Talk:Anglo-Saxon settlement of Britain

Gretzinger paper

• Gretzinger, J., Sayer, D., Justeau, P. et al. The Anglo-Saxon migration and the formation of the early English gene pool. Nature (2022). https://doi.org/10.1038/s41586-022-05247-2

There are some problems with the recent Nature paper that I would like to highlight, interesting though it is, and important as it is in adding further support to a substantial immigration from the North Sea littoral. To quote from Nick Higham "It is fairly clear that most Anglo-Saxon cemeteries are unrepresentative of the whole population ..." Looking at the paper, most of the remains investigated, originating from Britain, appear to be from before 700, and therefore are necessarily from pagan burials - this is emphasised by the many mentions of grave goods. This suggests to me that the investigators were looking at a sub-set of the population at the time, those buried in culturally pagan Germanic burial sites. Basically, if you look in hen coops, you will find lots of hens. The Iron Age in Britain, unlike the Bronze Age, is notoriously poor in burials, the remains only account for a small fraction of the Iron Age population. One estimate suggested that just 6% of bodies were disposed of in a way that would show up in the archaeological record. It could easily be argued that this lack of recognisable burials carried over within the native population into the post-Roman period, and therefore what the recent paper was looking at was a part of the population that included only incomers and those of the native population that had fully integrated into this imported cultural tradition. We are also left hanging concerning the apparent influx of substantial numbers of people with an Iron Age French (Gaulish) genetic profile, which does not seem to have or suggest a historic rationale. Urselius (talk) 09:14, 23 September 2022 (UTC)

Urselius, I have added a link. The paper is dated 21 September, roughly when would you expect to see comments on the paper from other researchers? TSventon (talk) 15:40, 23 September 2022 (UTC)

The usual 'newspaper column' comments are already out. Comments from other researchers take a bit longer, probably in a month or so. Urselius (talk) 16:26, 24 September 2022 (UTC)

Urselius can you give a link to Higham's comments? Dudley Miles (talk) 17:40, 24 September 2022 (UTC)

They are in the body of the text of this Wikipedia article. Urselius (talk) 18:11, 24 September 2022 (UTC)

Sorry, I misread you. I thought that your whole comment was a quote from Higham. Dudley Miles (talk) 18:26, 24 September 2022 (UTC)

The Leslie et al. paper also noted the French-like element to the modern British population, but ascribed it to pre-Roman immigration - the Belgae spring to mind. Interestingly, the Leslie et al. people claimed to have a dating method for analysing when genetic elements entered the British population. If the influx was indeed pre-Roman, the need for a historically unattested post-Roman immigration from France (Gaul) is gone. Urselius (talk)

It has long been known to mortuary archaeologists that furnished inhumation has nothing to do with paganism. This was first shown by Bailey Young and Halsall offers a summary in Worlds of Arthur, Chapter 10 for the extensive scholarship outlining why there is no inherent reason that it must have anything to do with paganism. There are many issues and there is much to be discussed with regard to this new paper, but let us at least make our objections based on contemporary understandings in contemporary scholarship. — Preceding unsigned comment added by 2A02:908:610:B640:98B:DCD0:1E90:F3F9 (talk) 12:45, 28 September 2022 (UTC)

Bede had a lot to say about Anglo-Saxon paganism and the process of conversion. He was a proponent of the Roman Church against the Celtic Church, but I doubt that his assertions about Anglo-Saxon paganism can be dismissed in their entirety, merely on the opinion of certain modern scholars. Urselius (talk) 11:26, 29 September 2022 (UTC)

Unfortunately, on Wikipedia we are more interested in what modern scholarship actually demonstrates (CF the Reliable Sources rule), than what we speculate can or cannot be dismissed in ancient sources with their own problems. 131.220.113.250 (talk) 11:01, 24 October 2022 (UTC)

If a bibliography is necessary for this point:

B.K. Young, "Merovingian Funeral Rites and the Evolution of Christianity: A Study in the Historical Interpretation of Archaeological Material (diss., Univ. of Pennsylvania, 1975); id., ‘Paganisme, christianisme et rites funéraires mérovingiens,’ Archéolo- gie Médiévale 7 (1977), pp. 5–81

