Talk:Anglo-Saxon settlement of Britain/Archive 2

Fostering, hostages and refugees
Both British and Anglo-Saxon societies indulged in the related social institutions of fosterage and hostage taking. Often a vassal chieftain would send some or all of his sons to the court of his overlord, to be raised and trained for war. The foster son was effectively a hostage for his father's behaviour, but also benefitted from being near to the source of political power. If an overlord were Anglo-Saxon and the vassal British this would also be a powerful way of anglicising the families of local British elites. It is also recorded that a number of Northumbrian kings sought refuge at Celtic courts, in Gwynnedd and Elmet and in Ireland. We know that King Oswiu was exiled in Ireland at a young age, and while there he seems to have become thoroughly integrated into local society, gaining an Irish nickname and probably a wife. Bede recorded these exiles because they were of Northumbrian kings, presumably the opposite happened and British princes sought refuge at Anglo-Saxon courts. I have not come across these being highlighted as methods of anglicisation in any literature I have read, which is odd as they seem rather obvious. Urselius (talk) 16:08, 16 February 2014 (UTC)


 * Nice thought. I had been thinking also about whether the article reads too "peaceful". We may need to say more about how people, especially the warrior / elite class, lived. There is evidence for trades/raids/hostages/tribute. I think the shifting nature of settlement and lifestyle in this period is a factor. This all would have created a complex pattern of loyalities and activities, tributes and family/tribal rivalries - hostages are a natural part of this. The most interesting part of the Spong Hill excavation is the over 2000 differently designed beads. The differentiation that went on in the early period was large and the possiblities of refugees and hostages even at the micro level. However these settlements and others show a large continuity. Could the people have spread, North and west into the frontier lands - Mercia? By the way we know that the Franks used these tactics as a matter of normal practice - and interesting it came from the need to find revenue, as no-one would pay Tax. probably why there isnt more about is that scholars use the word Slave and maybe your word hostage is what they sometimes mean - for example Patrick was taken in "slave raiding", but in reality if the raiders could have got what they wanted by selling him back I am sure they would have. How much slave raiding happened in early lowland england? Possibly a lot. J Beake (talk) 16:24, 17 February 2014 (UTC)
 * Lavelle, Ryan. "The use and abuse of hostages in later Anglo‐Saxon England." Early Medieval Europe 14.3 (2006): 269-296. Looks at the practice later in the period.J Beake (talk) 16:24, 17 February 2014 (UTC)


 * Nice find, I'll check it out. Urselius (talk) 19:52, 17 February 2014 (UTC)

A little note about Y Chromosome
Outside of the scope of this article, I think it is a real shame that people don't understand Thomas et al's paper. Particularly having seen Mark Thomas on TV, nor does Mark really get it I think. Let me put here the basics of what I am on about.

They did a rather nice piece of statitical analysis.
 * They looked at the populations Na & Nb where A is Incoming chromosomes and B is Indiguous chromosomes
 * They made S the reproductive advantage, where A/B = 1+S/1
 * They considered initial populations of 5%/10%/20%
 * Also those available to have child producing relations between communities (really only incoming women to indigenous men)
 * They looked at up to 15 generations and considered the present population of Y Chromosome possessors as being 50%.
 * Unfortunately they were fixated in the paper with 20% as an initial settlement number.

The results
 * the most selective advantage is 2%, meaning for every 49 "Indigenous" there is 51 "Incoming" boys born, which is very very low in my opinion when you consider economic advantage, opportunity health and age - not to mention preferences for boys in anglo-saxon culture - with possible abandonment of girls. I would expect the real figure to be be more 8-10%.
 * 10% indigenous men having the chance (forget laws) to marry/procreate with incoming women. This means for 150 years mostly only the elite indigenous could marry incoming. This is very possible.

The answer is around 4/5 (80-125 years) generations would take from 20% to 50% holders of Y Chromosomes, but line seems very geometrical in 7/8 (140-200 years) with the above parameters it seems would reach 50% from just over 1% initial. That means a population of 10,000-30,000.

Therefore it seems to me that very low initial populations can reach 40% plus percentage ( my prefer percentage, taking the Danes into account) just with the advantage of young, fit and desirable people - and that they are mostly men. (Without the need to say there was any political or law-based apartheid-like structure) More inportantly it proves how dynamic populations can be and the power of geometric progressions. J Beake (talk) 16:08, 20 February 2014 (UTC)


 * I would just like to go through some of the really huge holes in the scientific rigour of the population genetics studies that have been published so far. Most of them compare Irish/Welsh to North German/Dane, with the English in the middle. They are comparing just three variables and if you compare just three populations you will find apparently simple relationships between them. What all these studies needed to do was to also look at Belgian and North French samples, and compare all of these to an outgroup, such as the Finns. I suspect that much genetic input to Britain that has been imputed to historically recorded North German and Danish incomers is merely Continental versus insular, and not specific to the putative home of the post-Roman Germanic incomers. Also there is a blythe assumption made, with not a jot of evidence, that modern Irish and Welsh marker frequencies are reflective of pre-Roman Eastern England. This is appalingly bad science! The Trinovantes were much closer to the Continent than the Welsh or Irish tribes, and there is no reason to suppose that they did not have a much higher level of Continental gene marker frequencies than the Irish did even in 43AD. Urselius (talk) 18:22, 20 February 2014 (UTC)


 * About your specific point, I think that you are missing out the demonstrable porosity of Englishness. Once natives could 'pass for English' then any reproductive advantage attaching to Englishness as such would then be recruited to the propagation of their native genes.


