PDA

View Full Version : The Anglo-Saxons and Bernicia/Lothian?



Calum
06-01-21, 21:32
Hi all. I'm doing some intensive research into the history of Bernicia, Northumbria's Northern Kingdom, and of neighbouring Lothian, which came under Bernician rule in the 7th century. It seems to me that the genetic data could settle a lot of issues which have been the subject of debate for historians and archaeologists for several decades and I'm hoping that some of the experts here might be kind enough to help give me a little guidance, as someone who has almost no understanding of the genetic data, as well as point me to any relevant studies and articles.

I'm very aware of the historical and archaeological evidence for Bernicia and Lothian, as well as the usual interpretations of them, so I'm specifically and exclusively looking for insight into what the genetic data itself reveals, rather than a restating of what are believed to be the historical facts, or the various interpretations of them.

My particular focus is the issue of Anglian settlement in Bernicia and across the Tweed in Lothian. For decades now, archaeologists have been commenting on the distinct lack of evidence for the Anglo-Saxons north of the Tees. Historians too have concluded that Bernicia must have remained overwhelmingly British Celtic in terms of population with very little Anglian settlement, mainly at the 'elite level'.

It seems to me, as a layman, that the PoBI data supports this position, with Northumbria looking quite unlike England South of the Tees and with relatively low levels of GER3 (lower than Cornwall and even parts of North East Scotland).

Strangely, although most historians and archaeologists happily acknowledge the great lack of evidence for any significant Anglo-Saxon settlement north of the Tees, it is still common - when the context switches to Scottish History - to speak of heavy Anglian settlement in Lothian, which is obviously a bit of a contradiction, as Lothian would naturally have less Anglian settlement than the entity (Bernicia) which is supposed to have settled it - not more.

However, what little data there is for Lothian in the PoBI project also shows - as I understand it - very little Anglian ancestry, with none of the tell tale red squares of the PoBI map which litter Central and South Eastern England making it much further than Newcastle.

Anyway, I'm hoping some of the more experienced posters here might offer their thoughts on this subject of Bernicia and Lothian and the Angles and perhaps point me in the right direction.

I look forward to hearing any comments and discussing them with you.

I1a3_Young
09-01-21, 07:10
The POBI data was procured by LivingDNA so you may want to contact them. They have been refining their British subregion DNA estimates and may have the detail in which you're interested.

Sent from my SM-G960U1 using Eupedia Forum mobile app (http://r.tapatalk.com/byo?rid=89698)

Calum
09-01-21, 14:08
Thanks for the reply. When you say 'procured by' do you mean that they use the PoBI data to offer their clients a breakdown of their British region ancestry, or do you mean that they actually procured the original data for the PoBI? I've never seen them mentioned on any of the PoBI material, just the Wellcome Trust.

Calum
10-01-21, 02:46
Seeing as it relates to the thread, perhaps I can ask Maciamo about the following paragraph:

"Germanic people brought a whole new set of paternal lineages with them, namely I1, I2a2a-Z161, R1a (L664 and Z284), R1b-U106, and to a lower extent Q1a. Those haplogroups now make up over half of all male lineages in England and Lowland Scotland."

Seeing as Lowland Scotland would include part of the area under discussion, I'm wondering where the data behind the statement that Germanic haplogroups "now make up over half of all male lineages in... Lowland Scotland", can be found, as I didn't see any direct references?

The article also says, "England appears to be fairly homogeneous in terms of Germanic ancestry, except for Cornwall, which is slightly more Celtic."

However, the PoBI data seemed to show that Northern England or England North of the Tees was distinct from the rest of England. It also showed North East England as having considerably less Ger3 than Cornwall and about the same Den18. Wouldn't that make Cornwall more Anglo-Saxon than Northumbria, or to put it another way, wouldn't it reveal the far North East of England (and Cumbria) to be more Celtic than Cornwall?