G. Halsall, ‘La Christianisation de la région de Metz à travers les sources archéologiques (5ème–7ème siècle): problèmes et possibilités,’ in M. Polfer ed., L’Évangélisation des régions entre Meuse et Moselle et la Fondation de l’Abbaye d’Echternach (Ve–IXe siècle), (Luxembourg, 2000), pp. 123–46

G. Halsall, “Examining the Christianization of the Region of Metz from Archaeological Sources,” Cemeteries and Society in Merovingian Gaul (Leiden: Brill, 2010), 261–284;

G. Halsall, Worlds of Arthur (Oxford: Oxford University Press, 2013), 228–229.

D.M. Hadley, "Burial, Belief and Identity in Later Anglo-Saxon England," in Reflections: 50 Years of Medieval Archaeology, 1957–2007, ed. Roberta Gilchrist and Andrew Reynolds (London: Routledge, 2009)

Howard Williams, "At the Funeral," in Signals of Belief in Early England: Anglo-Saxon Paganism Revisited, edited by Martin Carver, Alex Sanmark and Sarah Semple (London: Oxbow Books, 2010)

Note e.g. Williams, who is a particularly regarded expert on the topic of Anglo-Saxon funerary ritual and its cosmological connotations, in the piece cited above. On the first page he states:

"The emergence of furnished cremation and inhumation graves is thus no longer regarded as reflecting a single and coherent ‘Anglo-Saxon paganism’; nor need the decline in accompanied burial relate directly or exclusively to Christian conversion (e.g. Geake 1997). Indeed, the very term ‘pagan Anglo-Saxon burial’ compounds the conceptually naïve assumption that there existed a one-to-one correlation between ethnic affiliation, religious beliefs and ritual practice that archaeologists have been so keen to move beyond (e.g. Carver this vol.). Therefore, on both theoretical and methodological grounds, pagan mortuary ritual is an area of study bedevilled with problems and recent considerations of religion that have sometimes avoided the burial evidence all together."

In the conclusion to the same piece:

"The mortuary rituals were not primarily about representation, nor were they primarily about symbolising religious or social concepts. Instead, they were mnemonic performances, enabling the living to transform the dead and to reconstitute the relationship between them." Also noteworthy: Williams has himself been challenged on the ability to show this commemoration of the dead/ancestors in ritual, by Halsall and Harland, among others.

The point here being, our arguments for inclusion of one piece, and discussion of its merits, should be based upon acquaintance with contemporary literature on these issues and the contemporary debates said literature is involved in, not our cherry-picked preferences and assumptions about what this society looked liked derived from decades-old scholarship. 2A02:908:610:B640:98B:DCD0:1E90:F3F9 (talk) 13:06, 28 September 2022 (UTC)

No one is arguing for the new paper to be ignored. You seem to have missed my main concerns with the paper, and be wholly engaged in a minor part of what I said. The beliefs of those buried in Anglo-Saxon burial grounds is almost an irrelevancy. What is relevant, is that their funerary goods were overwhelmingly of Germanic cultural forms and origins. I think that your ideas, with whatever textual support you can muster, are rather a minority view. Christianity tends to frown on funerary goods in general, barring the occasional devotional article, such as St. Cuthbert's cross. I do not think that this assertion can be seriously doubted. Urselius (talk) 11:16, 29 September 2022 (UTC)

It is interesting to note the very great difference between the Late Romano-British archaeological record, characterised by a shift from large numbers of settlements and few cemeteries, which are unfurnished burials, towards early Anglo-Saxon archaeology showing the opposite character with small numbers of settlements and many cemeteries of furnished burials. Cleary, S.E. (1993) Approaches to the Differences between Late Romano British and Early Anglo-Saxon Archaeology. Anglo-Saxon Studies in History and Archaeology 6, pp. 57-63. This suggests to me, that furnished cemeteries may not be representative of the population as a whole. Fewer settlements but more graves seems illogical, unless the number of Romano-British graves are unrepresentative of the population size to some extent, and that the same might be true for the native population in early Anglo-Saxon times. Urselius (talk) 10:08, 30 September 2022 (UTC)