 * I agree about the initial science and the amount of assumptions in the first studies are "polo-mint" shaped. I think the problem in the second paper is that they describe a uniform process.

It all goes to show that they research seems to prove the exact opposite to what they conclude J Beake (talk) 19:17, 20 February 2014 (UTC)
 * I think the first two generations must be different and the rate of population increase problem very large. Like "Americans during the war" - overpaid, oversexed, and over Here
 * Crazily they also assume the generations were 25 years - Lakenheath man who was 30 when he died in c. 500 should tell us that life was shorter and partnering multiple.
 * I agree with the Englishness point however to what extent did the Chromosomes follow this, in that what would have been better for a local landowner with daughters. In fact I think many fathers would have wanted daughters so they were linked to the new patrons. Suddenly genetically these families become "incomers".

New source
This article looks interesting and relevant, but I'll leave any editing to those more knowledgeable than me. Ghmyrtle (talk) 10:35, 27 February 2014 (UTC)

Thanks Ghmyrtle ... its seems the more strontium results we get the few non-local results we find. However the issue that Sam Lucy found is the period for earliest Germanic settlement seems to be earlier than 450. This means that of course people like the Lakenheath warrior who dies in 500 ad aged around 30 and who was located as local still could have been second or third generation. Yet every year the numbers lessen .. Mr Harke now thinks at the most 100,000 a few years ago he thought over a million. Personally I think the numbers could be between 10-20,000, but we need many more continental strontium results. I think the weak link in high figures notion is that they assume A-S genetics were just a little more successful. I think that is vastly underestimating. 10,000 with an increase of 2% per year (same as India today) makes 500,000 in 200 years J Beake (talk) 23:44, 7 March 2014 (UTC)

Plagiarism
Longtime reader, first time commenter. Be gentle.

An academic linguist, I'm brushing up my non-specialist knowledge of the history of England, having tumbled on Francis Pryor's Britain AD, which argues strongly against the Anglo-Saxon invasion & displacement story canonical in history of English texts. I had already heard of Oppenheimer's stuff, so I knew there was a fight going on somewhere, but hadn't found reports of the action. This page is a wonderful touchstone for me to explore from, so thanks to its creators and shepherds. However ...

Doing exactly that exploration, I found that a significant part of the section titled "'Saxon' political ascendency in Britain" is plagiarized from Nick Higham's article in History Compass 2:1, DOI: 10.1111/j.1478-0542.2004.00085.x. For instance, the clause "...Kenneth Dark, who has argued that Britain should not be divided..." appears verbatim in both. I haven't checked if any of the other sections on the page are plagiarized.

I actually like the text, in both places. It seems to present a balanced summary of two reasoned interpretations of the non-dispositive evidence. I doubt, though, that this is a WP-sanctioned method for achieving that end.

Now that plagiarism has put me over the tipping point into creating a user identity, I'll start trying to contribute where my specialty gives me standing. This, however, is not such an area. Can I leave it to those of you supervising?

Side issue: Francis Pryor reports (p 212-3) a tooth enamel study by Paul Budd & team that may be the one Urselius was thinking of: 10 of 24 dead in an E. Heslington Anglian cemetery came from west of the Pennines, but contra Urselius' memory 4 of the 24 were in fact from Old Anglia or Scandinavia. However, all 4 were female, and 1 was juvenile, a situation seemingly unaccounted for by the fostering/hostage hypothesis of language shift Urselius and J Beake were discussing. Still, addressing this conflict seems like original research rather than WP material. Frphnflng (talk) 21:35, 21 May 2014 (UTC)


 * As Dark is prominently named in this section, perhaps the easiest way to avoid plagiarism is to put the relevant passages in quotations. I do not have the work(s), so cannot check the wording myself. The fostering discussion was just me expressing incredulity that no one in the field has explored this obvious (to me anyway) method of acculturation. There was no intent to include it in the article itself. I can easily imagine the son of a British chieftain raised as a fosterling/hostage in an Anglo-Saxon king's retinue being thoroughly anglicised - if he then took over his father's lands there would be a single generation change of language and culture at the top of a British territory, with no blood spilled. Urselius (talk) 07:47, 22 May 2014 (UTC