MOESAN
11-01-21, 00:51
1- between the Bernicia old times and today S-E Scotland/N-E England genetics, time has passed, and resettlements of people has surelyoccurred in some way.
2- two pops of same origin but isolated later, even for a short enough time, can undergo serious drift, if they are small enough and (even partly) endogamous.
3- on what I have at hand (diverse plottings and clusters), the sample concerning Scotland seems small, what reduces the reliablety of these studies.
4- the toponymy of S-E Scotland (Lothians, but more tipically: Peebles, Selkirk, Berwick & Rowburgh) countains a high amount of old Angles placenames, common with Northumbry and farther southwards, denser than the P- Celtic placenames (Pictish comprised, for Gaelic names, they are very seldom there).
5- (weak point, I avow) in my opinion, the S-E Scotland pop is closer to allover England pop than S-W Scotland is.

&: concerning they attempt to compare "contiental" diverse imputs on Britain, their results are very not convincing, I think. Other surveys, where appear in a two dimension PCA France, Germany, England, Denmark people, show that England and Germany covers a same area, between France and Denmark, or separate England and Germany, but they overspan one on another, what is not the case for France, touching both but without overspan.

Calum
11-01-21, 13:24
Hi MOESAN, thanks a lot for your thoughful comments.

I think with regard to point one, the PBI study tried to address that by only taking samples from those in rural areas and whose grandparents had all lived close to one another in the same rural areas. In this way they tried to eliminate any population movement since the late 1800's. The fact that the seventeen clusters of the PoBI match the territories of 'dark age' kingdoms almost exactly was itself a proof that there had been remarkably little movement and resettlement in that time, although of course there must have been some, but not enough to change the regional signals.

On point two - if you are saying that Bernicia and the rest of England would appear genetically the same, except for the effect of isolation, which could have lead to an Anglian population in the north east of England gradually appearing distinct from that of England south of the Tees - then I'd have to ask what evidence there is for the isolation of the north east from the rest of England? And if there are geographic barriers, wouldn't the same barriers be the reason why the original Celtic population would have remained isolated and preserved from England south of the Tees?

Also, 'The Genetic Landscape of Scotland and the Isles' which came after PoBI found that the difference observed in Northern England was not due to isolation in an Anglian population, but that it was the preceding Celtic kingdoms (Rheged and Gododdin) which were 'driving the signal'. This Northern English cluster was prominent in Cumbria, and area which had very late - and very light - exposure to the Angles, yet it was no different to the North East, or Bernicia, indicating that the difference observed throughout the north to the rest of England is coming from the Celtic populations there and the lack of Anglian presence.

On your third point, I agree that the Scottish samples were few. however, they all came from close to the border and with the inherent randomness in this study, if there had been any significant Anglian settlement in this area - which should have been the most 'Anglian' part of South East Scotland - then it would surely have been apparent even in a few samples. Also, the pattern that south east Scotland shows is replicated in Northumberland, especially it's northern part, which adds weight to the few Scottish samples and showing that the outlying 'Anglo-Saxon' red squares of England south of the Tees fading out just beyond Newcastle.

Point 4 I'm not going to touch. First be because, as I said in the opening post, I'm only interested in the genetic data itself and, secondly, because I find place names to be a highly unreliable guide, especially in the interpretation of scientific data. Also, I don't quite agree with your assesment of the place names, which are quite heavily Celtic (I tried to post a link here but am not yet allowed) For instance, every single location that the Bernicians are definitely known through documentary evidence to have had a military outpost or ecclesiastical site in south east Scotland - as far as I'm aware - has a Celtic name. But I won't get into that further, as I see it as off topic.

On your fifth point, the PoBI in combination with the 'Scotland and the Isles' study show that the North East (and south east Scotland) is little different to Cumbria, a region which - as its name indicates - is heavily Celtic. The North east also shows similar levels of DEN18 and considerably less GER3 than Cornwall. - making Bernicia less, Anglo-Saxon than Celtic Cornwall. South East Scotland even has less GER3 than parts of North East Scotland, beyond the Forth, which is very telling!