Do you have any citations or evidence for the assertion that 'my ideas', which cover mortuary archaeology across the previous 20 years and offer a summary and historiography of approaches, are a 'minority view'? In Wikipedia we deal with citations, scholarship, and reliable sources, not wishful-thinking and speculative assertion. I refer you again to the Williams quote 'The emergence of furnished cremation and inhumation graves is thus no longer regarded as reflecting a single and coherent ‘Anglo-Saxon paganism’; nor need the decline in accompanied burial relate directly or exclusively to Christian conversion (e.g. Geake 1997).' He's describing a scholarly consensus there, not merely his own putatively 'minority' opinion (and it is far from minority) 131.220.113.250 (talk) 11:03, 24 October 2022 (UTC)

@Urselius: I finally got to spend some time on this. I think your remark is useful. It means that we should not write as if the samples of early medieval burials are a true early medieval population. We can't really know how representative they were, or whether the types of burials used might favour people with a certain ancestry. Read carefully I think the authours also do this. It is encouraging that the burials from the south sometimes seem be almost entirely non "Anglo Saxon" because it shows not all burial sites used were Anglo Saxon. The only way we estimate total genetic impact on the reproducing population is to switch back to modern data like Leslie, which they do; and maybe we should cite that. Other points:

• The authors do mention the Leslie article which of course did not use ancient DNA, and indicate why this study can go further than Leslie on some questions of timing.
• The iron age "French/Belgian" component is discussed in a lot more detail in the supplementary material. They a lot of work to exclude the idea that this was a pre medieval migration. They also don't think it can be caused by one single wave. The initial early medieval entry is seen as Frankish immigration via Kent. They think there might have been continuous trickle and other pulses over a long period. One obvious one they mention is the Norman conquest.--Andrew Lancaster (talk) 11:39, 19 August 2023 (UTC)

The Leslie paper used both modern population genetics AND ancient DNA. Urselius (talk) 22:19, 14 June 2024 (UTC)

Did they use ancient DNA somewhere in that article? There might be some mention buried in there somewhere but I don't see it. Anyway I don't think that changes anything.--Andrew Lancaster (talk) 10:53, 15 June 2024 (UTC)

I do not think Higham's 2004 paper should be cited against the 2022 paper. There has been enormous progress in research in the last 20 years. At this stage I think we can only report a very interesting reliable source and wait to see what support/criticism from reliable sources follows, not criticise according to our own non-expert views. Dudley Miles (talk) 15:22, 19 August 2023 (UTC)

Looking at the Gretzinger paper again, the figures for the genetic origins of the present-day English are interesting. The Nature paper (2022) actually says this: "We estimate that the ancestry of the present-day English ranges between 25% and 47% England EMA CNE-like [North Sea Germanic, AKA Anglo-Saxon], 11% and 57% England LIA-like [Native Celtic British] and 14% and 43% France IA-like [Pre-Roman Gaulish Celtic]". The first thing to note is that all the estimations have very large ranges, hinting at very considerable differences based on locality. Second, the maximum numbers are 47% (less than half) Germanic, 57% native British Celtic (more than half) and 43% Gaulish Celtic (less than half). The take-away message is that, whether native British, or incomer Gaulish, the majority of the ancestry of the present-day English people is pre-Roman Celtic and not Germanic.

Concerning their ancient DNA findings, they carefully make no claims about the entire contemporary population of the future England. They say that 76 ± 2% of the ancestry of specifically their sampled remains had Continental Northern European (Germanic) ancestry, not that the entire contemporary population of what later became England was of similar levels of Germanic ancestry. This is, I think an important caveat.

The Iron Age Gaulish descended incomers were presumably introduced to Britain under Frankish patronage and were also perhaps somewhat Germanised in culture and language. There were also 'Saxon' settlements of 5th-6th century date on the Channel coast of what is now France, especially in Normandy. These settlements may also have played a part in the export of Gaulish genetics to Britain. Urselius (talk) 07:55, 18 July 2024 (UTC)