Good to have your input. I think this is the only part of this section that has been there for more than a year. I think if it is a direct quote that was never properly ascribed, we can do as Urselius says. Higham, as often he does, sums up the situation well. As for your linguistic background, what do you think of PeterSchrijver, Language contact and the origins of the Germanic languages. This theory seems to provide a clear linguistic model.94.1.34.106 (talk) 08:15, 25 June 2014 (UTC)

oppenheimer,,,,again
I note earlier some protestation was (rightly) made about the weighting Oppenheimer (and Sykes) views are given in this article. To say that too much of a benefit of doubt has been given to him is to put it mildly. It could very well be that his 'autuchthonist- palaeolithic' views are dead wrong. Aside from a poor interpretation of the data, they are undoubtedly ideologically biased on the prevailing 'immobilist' zeitgeist in British scholarship. Coupled with: his view that R1b is some palaeolithic - "Basque" marker. Apart frmo the fact that there were no "basques" in the palaeolithic, the mounting ancient DNA evidence suggests that R1b spread through western Europe much much later. Certainly, isotopic evidence from the Bronze Age shows that whislt some settlements indeed showed overwhelmingly 'local' signatures' others (eg Cliffs End) had as high as 60% 'immigrant' proportions. Clearly, the Bronze Age, and even into the Iron Age, there was quite a bit of moving around going on, by specialist smiths and their families, etc, who were afforded special privillages, and perhaps even were responsible for the introduction of Celtic languages into Britain. Slovenski Volk (talk) 15:06, 6 October 2014 (UTC) .


 * The weighting seems quite even to me, Weale is given as much prominence as Oppenheimer and Sykes, more indeed as his material is presented first. There are as many opinions on the matter as there are investigators in the field. Urselius (talk) 10:25, 7 October 2014 (UTC)


 * HI Urselius. Thanks for your reply. I wasn't necessarily suggesting that more weighting should be given to Weale and more 'pro-migratory' papers; but rather, there is no mention of the drawbacks and pitfalls in all genetic studies dealing with topics such as this. Ie mention needs to be made of the type 'a' and type 'b' problems (to use Marek Zvelibil's critique on "genetic evidence") associated with egentics studies, problems with dating, stochasticity of haploid markers, the a priori & post hoc interrpretation of data, etc, which are part and parcel of the approaches taken by geneticists when dabbling with historical-archaeological reconstrcutions. I highlighted Oppenheimer simply because his results are the most widely known, but also the most outdated. There is no mention of recent anceint DNA results here. 152.76.1.244 (talk) 12:10, 7 October 2014 (UTC)

Hi, I think that any information directly linked to the subject of the article should be included, but reflections on the methodology of population genetics and ancient DNA studies probably are not directly germane. Personally, I think a lot of such work is badly designed and performed and the data dubiously interpreted, also such concepts as 'mitochondrial DNA clocks' and 'mitochondrial Eve' are pernicious and over-hyped. However, I think that articles more specialised in this field of work (such as Genetic history of the British Isles) are better suited to a critique of methodology. Urselius (talk) 12:40, 7 October 2014 (UTC)


 * You're quite right Slovenski Volk (talk) 13:34, 7 October 2014 (UTC)

Nothing on foederati
A paragraph begins: "Confirmation of the use of Anglo-Saxons as foederati or federate troops has been seen as coming from burials of Anglo-Saxons wearing military equipment of a type issued to late Roman forces which have been found both ..." but afaics there has been absolutely no previous mention of foederati at all, nor are they explained later. There wasn't even a link until I added it just now. I know this article has been through the wars, have the 2-3 paras on this needed fallen victim in an edit war? End of Roman rule in Britain is just as remarkably silent on the subject. Johnbod (talk) 12:06, 2 November 2014 (UTC)


 * A number of recent writers have cast doubt on the ethnicity of the people buried with military equipment. As ever, there was a circular argument involved - burial with equipment means not Christian, therefore means incoming Anglo-Saxon pagans. A view based on assumptions. The burials could have been merely of Roman soldiers or even Roman bureaucrats, who are known to have aped military dress. Much "chip-carved" metalwork, such as the Quoit Brooch Style, has been shown to have been provincial Roman, rather than Germanic, in origin. Urselius (talk) 14:47, 2 November 2014 (UTC)
 * Roman by background of the makers and most of the style, but still Germanic in terms of the owners, and at least some of the forms - per Webster, Leslie, Anglo-Saxon Art, 2012, British Museum Press, ISBN 9780714128092 etc. But this is completely beside my point, which is that the article does not mention what foederati were, or touch on debates about their role. And it should, as they have by no means vanished from the debate. Johnbod (talk) 16:37, 2 November 2014 (UTC)


 * Difficult to assign ethnicity except by molecular methods. Feel free to add material related to the Foederati angle. Urselius (talk) 17:22, 2 November 2014 (UTC)


 * Further, Agree with Urselius. The idea of Germanic federati pre-5th century is very much on the wane. It rested on dubious notions that belt-sets and weapons are "Germanic' cultural features. In fact, they were part of the new military style of the Late Roman empire. Of course, this is not to say that there weren't individuals of "Germanic" origin serving in the army, but their number has been likely been overestimated in the past. Well into the 4th century, the large marjority of the army were "provincial Romans", see  p 95-110. Slovenski Volk (talk) 05:08, 8 December 2014 (UTC)

Dating
It's all fine and good to point out alternative and more trustworthy sources from the Gallic Chronicle to archaeological findings to DNA examinations but we should still—somewhere in the article—provide the dates given or extrapolated from Gildas, Bede, and the Saxon and Welsh annals that formed the traditional understanding of these events. At the moment, we're pointing out that the Gallic Chronicle is "earlier" than Bede's "mistaken" date but not even providing it. We have two paragraphs on the nature and reliability of the Anglo-Saxon Chronicle(s) without any actual use of the text to discuss anything. — Llywelyn II   01:22, 18 February 2015 (UTC)