Thanks again for your response and I'll be interested to hear if you have any further comments.

MOESAN
17-01-21, 13:00
@Calum
Damned! You made an interesting answer to my post written a bit quickly and whose goal was just to emit some possible explanation, without to go deep into the question or to contradict you completely. It deserves a well weighted answer of mine, after I try to gather some solid arguments! Let's be patient, I 'll do my (poor) best with what I have at hand. Write you later.

Calum
20-01-21, 17:42
Thanks Moesan.

As I say, I'm certainly no expert but I have poured over those two studies a great deal for a few months and given them a lot of thought and I think my conclusions are sound. I think they show the North East - much like the North West of England - to be the most Celtic (or least Anglo-Saxon) areas of England, moreso even then Celtic Cornwall. This would necessarily mean the idea of a settlement of Lothian by an Anglo-Saxon population is a myth.

It's good to know that my responses weren't so far off track as to be ridiculous. I look forward to any further responses you have.

MOESAN
23-01-21, 01:33
@Calum
I exposed some points as stuff to weight, without to go deep into the question. I ‘ll try to answer you with my limited knowledge on the question, without any certainty, and my answer will not destroy seriously your arguments. By the way, my thought is that Beirnicia of the 600’s was more Anglian than it appears to us nowaday. That ‘s to say Celts moved back at first after fights before some « winner’s peace occurred and new alliances took form, and so, crossings.