@Urselius: might be better if you compare to specific parts of this article which you think need tweaking, but I find some of your remarks a bit hard to follow. A tolerance range can be be derived from the article, as can the estimates themselves, and indeed many other things. Should this non technical article really include discussion about tolerance ranges as opposed to simply making clear that these are rough estimates for now? If we did include anything about tolerance ranges how would we avoid this being OR? I am also not sure if the article mentions anything about Gauls? (As opposed to an unidentified population from somewhere near France, Belgium and western Germany.) You are of course right that the authors see evidence for regional variations, but again how far can we go into this, with research still so early, without this becoming OR? Concerning the most basic takeaway message the one which the authors themselves and many journalists drew was that, ignoring unclear details (which I think we also should for now?) there is now really evidence of a reasonably large movement of people, and not just a small elite group. Is there any reason for us to be opposed to that summary?--Andrew Lancaster (talk) 20:17, 18 July 2024 (UTC)

The paper uses the phrase "France Iron Age-like", certainly France in the Late Iron Age, as noted by Julius Caesar, was called Gaul. "Gallia est omnis divisa in partes tres". I think that the present wording gives the impression that about 75% of the population of what became England over the spread of the period covered by the sampled remains was of Continental North Sea Germanic origin, and the paper does not claim this, it only claims it of the modest-sized sample number it actually investigated (mostly from eastern England). This should be made clear. It would be useful to check the following assertion in the text, "The authors estimate the effective contributions to modern English ancestry are 41% north continental, 34% British Iron Age, and 25% French Iron Age", is correct?, or is it an OR-derived average, because it does not accurately reflect the wording in the paper, which is studiedly less precise and emphasises the local or regional variation by its use of ranges of percentages. A range, minimum to maximum, cannot in this context be used to create a meaningful average percentage for the population as a whole, as there is absolutely no indication of the real numbers of people involved. The geographic areas with the minimum values could hold the majority of the population, or vice-versa. This is very important! Urselius (talk) 20:27, 18 July 2024 (UTC).

I have just checked the Gretzinger paper, and could find no mention of the figures quoted in this Wikipedia article. Urselius (talk) 21:09, 18 July 2024 (UTC)

I'll make some more tweaks to try to address these concerns. Interesting point about the missing numbers. I presume the article has been adapted, so maybe we need to follow suit and use ranges. More generally: things like ranges are important, always, in any estimation, but they are also difficult to discuss. Whenever we say something is estimated then all this should be understood and I am all for generally careful wording. I think we should avoid the word Gaul, because we are talking about a much later period, and it would mean adding an original spin.--Andrew Lancaster (talk) 06:57, 19 July 2024 (UTC)

Mystery solved. Some of the information comes from supplementary data, which is an extra Word document. "Using ADMIXTURE, we estimate 40.7% ± 0.4% CNE, 33.9% ± 0.5% WBI, and 25.4% ± 0.4% CWE ancestry in England." I guess we should avoid using that file too directly for this article. (I think I adapted some of this text from work we did on a more genetics related article.)--Andrew Lancaster (talk) 07:21, 19 July 2024 (UTC)

The supplementary data also says this: "Summarising, we infer across all four qpAdm and supervised ADMIXTURE models approximately between 25% and 50% CNE [Continental Northern Europen] ancestry in modern English populations, which is consistent with previous estimates based on modern and ancient DNA56,120,137. Combining CWE-related [Continental Western European} and ancient CNE ancestry as “non-local ancestry”, we find that variation in nonlocal ancestry explains 99% of the variation in mean PC2 position of the tested English region populations in our Northwestern European PCA setup. We therefore conclude that in Britain, the genetic structure of the east-to-west (the “Anglo-Celtic”178) cline is mainly explained by introgressed continental ancestry that arrived within the last 4,500 years." I am somewhat at a loss to interpret the various interlocking sets of figures, which seem rather confusing, if not confused. The 4,500 year figure is puzzling for a paper concentrating on post-Roman populations. One paper, possibly Leslie, noted the French-like component in modern English population but their model attributed it to pre-Roman introgression. Urselius (talk) 08:49, 19 July 2024 (UTC)

Well I think we all agree that we need to take only the strongest, most relevant bits for this article. BTW we also have a passage in Genetic history of the British Isles.--Andrew Lancaster (talk) 12:11, 19 July 2024 (UTC)

Description of "small elite migration" theory as recently (but perhaps not currently) dominant

Seems a fairly minor point to bring to the talk page, but anyhow. My edit which tried to indicate that the "small elite migration" theory was recently widely accepted is being contested.