Molecular evidence
"Mitochondrial DNA (mtDNA for short) and Y-chromosome DNA differ from nuclear DNA ..." No, Y-chromosome DNA is nuclear DNA. — Preceding unsigned comment added by 139.68.134.1 (talk) 15:31, 13 March 2015 (UTC)


 * Quite correct. I have changed the relevant phrase from 'nuclear DNA, to 'diploid nuclear chromosomes', Y is always haploid wheras X is diploid in females. Urselius (talk) 21:06, 13 March 2015 (UTC)

New paper with ancient DNA
We need to look out for Iron Age and Anglo-Saxon genomes from East England reveal British migration history - Stephan Schiffels et al. due out in Nature. I have seen a pre-print and it is very interesting reading - but possibly not for Weale. The first systematic look at ancient DNA relevant to the Anglo-Saxon migration question. Urselius (talk) 15:27, 5 September 2015 (UTC)

Friesland vs. Frisia
Note that according to our articles Frisia and Friesland, the terminology is ambiguous: The modern province can also be called Frisia, and the historical region can also be called Friesland. Moreover, from the context, it is not at all obvious that the DNA samples have been obtained exclusively from the modern province in the Netherlands, rather than throughout the historical/geographical region (including places in Germany and Denmark where Frisian is still spoken, as well). --Florian Blaschke (talk) 23:39, 3 September 2015 (UTC)
 * I have checked the original paper, and the wording in the paper is "Friesland", not Frisia. As the relevant passage in the article describes the findings presented in the Weale paper, your preferences on nomenclature are not relevant. The usage in this part of the article needs to accurately reflect the usage in the paper; DNA samples were taken from natives of the Dutch province of Friesland, I quote from the paper: "we also  collected  DNA  samples  from  94 males in Friesland (northern Netherlands)". Obviously, Weale was not collecting DNA from natives of Groningen, or the German East Frisian Islands, or anywhere else in what at one time or other formed part of historical Frisia. I will return the article to its original, and correct, usage. Urselius (talk) 07:14, 4 September 2015 (UTC)
 * So when Frisia is mentioned later in the Oppenheimer paragraph, what is meant there? I find the terminology confusing. Replace by "Friesland", too? --Florian Blaschke (talk) 19:16, 4 September 2015 (UTC)
 * You are right, there is an inconsistency. The sentence with "Frisia" in it also refers directly to Weale's paper, rather than Oppenheimer's book, so 'Friesland' should be used. I have changed the term so that it now reflects the original paper and is consistent with the earlier occurrence in the article. Urselius (talk) 07:06, 5 September 2015 (UTC)
 * ✅ Thanks, that addresses my concern. --Florian Blaschke (talk) 00:39, 6 September 2015 (UTC)

Oppenheimer, again
Following up on Talk:Anglo-Saxon settlement of Britain/Archive 1, I must point out that the article references Oppenheimer as apparently (with reference to the worse-than-poorly-received among linguists work of Forster and Toth) endorsing some variant of the Paleolithic Continuity Theory, without the reader being given the slightest indication of how marginal that view is.

And marginal it is: In fact, it crosses the line from fringe science to pseudoscience (to avoid saying pure, unadulterated BS). That we don't have much in the way of criticism in the PCT article is just because linguists generally don't even bother refuting or even remarking on it – check the talk page over there to get straighter talk. Seriously, experts may disagree on a lot of things, but Indo-European breakup in 10,000 BP, (Insular) Celtic in 6000 BP or English earlier than 2000 BP is just not endorsed by any professional Indo-Europeanist (I'm aware of) except Ballester, and to pretend otherwise is incredibly unprofessional and flies in the face of N-HH's wise remark in Talk:Anglo-Saxon settlement of Britain/Archive 1: "Many WP pages have impressive footnotes but in fact are pushing fringe ideas or giving them equal weight with mainstream ideas. Too many WP editors have an overly simplistic view of sourcing policy and believe that if you can "source" a statement to a seeming authority, in it goes, without regard for overall balance, presentation or structure. Often that can lead to content that's as flawed and misleading as someone writing it all off the top of their head without any sources or footnotes. With this sort of topic in particular, where there's a lack of definitive evidence and a lot of disagreement among academics, it's incredibly difficult of course to ensure a balanced write-up." Moreover, since the article is concerned with the 400–800 AD period, the nutty Belgae thesis is beyond its scope, anyway. --Florian Blaschke (talk) 23:24, 3 September 2015 (UTC)


 * I don't buy Oppenheimer's hypothesis either, but it is out there, and as such is admissible to the article. Oppenheimer also performs one useful function, he highlights the lack of a reliable chronology in the the various population genetics works. That English SNP frequencies are a particular percentage similar to those of the inhabitants of Angeln or Friesland tells us nothing about when this similarity arose. Circular argument, pinning Genetics to particular historical events, is lamentably all too common. Urselius (talk) 08:00, 4 September 2015 (UTC)