You Calum
I think with regard to point one, the PoBI study tried to address that by only taking samples from those in rural areas and whose grandparents had all lived close to one another in the same rural areas. In this way they tried to eliminate any population movement since the late 1800's. The fact that the seventeen clusters of the PoBI match the territories of 'dark age' kingdoms almost exactly was itself a proof that there had been remarkably little movement and resettlement in that time, although of course there must have been some, but not enough to change the regional signals.
Me
Late 1800’s is not very far in past. That some traces of ancient structures could be found is not so surprising, specially at very fine scale, even after moves and crossings, these last ones can’t erase every difference so easy if they are old enough and followed by a relative stability : they modify the differences and distances but don’t erase them completely. Concerning the rural aspect, I don’t know its effectiveness in region like Tyne or Yorkshire. Were they all agricultors or someones only countryside inhabitants (with some ascendants moves)? Maybe a surnames survey could give me clues about moves of population these last centuries, before our recent times ? Specific moves evidently, and not allover go an return which could become the rule in future. To go back to the depth of differenciation, it doesn’t seem on the PCA that the most of the SE Scotland/NE England cluster plot so far from the bulk of the English pop, compared to people of Orkney and Wales (and surely Ireland, out here). It’s rather at a finer level when appear a N-Ireland/Scotland cluster (yellow circles), that some rare inhabitants of NE-England SE Scotland, show some attractiveness towards Wales people, when the most of these yellow circles concern rather Scotland.
You, Calum
On point two - if you are saying that Bernicia and the rest of England would appear genetically the same, except for the effect of isolation, which could have lead to an Anglian population in the north east of England gradually appearing distinct from that of England south of the Tees - then I'd have to ask what evidence there is for the isolation of the north east from the rest of England? And if there are geographic barriers, wouldn't the same barriers be the reason why the original Celtic population would have remained isolated and preserved from England south of the Tees?
Me
I confess I hav’nt understood their way to create their clusters (I suppose I’ve only a part of their work at hand). I find some discrepancy between the PCA’s plottings and the clusters, I cannot explain. More recent common drifts (clusters?) upon different backgrounds (PCA’s?) ??? I give up here.
It’s true that some pecularities can be born as well by crossing with neighbours or by drift linked to partial isolation (problem which concerns also my point above). Are the peculiarities of some Celtic communities conserved by isolation of already different pops, or born by drift by isolation of close pops ? Both explanation can exist. Personally I think they existed before what doesn’t exclude later ddrifts. And I don’t deny that a Celtic element play a role in the formation of the NE (and NW) England cluster. But it isn’t to say it played a so big role, because Celtic elements have influenced England as a whole, spite in different degrees.
You, Calum
Also, 'The Genetic Landscape of Scotland and the Isles' which came after PoBI found that the difference observed in Northern England was not due to isolation in an Anglian population, but that it was the preceding Celtic kingdoms (Rheged and Gododdin) which were 'driving the signal'. This Northern English cluster was prominent in Cumbria, and area which had very late - and very light - exposure to the Angles, yet it was no different to the North East, or Bernicia, indicating that the difference observed throughout the north to the rest of England is coming from the Celtic populations there and the lack of Anglian presence.
Me
I have some hard work to digest the fact N-Ireland and Scottish people of diverse regions are so homogenous. Even typology and phoenotypes metrics seem showing differences (if slight). The PCA’s are instructive, I think. As zooming increases and new colors and forms appear, they show the bulk of SW+W+ Scotland + even Inverness (small number here!) and N-Ireland is more attracted towards Wales than Cumbria, Northumbry and other parts of Scotland, with E-Lothians and E-Borders (Berwik). At the finest scales, we see Cumbria with its white triangles is even more ‘germanic’ or rather more ‘norwegian’ than the red squares of E-England. The Same with E-Grampians, partly germanized, and their pink circles and white squares, just a little attracted at the same time towards Wales and Orcadians (Norwegian input?). We see also Argyle and Some parts of W-Ulster are even more attracted towards Wales : here is the strongest Celtic input, and nowhere else, better reflected by PCA’s (so general auDNA?) than by their clusters system.
The clusters as illustrated in what I have of PoBI show just some heterogeneity in N-England/SE Scotland and surely moves between Middle Ages and 1900’s. From the few we have from Berwick we see thay have only ONE yellow circle (more « Welshlke ») drown among orange circles (more « Englishlike »), when even Northumbry and Cumbry have more.
My todate point is : my feelings are that the clustering system and dendograms don’t reflect ancient genetic affinities exactly, less than the PCA positions. At fine scale, I think I see more recent moves (« out of place symbols ») in Northern England, what doesn’t exclude some traces of ancient periods of isolations. As a whole, I see Eastern Scotland is less Celtic and more Germanic than Northern reland and Western Scotland, ths more visible in Agyle by instance. But I see also a little more of Celtic input in Northern Egland and Berwick than in Cntral and Eastern England, but it isn’t a surprise. If I have time I ‘ll try to confirm or infirm all this with the help of the continental admixture proposed by PoBI. If I rely on these admixtures !!! Sorry for my stuffy prose.
Have a good week-end.

Calum
23-01-21, 15:05
Hi Moesan. Thanks for your response. I'll read it a few more times and digest it. But for now here are my initial thoughts.

I agree that the degree of differentiation is subtle, but it is subtle between most of the clusters and England, including those that never saw Anglo-Saxon settlement or Anglo-Saxon rule. I don't think Northumbria is any different from them in that. Cornwall is still considered a Celtic region yet it clusters closer to England than Northumbria does. This is highly significant, I think. Orkney and Wales are at the extreme end of the spectrum and most clusters can't really be compared to them, in my opinion.

I'm also quite strictly avoiding notions of 'Germanic' and sticking to 'Anglo-Saxon' and the PoBI's way of measuring it. The reason is that many peoples - Norwegians, Danes, Flemings - even Normans - could leave a 'Germanic' signal yet it would have nothing to do with the Anglo-Saxon settlements or the Anglo-Saxon age. Even some of the Celtic speaking peoples in Britain long before the arrival of the Anglo-Saxons could have left a 'Germanic' type signal. This can really confuse things and lead to seeing "English" or "Anglo-Saxon" genetics where there are none.

