for a speculative hypothesis that at least some amounts of EEF ancestry crossed the Sahara southwards. Genetic analysis of Neolithic human remains in the Sahara from the Neolithic Subpluvial would provide key insights into the timing and specific route of R1b-V88 into Africa - and whether this haplogroup was associated with a maritime wave of Cardial Neolithic along Western Mediterranean coasts and subsequent movement across the Sahara.
Overall, our analysis provides evidence that R1b-V88 traces back to eastern European Mesolithic hunter gatherers and later spread with the Neolithic expansion into Iberia and Sardinia." [And from there into Africa]
https://static-content.springer.com/esm/art%3A10.1038%2Fs41467-020-14523-6/MediaObjects/41467_2020_14523_MOESM1_ESM.pdf
Map: https://i.imgur.com/ool6zDJ.png
Yes, I would estimate the DNA of the Western Mahgreb as fitting best with minor admixture from the South Balkan/Greek Chalcolithic, which then appears to have drifted into Western Europe, possibly bringing some Chadic influences with it.
DeleteR1b-V88 looks probably to trace back to Europe earlier than this, as the small amounts of non-indigenous DNA in African samples seems to bear out.
"Early Neolithic Moroccans are distantly related to Levantine Natufian hunter-gatherers (∼9,000 BCE) and Pre-Pottery Neolithic farmers (∼6,500 BCE).… However, Late Neolithic individuals from North Africa are admixed, with a North African and a European component. Our results support the idea that the Neolithization of North Africa involved both the development of Epipaleolithic communities and the migration of people from Europe… The cultural and genetic similarities between Iberian and North African Neolithic traditions further reinforce the model of an Iberian migration into the Maghreb. ...
DeleteKEB (Kelif el Boroud, Morocco, 3700 BC) is placed in an intermediate position, with ∼50% each of European Early Neolithic and North African ancestries. ... KEB can be explained as having both IAM-like (Neolithic North African) and Iberian Early Neolithic components. The same admixture profile is observed in the Guanche samples"
https://www.pnas.org/content/115/26/6774
"we generated the first genome-wide sequence data and mitochondrial genomes from eleven archaeological Guanche individuals originating from Gran Canaria and Tenerife. ... All individuals predate the European colonization of the Canary Islands (15th century) and one predates the Muslim conquest of the Maghreb (7th–8th centuries), events that have had significant impact on the gene pool of the Canary Islands and North Africa, respectively. Hence, our sample is a good representative of the pre-conquest aboriginals. …
A principal component analysis (PCA) of the five samples with the highest autosomal genome coverage, performed using genome-wide autosomal SNPs overlapping with Human Origins (HO) data, reveals close affinity to modern Northwest African populations such as Tunisians and Algerians, but with a tendency (especially for individuals from Gran Canaria) to occupy a space outside modern Northwest African variation, closer to Europeans ... The results of the ADMIXTURE analysis furthermore show that the Guanches carried early European farmer (EEF)-like ancestry”
https://www.cell.com/current-biology/fulltext/S0960-9822(17)31257-5
I agree, and have honed down this European contribution to find that the best fits for it in Moroccans, Guanches and Tunisians is the kind of DNA profile that we see in the South Balkans/Greece first, and only arising in Iberia later (in the mid 4th millennium BC).
DeleteAccordingly, I would hypothesise that the bulk of the EEF DNA in North Africans most likely migrated from SE Europe via the Maghreb to Iberia during the 4th millennium BC, rather than from Iberia to Maghreb earlier in the Neolithic.
This, in my view, could be relevant to the Irish language question, as it indicates a clockwise migration via likely Chadic-speaking areas up into Western Europe.
This comment has been removed by the author.