 * I don't mind that, I already saw that point; but (per N-HH) the reader should not come away from the article believing that Oppenheimer's views, though widely publicised, are anywhere near mainstream. Neither fellow geneticists nor linguists take his views particularly seriously. Oppenheimer is pop-science, which makes it all the more important to heavily qualify his ideas. (Compare Talk:Celtic Britons. Talk pages are riddled with laypeople taking his views as cutting-edge science.)
 * As for circular reasoning, a general remark: especially in a field like this, some degree of circularity is hard to avoid, as undesired it may be methodologically (Popper pointed out that this is a general problem of the historical sciences; never mind the problem of induction). --Florian Blaschke (talk) 19:30, 4 September 2015 (UTC)


 * Speaking as a biologist, with a slight working knowledge of 'Ancient DNA', I would say that Oppenheimer's genetics is not as fringe as his linguistics. His population genetics are not hugely dissimilar to those of Sykes, who is considered as being academically sound - with the proviso that equally sound geneticists come to radically different conclusions through the use of different sampling strategies and data processing methods. I think that a sentence about Oppenheimer's linguistic theories being unsupported by academic linguists would be a useful addition (with a relevant citation). However, I would not support anything similar being added concerning his genetics. Urselius (talk) 07:24, 5 September 2015 (UTC)


 * On circularity, I would point to a recent paper that found the English to be 20-30% similar to the North Germans, the same paper also found the English to be ~40% similar to the North French. Whereas the French similarity was dismissed as being of prehistoric origin, the authors went to considerable pains to try to date the German connection to the Anglo-Saxon adventus. This they succeeded in doing, though their conclusions were only as robust as their mathematical model of the rate of diffusion of genetic markers allowed. If there had been a historical French invasion with, unlike the Norman conquest, rumours of large-scale immigration, I think they would have spent as much energy on trying to pin the genetics to this as they did for the Anglo-Saxons. Urselius (talk) 07:43, 5 September 2015 (UTC)


 * Our NPOV policy requires us to show significan minority views but also to show that they are minority views. Doug Weller (talk) 10:19, 5 September 2015 (UTC)


 * The Oppenheimer section could be cut back a good deal, with some commentary on the findings of Sykes introduced: *. Sykes also upholds the English are most similar to the Welsh and Irish viewpoint, and the author is considerably less controversial and has no linguistic hobbyhorses. — Preceding unsigned comment added by Urselius (talk • contribs) 12:48, 5 September 2015‎ (UTC)


 * That's fine to me. Just let me point out that even from a linguistic point of view, I don't oppose the idea at all that the West Germanic immigration (and thus genetic input/admixture) in Britain was only relatively limited. It sounds reasonable that the bulk of the population (and partly even the elites) were native and remained so, despite language shift. This interpretation is suggested by the historical sources and some linguistic clues.
 * Perhaps Latin America (especially Mexico), where an apartheid-like system (the colonial sistema de casta) was in place for a long time as well, provides a close comparison: There was substantial immigration (enough to effect language shift, at least – however much might be needed for that), but no numerically overwhelming one, unlike in North America. Morocco with its rather superficial Arabisation might be comparable as well. I'm not sure if early modern Ireland is a close comparison. However, the partial survival of the native language and culture, if only at the margins, is a common feature of these. A good and very useful essay on the reconstruction of prehistorical ethnolinguistic changes, using the example of Fenno-Scandinavia and the Saami problem, is found here.
 * As for Northern France, it bears keeping in mind that there does seem to have been substantial immigration from east of the Rhine. Your point that care is needed since we have no good handle on the ancient situation is very much true and well taken. More ADNA evidence is sorely needed for a fuller picture. --Florian Blaschke (talk) 00:27, 6 September 2015 (UTC)


 * The paper I have mentioned below has some really great content: 4 early Anglo-Saxon burials, all with Anglo-Saxon grave-goods, two with continental DNA markers, one with native "British Iron Age" markers and one showing a recent admixture of the two. This shows that early intermarriage occurred - no apartheid - and the 'native' had the most prestigious grave-goods, showing that native origins were no barrier to elite status. Urselius (talk) 13:57, 6 September 2015 (UTC)


 * Back on point: What kind of citation could be sufficient for this purpose? Pereltsvaig & Lewis (2015) appears to be a good and useful new book, though it does not seem to directly reference – let alone explicitly refute – Oppenheimer, either (I haven't read it). --Florian Blaschke (talk) 00:49, 6 September 2015 (UTC)


 * I think virtually anything from a reasonably academic source would be fine, a paper, a review in a jounal, or a mention in a book. Even a presentation at an academic conference, if it is available online. Urselius (talk) 13:57, 6 September 2015 (UTC)


 * Actually, p. 96 explicitly addresses the criticism of Forster & Toth. --Florian Blaschke (talk) 23:17, 7 September 2015 (UTC)


 * This might be a little too far removed from a direct criticism of Oppenheimer. Urselius (talk) 07:15, 8 September 2015 (UTC)

Britain below Hadrian's Wall
Ouch! My geography teacher would have scolded me for using such a lax term! "Britain SOUTH of Hadrian's Wall" is much more geographically "proper", I think... 81.99.236.225 (talk) 02:21, 19 December 2015 (UTC)

Title
This is a very important article, but the current title is unnecessarily convoluted and obscuring the article's importance. The event/process is easily identifiable as the Adventus Saxonum. I suggest the article is moved to that title. Deacon of Pndapetzim ( Talk ) 19:29, 19 December 2015 (UTC))


 * Unfortunately, not very many people can say Latine loquebar these days. The title really needs to be in English. Urselius (talk) 13:08, 16 February 2016 (UTC)

Population estimates
I find the first paragraph of the current section #Estimating continental migrants' numbers rather confusing.