That is why the PoBI map and it's clustering is so useful, IMO. We can see difference - even if subtle - and geographic distribution, as well as the identification of GER3 as the primary marker of the Anglo-Saxons.

By all these measures we see Northumbria as different to England. 1. It clusters differently (more differently than some accepted Celtic regions). 2. The red Anglo-Saxon cluster stops before Bernician territory, with just a few outliers, and 3. Northumbria has less Ger3 than at least two accepted Celtic areas which were never part of an Anglo-Saxon Kingdom and for which there is no known Angli-Saxon settlement. I see all of this as highly significant.

Calum
24-01-21, 16:48
Hi again MOESAN.

Just a few more thoughts and questions after looking at the PCAs etc.

I find it very telling that Northumbria resolves as different to England (and clusters with Scotland) several steps before both Devon and Cornwall separate from England.

This must be significant. Devon and Cornwall were Celtic areas for a long time and Cornwall still is - yet Northumbria emerges as different to England several steps before them? I do see what you are saying about the PCAs and that the Northumbrian signal is closer to the English signal than to the North Welsh one. However - again - Celtic Cornwall is more similar to England than Northumbria, so this apparent similarity is relative and not necessarily indicative of 'Anglo-Saxon'. And North Wales has more genetics from the earliest pre-Celtic inhabitants of Britain, does it not? Is that why is is appearing different to the rest of Britain, like Orkney, very early in the analysis? If Wales has a unique pre-Celtic contribution should it really be used as the 'Celtic' standard?

Also, where does this degree of similarity between Northumbria and England come from? What creates it? The Red English "Anglo-Saxon" cluster is still 60-90% Celtic, according to the PoBI. Could it not be that the degree of similarity is not because Northumbria is like England (And therefore "Anglo-Saxon') but that England is still somewhat like Northumbria because of it's absorption of a majority British Celtic population, If you see what I mean?

Perhaps the degree of similarity is created by England's Celtic element, rather than Northumbria's Anglian one?

MOESAN
25-01-21, 19:56
Hi again MOESAN.
Just a few more thoughts and questions after looking at the PCAs etc.
I find it very telling that Northumbria resolves as different to England (and clusters with Scotland) several steps before both Devon and Cornwall separate from England.
This must be significant. Devon and Cornwall were Celtic areas for a long time and Cornwall still is - yet Northumbria emerges as different to England several steps before them? I do see what you are saying about the PCAs and that the Northumbrian signal is closer to the English signal than to the North Welsh one. However - again - Celtic Cornwall is more similar to England than Northumbria, so this apparent similarity is relative and not necessarily indicative of 'Anglo-Saxon'. And North Wales has more genetics from the earliest pre-Celtic inhabitants of Britain, does it not? Is that why is is appearing different to the rest of Britain, like Orkney, very early in the analysis? If Wales has a unique pre-Celtic contribution should it really be used as the 'Celtic' standard?
Also, where does this degree of similarity between Northumbria and England come from? What creates it? The Red English "Anglo-Saxon" cluster is still 60-90% Celtic, according to the PoBI. Could it not be that the degree of similarity is not because Northumbria is like England (And therefore "Anglo-Saxon') but that England is still somewhat like Northumbria because of it's absorption of a majority British Celtic population, If you see what I mean?
Perhaps the degree of similarity is created by England's Celtic element, rather than Northumbria's Anglian one?

Just some details
- Cornwall today is no more the ancient Celtic country. The Cornish surnames and even the christian names turned into surnames as in Wales are a minority. And I think it goes back to ancient enough times : Cornwall was a region prised by the gentry of England. How much of the people tested were named Baragwineth, Carbeen, Treglown, Penrose or Chynoweth ? Or even only Johns, Williams, Lewis ? In this study I find surprising the big discrepancy between PCA’s and Clusters, for a big number of cases ; + Cornwall people here, cover as a whole a big area in PCA’s, not clearly attracted towards Wales, less than N-England/SE Scotland, less even than Devon neighbours (here it doesn’t contradict your points).