Delete@NicolasPaul To a great extent I am with you on your conclusions, but the suggestion that they would have been Chadic speaking is very unlikely, even though they almost surely derived from the same 6th millenium BCE source population. Both linguistic evidence and non-Y-DNA genetic ancestry strongly suggest that the Chadic population and language family are derived from a Cushitic population in Ethiopia with whom the male R-V88 bearers derived their language and almost all of their wives. Neither the linguistic connection nor the genetic one would be present if their route didn't include a soujourn in Ethiopia and that wouldn't happen in a West to East migration or a non-Nile based route across the Sahara.
Delete@A.June Thanks for that material which helps a lot to clarify the overall story of R-V88.
DeleteI agree that it is unlikely these people would have spoken a Chadic language, although they might have picked up some non-standard inflexions/linguistic traits from people they associated with in the Mahgreb who spoke perhaps a brother language to Chadic.
Delete"I would hypothesise that the bulk of the EEF DNA in North Africans most likely migrated from SE Europe via the Maghreb to Iberia during the 4th millennium BC, rather than from Iberia to Maghreb earlier in the Neolithic."
Delete4th millennium seems too late.
SE Europe had metallurgy long before then, but KEB in Morocco 3700 BC is still Neolithic.
The Cardial culture had already arrived in Spain before 5000 BC. Which culture are you associating with a 4th millennium BC migration?
And there appears to be a migration of mtDNAs in the opposite direction. Such as U5b2b5, which appears in the record first in Spain (c.3500 bc) then later in an Egyptian ruler (2000 BC). The same with H4a1, seen first in Spain (c.5300 BC) in the same area as U5b2b5, then later appearing in an Egyptian noblewoman (600 BC). https://i.imgur.com/5gNqR5V.png
Similarly the U5b2c1 found in Carthage (600 BC) and other mtDNAs found in Berbers appear to have come from Iberia.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4880306/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1199377/
The evidence on R1b-V88 also indicates a migration from Iberia or Sardinia into Africa, with it later becoming dominant among Chadic speakers south of the Sahara.
There also appears to be a migration of the Megalithic culture into northwest Africa from western Europe.
Yes, the migrations from SE Europe began earlier with the Cardial culture, and continued thereafter. I am examining the aDNA element that persisted in the Maghreb, which looks to fit best with SE Europe c. 4,000 BC. It is different to Iberian early Neolithic DNA, and I am guessing it arrived in Northern Iberia by mid-4th millennium BC, as it arises almost undiluted in Atapuerca at that point. The same aDNA is found in heavy concentrations in Southern Italy (including Sardinia) today. I haven't examined KEB in Morocco, whose Neolithic DNA element could either be an early manifestation of it or a partly-related precursor to it.
DeleteI am only looking at the genetics and do not associate it with any named culture, although it might have been significantly culturally appropriative.
Yes, there would have been back migrations and indeed migrations in all directions, particularly of mtDNA during and after the Bell Beaker period, as we can see that Beaker men moved around a lot and took on females from lots of different source populations.
I see an element of this aDNA in Bell Beakers, although it was heavily diluted as BB expanded.
In its early days (pre-dilution), there were more heavily Natufian and Levantine varieties of it, and I believe there is a significant possibility that these could have skewed lingusitic development in some parts of Western Europe, just as immigrant populations have subtly altered inflexions in the English spoken in the UK today.
Yes, perhaps Semitic features persisted in the insular fringes for the same reasons that the Celtic language itself (and indeed that prior Beakerish R1b-L21) has persisted in these fringes. Not sure whether this would be due to innate conservatism, or whether these areas were relatively unattractive to all newcomers apart from the Beaker folk. (Cassidy's data doesn't seem to indicate the presence of any pockets of people who were not L21 Beakers, so perhaps the Semitic features arrived with them?)
ReplyDeleteIn which varieties of Celtic are the Semitic features present or absent? Are they a feature of Welsh, Manx and Breton as well? And do we know that they were absent from Gaulish?