Apparently we start by estimating the population of Roman Britain. From this an estimate is given of the population of Southern and Eastern England. (How? It doesn't say). Is the number of Britons at the time of the Adventus just allocated pro-rata from the total; or does it assume that the economic catastrophe in late 4th/early 5th century Britain went hand-in-hand with a population catastrophe? (eg according to Michael Wood the British population may have fallen from 4 million to 1 million, though a quick read of our article suggests we may be treating it as constant.)  The article doesn't explain, just that apparently this then helps estimate the number of migrants -- but why it should help seems completely obscure.

Then, from a completely different direction, one estimates the number of settlements 200 years later. (200 years after what? AD 410?  AD 449? AD 520? The article doesn't say). From this one estimates an Anglo-Saxon population of 250,000. (Can one similarly count British settlements? Were there 4 times as many?  Or were they invisible?  Or had the British culturally assimilated by this point, to be archaeologically no longer distinguishable?  Again the article is silent).

One then introduces a population growth rate of 1% p.a, and compounds this over 200 years to produce a growth factor of 7.3x; or alternatively, from 2% p.a. one gets a growth factor of just over 50x. Where these growth rates are presumed to come from is not made clear -- nor how they are supposed to be compatible with a total population that was still only 1.7 million in 1086. Replacement rates per generation might be a more transparent way of presenting the hypotheses -- eg 2% p.a. corresponds to a replacement rate of 1.65 for each person, for a generation length of 25 years. But perhaps all that these numbers are meant to show is that if there was some differential population growth, which then saturated, then this would make it even harder to estimate the number of migrants from any final figure, or any final DNA proportion?

Then there's a number based on cemeteries, but as noted here:, "what proportion of the actual historical total of burials is represented by the modern finds is really anyone's guess".

All in all, it seems to me that the writing of this section could use another look, to make the overall logic and the different assumptions a lot more apparent. Jheald (talk) 19:05, 15 February 2016 (UTC)


 * cf also various numbers at Anglo-Saxons.
 * The Wiley-Blackwell Encyclopedia of ASE pretty much ducks the whole question of population estimates: Jheald (talk) 13:01, 16 February 2016 (UTC)


 * Go ahead and overhaul the section, the whole population size estimation thing seems to consist of various people who claim some arcane insight just waving their arms about. Other than to say the natives must have hugely outnumbered any people rowing over the North Sea in glorified boats, I don't see the numbers game as being all that important. Urselius (talk) 13:16, 16 February 2016 (UTC)


 * I would feel happier to rewrite it if I had more of an idea what to rewrite it to; and what the original editors were trying to say. Jheald (talk) 13:35, 16 February 2016 (UTC)

According to genetic and archaeological evidence, only five percent of the British today have Anglo-Saxon ancestry. Is this helpful in contriving population estimates? Or should we ignore genetic evidence altogether in discussing population in the 5th through 7th centuries? Supposing the Anglo-Saxons constituted only five percent of the population at the time, how did so few of them dominate and/or conquer so many of the native Britons? — Preceding unsigned comment added by 45.47.192.105 (talk) 12:30, 14 April 2016 (UTC)


 * 38% according to the most recent genetics article. There is a huge range of opinion, a huge range of estimates concerning percentages of incoming DNA in the modern population. Do some reading, become informed. Urselius (talk) 19:18, 19 April 2016 (UTC)

@Everybody: How does "Knowing the number of migrants who came from the continent provide[s] a context from which scholars can build an interpretation framework and understanding of the events of the 5th and 6th centuries"? Why is population so important for this article? What are we trying to prove? Gordon410 (talk) 15:42, 26 April 2016 (UTC)

What is the subject of this article?
I would partition the transformation of Post-Roman Britain into Anglo-Saxon England into two phases. The first from c.450 to 550 AD, as a 'settlement phase' or an 'ethnogenesis phase' and then the second, from c. 550 to 650 AD, as an 'expansion phase', when Anglo-Saxon proto-kingdoms expanded from their coastal enclaves and took over the greater part of what is now England. If, as the recent changes to the lead suggest, the article covers both periods, then the article should be renamed - possibly to "The formation of Anglo-Saxon England". Urselius (talk) 12:14, 2 August 2016 (UTC)