My understanding of the evolution of the scholarly views on this subject is that 1) mass migration theories were favoured in the mid-to-late 20th century, 2) a revisionist view from archaeologists and historians arguing for a much smaller migration became popular and arguably dominant in the early 20th century 21st century [typo; corrected on 2024-02-09], and 3) that the old mass migration narrative has become more accepted after 2020 or so, largely on the back of strong (ancient) DNA evidence.

Perhaps this account is missing some nuance. I observe for instance that on this very talk page there has been some disagreement on how exactly to integrate the ancient DNA evidence, so perhaps it has not been fully accepted as swaying the debate. But regarding the dominance of the "small elite migration" theory, some version of that point existed for a long time in the article with various wordings and supporting sources:

February 2015: "The view that the Anglo-Saxons arose from insular changes and developments, rather than as a result of mass migration and displacement, is now widely accepted." Supporting source: Härke, Heinrich. "Anglo-Saxon Immigration and Ethnogenesis." Medieval Archaeology 55.1 (2011): 1-28.

February 2019: "However, another view, probably the most widely held today, is that the migrants were fewer, possibly centred on a warrior elite." No specific supporting source.

March 2023: "However, another view, the most favoured among 21st-century scholars, is that the migrants were fewer, possibly centred on a warrior elite." Supporting source: Higham & Ryan 2013:104–105.

This last citation now lives at the end of the paragraph which I edited, which is why I did not add one.

It seems illustrative that 1) the "small elite migration" theory was formerly described in the article as "most favoured" but is now instead merely "another view", and 2) the current introduction ends on a fairly conclusive tone in favour of mass migration. To me this does imply that the breadth of acceptance of the "small elite migration" theory really is waning. And I think it would be useful and interesting to note this somehow; it's an interesting bit of historiography on a page which is framed as describing an ongoing debate. Perhaps my attempt at an edit was seen as inelegant or misleading, and someone else can better effect the gist of it. Massivefranklin (talk) 14:07, 8 February 2024 (UTC) ; edited to correct typo 11:11, 9 February 2024 (UTC)

• The key issue is not editors' views on the academic consensus, which is personal opinion, but what the source says. Neither version reflects Higham & Ryan 104-105 (and nor does the March 2023 edit). You say limited immigration is widely accepted in the 21st century. The source says the theory dates to the late 1980s and very tentatively endorses it, but does not say how widely it is accepted it is. The rest of the paragraph goes well beyond the source, refering to a warrior elite, intermarriage, settlement patterns and land use, none of which are in the source cited. Dudley Miles (talk) 15:22, 8 February 2024 (UTC)

• I agree that the key issue is not editors' views, but what the source says, so I asked for sources when I reverted the edit.

My understanding of the evolution of the scholarly views on this subject is that 1) population replacement theories were favoured in the early 20th century, 2) an argument for a smaller migration became popular in the later 20th century, and 3) that a larger migration narrative has become more accepted in the 21st century on the back of DNA evidence. Highham and Ryan is published by Yale University Press so it is a good source as of 2013, I don't know if an overview has been published more recently.

There are references to a warrior elite, intermarriage, settlement patterns and land use in the body of the article, although I haven't checked how well referenced they are. TSventon (talk) 03:38, 9 February 2024 (UTC)

That is my point. I have checked the source and warrior elite etc are not in the source so the references in the article are wrong. Dudley Miles (talk) 09:45, 9 February 2024 (UTC)

FWIW I think the DNA evidence is starting to settle down and be more respectable now, and this will start to be spread into different fields. Concerning the history of debate, I think some of the early genetic studies pushed for extreme replacement scenarios but those quickly became controversial. The consensus is once again somewhere in between the extreme scenarios. As already cited in our text the geneticists are now confident there was a major immigration into eastern England in post Roman times, contributing to about 40% of the modern English genetic make-up.--Andrew Lancaster (talk) 09:55, 9 February 2024 (UTC)

I wonder if Marc Morris's The Anglo-Saxons: A History of the Beginnings of England pp 34-35 might be an acceptable source for "state of the field" claims. The book is more of a pop history but he is fairly well credentialled.