I tend to agree with the more mainstream narrative that Gaelic (which represents a more archaic basal clade of Celtic compared to Gaulish/Brythonic) arrived around the time of the early Iron Age transition as opposed to the likes of Koch’s in house “Celtic from the west” theory that it developed in situ from whatever BB spoke, which I don’t believe is credible as there is clear “southern” drift that came about somehow as the island’s population shrank substantially after the collapse of the BA. I believe this is probably how Celtic speech possibly came about on the island. We certainly are at least in terms of direct ancestry the most autosomally bell beaker population alive to date as later Hallstatt waves certainly swept through the remnants of Atlantic Bronze Age Europe. This is also evident in terms of Y-HG as we possess the highest levels of BB derived R-L21.
ReplyDeleteWe’ve adopted the speech of invaders more than once it seems.
This VSO word order isn’t exclusively just an Irish thing, but more so an insular Celtic phenomenon also seen in Welsh interestingly. Vasconic is of course SOV so Venneman’s hypothesis is unlikely here.
If Gaelic were a basal clade of Celtic, perhaps it acquired VSO before it migrated to Ireland?
Delete(Although other languages with occasional VSO do all seem to be Western European fringe.)
I would identify the Irish as autosomally West European Bronze Age, rather than autosomally Bell Beaker. I see BB proper as rapidly overstretching itself, causing its aDNA to be spread thinly.
The mysteries are how Beakerish L21 in Ireland (i) took in lots of NE Central European BA aDNA without losing its stranglehold on male lineages in Ireland, (ii) absorbed Celtic language, but apparently not much Celtic DNA, (iii) acquired a Semitic slant to its language when fellow Beaker-descendants and Celtics apparently did not.
Well ultimately something increased our EEF ancestry between now and then but I’m not advocating wide scale population turnover here on the scale seen with the LNBA genomic transition. There seems to have been a collapse of late neolithic north Atlantic populations which made their replacement all the more easier for BB.
DeleteThe turnover isn’t as noticeable with the Iron Age as these weren’t as drastically different groups autosomally.
R-L21 today also doesn’t quite reach BA levels amongst today’s Irishmen.
Celtiberian samples are also more “northern” relative to the previous BA population.
These movements affected SW and NW Europe simultaneously. But again nothing on the same scale in terms of demographic replacement.
BA British/Irish samples overall share most drift with Scandinavians, I’m talking about more direct kurgan bell beaker (all R1b-P312 thus far) introgression. Barbed wire beakers in Scandinavia had a limited presence and there were already CWC like populations there (Battle Axe, Danish Single Grave etc)
Using the new Gallic genomes from the recent French paper illustrates this better. We are predominantly a BA derived population but not entirely, there was later dilution. I’ve seen some claim the VSO word order is more to do with an emphasis on poetic shorthand rather than any supposed Semitic substrate, but I don’t know enough about linguistics to comment any further on this.
Target: Irish
Distance: 2.1977% / 0.02197657
69.8 IRL_BA
30.2 French_IA_central
Target: Irish
Distance: 1.8908% / 0.01890777
76.0 IRL_BA
24.0 French_IA_south
Target: Irish
Distance: 2.2563% / 0.02256308
90.0 IRL_BA
10.0 IRL_MN
I model the increased EEF as principally the arrival of Iberians, French and Celts during the late BA and IA, rather than a resurgence of local Neolithics.
DeleteEarly L21 BB society in Ireland appears to have been flooded with NE Europeans (females, at least) during the EBA, and a fair amount of S/C Europeans during LBA and IA, without its original lineages losing primacy.
If there were a Afro-Asiatic influence, my guess is that it most likely came through Spain or Italy during the orientalising period, but it might even hark back to a L21 BB origin. L21 dynasties were resolute survivors through periods of BA, IA and Anglo-Norman inward migration, so it would not be surprising if aspects of their language and culture survived along with them.
Big news: we now have our first Bronze Age continental L21, in France: PIR3037AB. He's DF21 same as the Rathlin islanders. Apparently there is also an M222 in the Iron Age La Tene culture, which is HUGE news as this is the most prevalent Y-SNP amongst Irishmen today!