 * Actually, perhaps the history of Britain in general should be divided into "pre-Roman Britain" including the various constituent influences on Britain's different regions, then "Roman Britain" (a relatively short period with varying amounts of influence on the regions involved from 43 AD to 200 AD, with a coda by 400 AD), and then sections on the Angles, Saxons and Jutes, with varying histories in various sections of Britain. The current "Britain is a single subject" implications are problematic, especially with the latest DNA studies. Collect (talk) 12:31, 2 August 2016 (UTC)
 * (ec) I don't really see why. Both phases involve "Anglo-Saxon settlement", and the obvious thing is to keep them together, although a clearer distinction can be made within the article. Johnbod (talk) 12:33, 2 August 2016 (UTC)
 * One problem is DNA studies which indicate some earlier genome (pre-400 AD) influx from "Germania"  might have occurred.  http://www.prospectmagazine.co.uk/features/mythsofbritishancestry  might be of interest. It posits that much of the "German" DNA was introduced before the Romans arrived in the first place. Collect (talk) 22:30, 2 August 2016 (UTC)
 * He's got a very long way to go pursuading specialists that Iron Age Britain, or England, spoke a Germanic language. Johnbod (talk) 02:48, 3 August 2016 (UTC)
 * The problem with studies of modern population genetics is that they all depend on mathematical modelling. The interpretation of data is therefore almost entirely dependent on how 'fit for purpose' the mathematical model is. Few geneticists are first rate mathematicians, so their models are quite often less than ideal. Research groups working on the same data, or very similar data, can come up with radically divergent interpretations. Ancient DNA work, on human remains from archaeological sites, is now beginning to address this problem. The view that Palaeolithic DNA is predominant in European populations has received recent blows, English DNA seems to be about equally split between Mesolithic hunter-gatherer, Neolithic farmer and Bronze Age Yamnaya steppe-herder origins. Urselius (talk) 07:24, 3 August 2016 (UTC)
 * A good (fairly) recent summary is Hedges in the The Oxford Handbook of Anglo-Saxon Archaeology 2011 at . His view is that molecular genetics will provide definitive answers, but not in the near future. I do not however understand why large scale emigration - for example to Britanny - is generally dismissed as a significant factor. Dudley Miles (talk) 09:09, 3 August 2016 (UTC)

As language is not genetically linked, to ask that one prove Neolithic groups in Germany spoke a "Germanic" language at a period when we do not even know who spoke any "Germanic" language is difficult. Collect (talk) 13:48, 3 August 2016 (UTC)
 * this^ Gordon410 (talk) 14:15, 3 August 2016 (UTC)

One of the reasons that the genetic work has produced so many variable scenarios is due to poor experimental design. If you take three populations say Irish, English and Frisian. Then down to purely mathematical reasons you are going to see that the English are closer to one or other of the two remaining populations. This is guaranteed, but does it mean anything? Not really. If you add an out-group, say the Finns, that have no historical connection to the other populations, this will give you more information on how close the relationship between the English, Irish and Frisians really is. If you then add, say, the Belgians or northern French, then it will give you a handle on what level of identity between the English and the Frisians is just the result of genetic similarity to populations from the near-continent and not to supposed episodes of historical immigration (i.e. the Anglo-Saxon settlement). That so many population studies have missed this fundamental aspect of good experimental design is very damning in my opinion. Urselius (talk) 07:32, 4 August 2016 (UTC)


 * "Germanic languages" did not even exist in the Neolithic, just like Romance or Slavic languages did not. This is silly. Also, not only the Germanic peoples were (partly) descended from Bronze-Age Yamnaya steppe-herders, so a Yamnaya steppe signal is not indicative of specifically Germanic heritage. AFAIK, Ancient Celts were genetically no less Indo-European than Germanic peoples. Finns are not a suitable outgroup in view of the strong Germanic admixture. Sardinians may be better. Can you comment here? --Florian Blaschke (talk) 18:09, 28 August 2016 (UTC)


 * Language is only as relevant as the experimental designers decide it should be. I do not defend this approach. Yes, the Celtic languages are Indo-European, no one here was suggesting otherwise. The Finns have, in a number of population genetic studies, been shown to be a useful, genetically distinct, out-group for non-Finnish Europeans to be compared to. Urselius (talk) 13:08, 29 August 2016 (UTC)