Beginning in the 1960s, this view was subjected to a thoroughgoing re-evaluation. ... Instead of a mass migration, scholars developed the idea that Britain was invaded by only a few Saxons who were disproportionately powerful. ... Latterly the pendulum has swung back in the other direction, and the scale of migration is now once again generally reckoned to have been very sizeable. This revisionism has little to do with DNA.

(I am not sure that the genetic studies had such a small role in the debate - he seems strangely dismissive of the idea that genetics could ever have anything to say about history, and his only engagement with the genetics literature in the book is to breezily call it problematic - but if that's what the source says, I guess that I shouldn't contradict it. Perhaps notable that the book came out a year before the Gretzinger paper. But this is rather tangential, as my edit didn't explicitly ascribe changing views to genetic evidence anyhow.) Massivefranklin (talk) 11:51, 9 February 2024 (UTC)

Unfortunate typo (which I will correct) in my OP - I meant "dominant in the early 21st century", not "20th century". Massivefranklin (talk) 11:07, 9 February 2024 (UTC)

Massivefranklin your edit said a revisionist view that gained wide academic support in the early 21st century claims. I could support a view that gained support in the late 20th century suggests. "revisionist" and "claims" are loaded terms, "early 21st century" was an error and "wide academic support" was unsourced. Also the sentence about "Genetic studies in the late 2010s and early 2020s" added by Dudley Miles here is more important than the apartheid theory and could be moved to the start of the fourth paragraph. TSventon (talk) 13:22, 9 February 2024 (UTC)

I don't feel that "revisionist" is an especially loaded term these days, although I know of course that it has been used pejoratively. For instance, the source I suggested above uses the term as a neutral descriptor, and the Wikipedia article which you raise portrays it as often being legitimate. I also assumed that the fact that the wording survived for 8 years meant that the sources did support it, but I suppose that ideally I would have checked this myself. I'm happy to concede that the timepoint at which the elite migration theory tipped over to dominance was probably a little earlier than I thought, although I'm still not clear on which decade or two this happened. At any rate, I don't feel especially confident anymore that I'd be able to provide a phrasing of this point that the other participants in this discussion would be happy with. Massivefranklin (talk) 14:42, 9 February 2024 (UTC)

Massivefranklin, I have self reverted, then adjusted your text. Wikipedia tries to use neutral language, see MOS:EDITORIAL and MOS:CLAIM. Of course historians don't have to follow Wikipedia's Manual of Style. TSventon (talk) 17:39, 9 February 2024 (UTC)

It is increasingly clear that the anglicisation of lowland Britain was far from uniform in nature. In its earliest phases it was not a folk movement such as happened in continental Europe, where entire peoples, such as the Vandals, Visigoths, Burgundians etc. moved and settled as units, with more or less centralised authority figures in charge. Instead it was very piecemeal, with small folk groups arriving and settling in Britain, with kingship and kingdoms developing later. As such, the settlement pattern probably differed greatly from one district to the next. This does not even include the obvious differences between areas anglicised at an early date and those anglicised later due to conquest by Anglo-Saxon kingdoms, where whole British polities were absorbed in a single event. In the latter case, if not in some former cases, elite dominance almost certainly played an important part. The laws of Ine and their allowance for Wylisc (British) subjects of the West Saxon king, make this crystal clear. Urselius (talk) 21:04, 9 February 2024 (UTC)

I agree it probably wasn't uniform, but I would emphasize that one of the most important evolutions in academic thinking is also that we don't really know the details. There are no contemporary records which can really help, and so many older academic works which sound very certain about their position did not really have much evidence to work with. In such a situation it is no wonder that academics have historically swung between extreme ideas. By the way I'm also not convinced that we really have a clear idea about Vandals, Visigoths and Burgundians being "entire peoples", although I think what you mean in these cases is that there were definitely records of large organized groups moving at one moment. That of course does not mean that these groups represented entire peoples (with none left behind in other regions so to speak), nor even ethnically homogeneous groups. Alaric's Visigoths are for example generally seen as a very mixed group of military families ruled by Alaric, who can also be seen as a Roman military leader. The Burgundians who were moved into Burgundy came from a region on the Rhine frontier and were clearly connected to the Roman military. They were very far from the classical Burgundian homeland. Also in England it seems likely that Roman approaches to military recruitment in Northern Europe were quite likely to be involved in whatever happened. The way in which military populations could apparently be identified as peoples by the later Romans is something which is now difficult to work with or reconstruct.--Andrew Lancaster (talk) 08:36, 10 February 2024 (UTC)