ReplyDelete[https://www.pnas.org/content/early/2020/05/20/1918034117]
Good find. We have Anatolians in Marseille from 600 BC, Celts there in 517 BC and an Irish L21er just along the coast from there 450 BC. I suppose relationships between these kinds of people could have led to a Semitised version of Celtic being brought to Ireland.
DeleteI can find where Iron Age PECH8 is identified as M222, but not where PIR3037AB is shown as DF21. Where does that appear?
DeleteApart from the ultra-specific classification of PECH8 as R1b1a1a2a1a2c1a1a1a1a1, all the other R1b samples seem to have been assigned relatively unspecific classifications (e.g. R1b1ala2).
For some reason they don’t list his terminal SNP in the supplementals but in the graph from the study itself he is indeed listed as R1b1a1a2a1a2c1a5c3c1 which is downstream of DF21.
Deletehttps://anthrogenica.com/attachment.php?attachmentid=37760&d=1590599458
I wonder why they don't they list the terminal SNP in the supplements, and indeed which other potentially informative SNP readings have they omitted to mention.
DeleteI note that PIR3037AB is dated to 2,058 BC, whereas yfull estimates the age of even its parent SNP as only 1,500 BC.
Oddly, ancient samples and modern phylogenies both suggest that the terminal SNPs of both PIR3037AB and PECH8 are Irish/British, and yet both samples were from the same far corner of France (Eastern Pyrenees) 1,500 years apart. Is it possible that migration from Ireland to the Eastern Pyrenees could have continued unabated over such a long period of cultural upheaval?
Not sure how to interpret this, nor why these people would have taken a Semitised Celtic language back to Ireland without taking any substantial Semitic or Celtic DNA with them.
French ancient DNA has been quite late to become available, with the paper just this month being one of the most important additions to the data. So, if Irish/British have something and you have no data from France, it makes sense to assume a Insular rather than a Continental origin for it.
DeleteHowever, I suppose an older Semitic influence is still an alternative possibility.
ReplyDeleteFunnel Beaker samples I0551 and I0802 at Salzmunde 3,200 BC look 20-45% Levantine/Natufian.
Spanish sample ATP3 3,400 BC looks 5% Levantine/Natufian and is positive for a L21 SNP (Y17204).
It seems quite possible that these people were in some way related, and indeed related to the Beaker-derived people that we find in Ireland (with their L21 yDNA and the Semitic-looking features in their language).
This comment has been removed by a blog administrator.
ReplyDeleteThis comment has been removed by the author.
Delete@Nicholas Paul
ReplyDeleteFrom where are you taking your data re PECH8? On dataset S11 he is assigned only P310*.
From dataset S1, cell W233
DeleteThank you. Very interesting if confirmed.
DeleteI have looked at DNA profiles of the NW European Neolithic, and there are strong associations between Late Irish Neolithic and Western Funnel Beaker.
ReplyDeleteWe can also see outlier samples at different FB sites with heavy Levantine-looking contributions. I have already mentioned Salzmunde in this respect. Gokhem in Sweden is another example - it's oldest sample provides an almost exact match with Irish Neolithic, but another of its samples dated to 2,900 BC looks like an almost entirely non-local mix of Spanish Chalcolithic and Levantine Neolithic.
Although these Levantine components arriving at the end of the 4th millennium BC seem to have died or diluted out, it is quite possible that they left a cultural and/or linguistic legacy in FB populations, and indeed in similar neighbouring populations in Britain and Ireland.
Good catch!
DeleteDo you have any more info on those samples?
DeleteI think the relevant Gokhem sample was Gok7, and the Salzmunde samples I0551/Salz3b and I0802/Salz77a
DeleteThanks for the link. I cleaned up some of the formatting in the post and updated a side question I asked that I subsequently learned the answer to out of vanity as a result.
ReplyDeleteHeads up to a new find relevant to Beaker archaeology and Indo-European linguistics. https://languagelog.ldc.upenn.edu/nll/?p=47335
ReplyDeleteOh wow. That's awesome and insightful. Thanks for linking
Delete