 * The point is that defining terms such "Germanic" or "Celtic" – as necessary in this context – without reference to language at all is not possible. After all, you need to know what you are looking for, and these terms only have clearly defined meaning in linguistics; the archaeological definition of "Celts", for example, often leaves out the linguistically Celtic inhabitants of Britain (because they were never called "Keltoi" in antiquity) and therefore is evidently useless in context. I don't know how a genetic definition of "Celtic", especially one that is helpful in context, without even any indirect reference to language should even work.
 * The debate revolves, after all, on the relative contributions of (possibly Romanised) "Celts" and immigrant "Germanics" to the "English nation" as it existed in, say, 1900. So, the only useful definition of these terms with regard to these issues is ethnolinguistic, referring to people who spoke either a Celtic or Romance language or a (West) Germanic language as their first language in Britain by c. 450 AD. (I understand that what happened in Britain after 800 AD, namely the Scandinavian contribution, is not really relevant to the debate.)
 * At the very end of the Neolithic, "Germanic" simply did not exist yet in any meaningful sense, so it makes no sense to talk about "Germanic" vs. "Celtic" with reference to the Neolithic; it's a pure anachronism. At the time in question, Central Europe was dominated by the Corded Ware culture, whose bearers are thought to have been largely steppe-derived and linguistically Indo-European, and are frequently surmised to have been the (linguistic, and largely genetic) ancestors of both the ancient Germanic- and Celtic-speaking peoples (and possibly others); in any case they were hardly exclusively ancestrally linked to the "Germanic" component of Europe. Chance is that the contemporary Indo-European dialect from which the Proto-Germanic language later descended is also the direct ancestor of Proto-Celtic. This is necessary to keep in mind as background information.
 * Distinguishing between "Germanic" and "Celtic" genetically is a priori far from trivial. The equation "Celtic" = Mesolithic forager / Neolithic farmer vs. "Germanic" = steppe-herder is fallacious. There's a good chance that "steppe" DNA was introduced to Britain from the Continent together with Celtic language (most probably in the Iron Age). This can very well be the very influx Collect mentions. --Florian Blaschke (talk) 14:57, 29 August 2016 (UTC)
 * Again, yes. What you say is correct. But what connection does this have with population genetics research that compares the modern populations in the British Isles and various parts of the continent? It is increasingly obvious that the last major genetic influx affecting Western Europe occurred in the Bronze Age, but this does not stop people trying to pin genetics on to subsequent historical events, those being the Anglo-Saxon adventus and the settlement of various Viking Norse groups in Britain. Urselius (talk) 07:55, 30 August 2016 (UTC)
 * What exactly is your problem with the article and what edit do you suggest be made? Gordon410 (talk) 19:42, 30 August 2016 (UTC)
 * I have no problem, I have no desire to make specific changes to the article. I am in a conversation with Florian Blaschke about the distinction between language and ethnicity. Urselius (talk) 07:28, 31 August 2016 (UTC)
 * Discuss this in a different place. "The purpose of an article's talk page (accessible via the talk or discussion tab) is to provide space for editors to discuss changes to its associated article or project page. Article talk pages should not be used by editors as platforms for their personal views on a subject." - WP:Talk_page_guidelines Gordon410 (talk) 11:25, 31 August 2016 (UTC)
 * It isn't the place for the exposition of very extensive screeds of own research, but that didn't stop you from using it as such. The connection, or lack thereof, between language and ethnicity is of central relevance to theories of English ethnogenesis. Urselius (talk) 12:29, 31 August 2016 (UTC)
 * If you are not discussing changes to be made, it should be discussed somewhere else. Gordon410 (talk) 15:32, 31 August 2016 (UTC)
 * I understand your question now. I opened this discussion with a desire to see the name of the article changed. However, the consensus of opinion was against this so I quickly dropped the idea. It is something that you could emulate, to everyone's benefit. Urselius (talk) 18:39, 31 August 2016 (UTC)
 * Honestly, I'm not seeing anyone here suggesting real, actionable changes in this discussion.--Cúchullain t/ c 15:40, 31 August 2016 (UTC)

Readability
I have pasted the lede into http://www.online-utility.org/english/readability_test_and_improve.jsp. It calculates, among other things, the Gunning Fog index. This is an indication of the number of years of formal education that a person requires in order to easily understand the text on the first reading. I don't think an encyclopedia article should require its readers to have 21 years of formal education. I appreciate that these indices are crude tools, but they agree with my perception of the article. The present version gives a very good encyclopedic account of the current state of academic discourse - if you have an understanding of the nature of the underlying evidence, and if you are prepared to comprehend charming, civilized, sophisticated, occasionally allusive, and remarkably opaque prose.

I suggest that a rewrite for readability would be very useful. It might also solve the current wrangle over the exact status of various theories. Unless I get any objections, I propose to do this, gradually, over the next few days. Richard Keatinge (talk) 09:18, 1 September 2016 (UTC)
 * Go ahead, but there is a tendency for simplified language to make uncertainties and nuances artificially more certain, and to make complex concepts, themselves, appear falsely clear-cut. This would have to be rigorously avoided. Dumbing down is not an option. Urselius (talk) 10:30, 1 September 2016 (UTC)
 * I support Richard Keatinge's proposition. I also agree with Urselius in that dumbing down should be avoided. I believe that we can work out an agreement and make some edits. Gordon410 (talk) 11:18, 1 September 2016 (UTC)
 * That this "might also solve the current wrangle over the exact status of various theories" is a very forlorn hope. Most WP articles on complex topics score very highly on these tests - anything that strays into talking about DNA or isotope analysis automatically gets a score suggesting only PhDs can read it. I suggest you concentrate on the lead first - that is all most people look at, which we are inclined to forget. If you are not actually working with the sources, I would be very cautious indeed. Johnbod (talk) 13:53, 1 September 2016 (UTC)
 * I don't think anyone is discussing DNA. We are mainly concerned with the lead and section 6.3. Gordon410 (talk) 14:52, 1 September 2016 (UTC)

Articles which are not understandable by readers, even if accurate, do not help anyone. Wikipedia is not intended for people with doctorates in a field to ensure that every word conforms to the correct cant of the filed, it is intended to explicate topics in a manner usable by the actual readers of articles. This article currently is slightly harder to read than the Wikipedia average. I suggest that we can improve it to a level where it is easier to read than the Wikipedia average without harming accuracy. Collect (talk) 13:39, 4 September 2016 (UTC)
 * 100% Agree. Gordon410 (talk) 01:52, 5 September 2016 (UTC)
 * The "Wikipedia average" presumably being the bio of a pornstar, or a village in the Ukraine! Johnbod (talk) 18:21, 4 September 2016 (UTC)
 * An irrelevant response. Gordon410 (talk) 03:36, 16 September 2016 (UTC)