The Goths, on linguistic grounds at least, left behind populations in Scandinavia and the Crimea. That the major migratory 'Germanic tribes' included other barbarians, including Sarmatians, and disaffected Roman provincials and former Roman soldiers is well attested. However, this does not affect the major difference between continental Germanic settlement of large, organised groups under some form of centralised authority (be it a charismatic king or federated chieftains), and the piecemeal settlement of small independent warbands and folk groups that seems to characterise the earliest phase of Germanic settlement in Britain. Though there is later description of kings in this period they seem to be back projections and rather mythic in character, like two brothers named 'Stallion' and 'Horse'. Urselius (talk) 09:37, 10 February 2024 (UTC)

The Continent

Dudley Miles, the text currently has a mixture of "the Continent", "the continent", "Continental Europe" and "continental Europe". It is likely that some non-British English speakers are confused by "the Continent". My suggestion is to change the first mention of "the ... Continent" to "continental Europe" and later mentions to "the continent". As you reverted my edit, can you say what you would suggest? TSventon (talk) 18:50, 13 February 2024 (UTC)

That would be incorrect I think. "the continent" should still be "the Continent" if meaning mainland Europe even with a first mention of CE. I agree for non-British English speakers (and probably many British English speakers too) this is confusing and suggest just using "mainland Europe" throughout. DeCausa (talk) 18:59, 13 February 2024 (UTC)

"The Continent" is standard and a widely used term. I do not think that is more confusing than many other usages, but I agree that it is better to use "mainland Europe" or "continental Europe" throughout than to argue about the issue. Dudley Miles (talk) 19:10, 13 February 2024 (UTC)

How the article looks

At the moment this article is clearly dominated by "meta" discussion about different types of evidence. I have started working on a short skeletal history before these discussions, so that there is context. In the background I am thinking a lot about how we should better divide this topic up between the 3 over-lapping articles we currently have. (See the hatnote.) Some discussion has started on the Anglo Saxons talk page about this. Andrew Lancaster (talk) 20:55, 14 June 2024 (UTC)

Short update. Unless someone else wants to try, or if there are good counter-arguments, I expect to eventually trim a lot of this article. The article is dominated by relatively unstructured discussions about the history of publications about the evidence, rather than short summaries of the evidence and its implications for the topic of the article. I think the article needs not only trimming but also restructuring, so that for example different types of evidence can be discussed together in sections about specific topics/questions, such as where the immigrants came from, or how many came. If anyone has concerns with my thinking can you please register them? A related question is whether anyone can think of other articles where some of the excess materials should go. Potentially we could even consider creating new articles I suppose. For editors in the future looking for deleted materials then this post might also become a handy reference point for the approximate date before some materials were trimmed.--Andrew Lancaster (talk) 10:13, 18 June 2024 (UTC)

Andrew Lancaster, I think the current structure makes sense, but am willing to be convinced otherwise. Do recent reliable sources agree on answers to questions like where the immigrants came from, or how many came? If they do, are they available through the Wikipedia Library? TSventon (talk) 09:54, 19 June 2024 (UTC)

The "meta" problem might not only be the structure. I will also try to keep an open mind to different approaches but I wanted to put out a warning and call for comment. I take your second sentence as a distinct question. In answer I think no one knows exactly where the immigrants came from but there is a fair bit of consensus about what the available evidence is and what it indicates. In a rough way we can say most came from areas between the Franks on the Rhine and Denmark. I think that Bede's two passages on this topic are the basic starting point for most discussion. An interesting side issue is the question of how many such people already lived in Britain as Roman soldiers before the Saxons actually started being a power in Britain, and not just raiders, so that should always be mentioned. A more recent type of evidence that needs consideration is the latest DNA evidence. And so on. I am not sure if this answers your questions?--Andrew Lancaster (talk) 11:01, 19 June 2024 (UTC)

For an evolving topic, and this is certainly evolving, a 'history of concepts, evidence and conclusions' approach seems useful to me. As a definitive narrative is impossible to present, here or anywhere else, throwing out the baby with the bathwater needs to be avoided. Urselius (talk) 13:05, 19 June 2024 (UTC)