- I say ‘Germanic(s)’ when I speak at a general level, including Scandinavians and ancient Germanics (today N-Dutch/NW-Germans only), Germanics who was already distinct from continental Celts since long ago IMO. No way to confuse them, I think, spite they were not so far one forme another. Surely at finest scale, Scandinavian were distinct from N-Germanics, even both distinct within them.
I 'll write later about what these clusterings (as opposed to PCA) evocate to me, when I find inspiraton and time, and, why not, clues.

Calum
27-01-21, 13:47
Hi MOESAN,

I meant that Cornwall is still considered a Celtic region today according to dictionaries, encyclopaedias and various cultural organisations. The Cornish language survived there for many centuries after the Anglo-Saxon age and no historian speaks of any notable Anglo-Saxon settlement in Cornwall. I don't think there is any real evidence for any significant moment of English people into Cornwall before the 20th century, and this study's samples effectively come from the 19th century, prior to that.

As with placenames, I find surnames a highly unreliable guide, especially when they reflect a move from Celtic to English. I have seen too many documented instances of Scottish Celtic names being anglicized or simply changed completely to English names in relatively modern times to believe that an English sounding name automatically reflects English ancestry. I think that process of Celtic names being anglicized has been profound and happening for almost a thousand years in the Celtic regions and so makes assesment based on surnames absolutely impossible. While a Celtic name - like MacGregor - will likely reveal Celtic ancestry, an apparently English sounding name - like Smith - is just as likely to reveal Celtic ancestry, because of this longstanding process of anglicisation in the Celtic regions.

MOESAN
28-01-21, 16:38
Hi, Calum.
I doubt Surnames as a whole underwent replacement or mispronounciation at the same level as placenames (it's even true for other countries).
Concernng Cornwall, I have some clues concernng language, Cornish surnames and (less), placenames. But I lack clues about surnames and origins linked to social level and occupations. Maybe the loss of Cornish language (18th Cy at the last) is a question of relations and mating??? Slow osmosis, spite Y-haplo's didn't move to fast??? No clue todate.

Concerning Bernicia or other places studied by toponymists, I think the cleaver ones were aware of replacements and so on, and studied specific placenames testifiing diverse levels of depth (military strongholds, big towns, rural agricultural places...).
They dated the use of placenames which can disappear or evolve in meaning, in the same way that the Breton 'kêr' (Celtic 'cagro' > 'caer') which before the 7th Cy signifies stronghold, fortress, and in the 15/16th took the meaning of farm! (look at 'villa' > 'ville' = town and farm, as you know).
So, and it is of light weight concerning later pop's, my take concerning toponymy of Beirnicia was the region received an important input of Aengels at some time (6/7° Cy?), what doesn't prejudice of what occurred later between both pop's, "Germanics" and Brittons of North.
But all this is for the game, because what is of worth is the value to accord to too precise clusterings hyerarchies in today populations, and here, I avow I have not got the clues I was looking for, todate.
I appreciate your cool and sensble contradiction. As I said already, I had only some doubts, no certainty.

MOESAN
28-01-21, 16:56
Just speakng of surnames, Scotland is a very bad example, because a lot of names are the medieval owners ones, often of Normand origin, are formed on placenames of every kind of linguistic origins. Become clans names, many of them have been taken by the clans chiefs men without too much genealogical and genetic direct link. It's funny to see that a big part of the typical 'Mac' names are the ones taken by Vikings chiefs in the Western isles. Some typical Gaelic clans have too forms, one Gaelic and one English (translation); and, sometimes, the chief taken as reference as clan foundator is not the same in the Gaelic and the British official traditions, and even if Gaelic by the form, the same clan has two different "Mac" names in this case. In Lowlands it was another story, apparently. More English names, without it would be easy to distinguish the translations of Scottish Celtic names of the original English names. As matter of fact, very often forms found in Northern England too.