Yes it is not my intention to create false certainties if that is your concern. But when I say this article is too "meta" I do NOT mean that it is giving a straightforward summary of existence. If we can get something like that I think it would be an improvement. Hope that clarifies.--Andrew Lancaster (talk) 13:26, 19 June 2024 (UTC)

Andrew Lancaster, to go back to my questions, The Anglo-Saxon World of 2013 has a chapter on The Origins of England, mostly available via Google books, which looks at various forms of evidence, starting with written sources, which is what our article tries to do. It ends with some tentative conclusions, but different scholars would draw different conclusions. So I think the current structure makes sense, but am interested if there are more recent surveys that cover the material as you are suggesting. The origin of migrants was just your example, but the Gretzinger 2022 study suggested migration from France as well as Netherlands/ Germany/ Denmark. TSventon (talk) 23:16, 19 June 2024 (UTC)

Yes the Gretzinger study shows migration from France/Belgium, or somewhere in that direction, but note that it is also a later-starting and long-term stream, so not necessarily relevant to pre 1066. (It is also a migration they were not expecting and so it will need more research to really understand it.) Concerning the structure we do not necessarily need to follow a source, because structure is an editing decision and what works in a book won't necessarily work here. OTOH I am not necessarily opposed to going through different types of evidence, and I'll keep looking at it. For me it at least seemed important to precede such specialized discussions with some kind of introductory section which gives the reader a basic understanding of the historical context before they are have to dive into technical details. I've started with that. So in terms of tweaking the structure, within that section I have already discussed some of the basic information found in texts, because it is after all the starting point of most other discussions. I am thinking that a specialized texts section should be about more complicated and uncertain points. I still need to work on that texts section.

Looking further ahead than the texts section, the article is not really structured based on types of evidence. There are big sections called "Migration and acculturation theories" and "Aspects of the success of the Anglo-Saxon settlement" which both had several sub-sections. It is difficult to follow how these section names were chosen and what should go in them. This seems to me to have led to duplication (the same questions such as estimates of the migrant numbers, handled more than once) and a very UNstructured feeling when you try to read through the article, like reading a collection of notes. Are those sections based on Higham and Ryan?--Andrew Lancaster (talk) 07:54, 20 June 2024 (UTC)

Perhaps I also should point to sections like the one on the Tribal Hideage and Old English Poetry, which never seem to come to any point. These seem to be just quick notes about types of evidence which exist, and which someone might possibly think could be relevant to settlement, although there is basically no relevance. So I guess they were added on the basis of wanting to follow this type-of-evidence-based structure, but it seems to have been taken too seriously. I am thinking that such sections are not needed in this specific article about settlement. They might well make more sense in a book length treatment of the whole Anglo Saxon period.--Andrew Lancaster (talk) 10:01, 20 June 2024 (UTC)

Andrew Lancaster, I agree with Alarichall that what you are doing is useful. I am not trying to defend every detail of the existing structure: the "Migration and acculturation theories" and "Aspects of the success of the Anglo-Saxon settlement" sections follow the evidence sections and could be structured differently. I didn't say that those sections followed Higham and Ryan, just that a chapter in H&R has sections on different kinds of evidence, so that is one valid way of organising the material. For what it's worth, I would keep the Tribal Hideage and and lose the Old English Poetry section, it depends where you draw the line between this article and the Heptarchy. TSventon (talk) 09:47, 21 June 2024 (UTC)

Indeed, keeping tribal hideage in this article sounds reasonable, but possibly not in a small sub-section about minor sources, but rather in a section about what we know (from various types of evidence) about the earliest Anglo-Saxon communities. But this is also just one idea. I haven't gotten to it yet.--Andrew Lancaster (talk) 11:06, 21 June 2024 (UTC)

It seems fairly clear to me that there won't be "other articles where some of the excess materials should go". If you set up a new one, what might that be called? I suppose Historiography of the Anglo-Saxon settlement of Britain. To me this seems rather a sideshow to the main Anglo-Saxons "problem". Johnbod (talk) 13:00, 19 June 2024 (UTC)

You are probably right about that. However, my arrival at this group of topics started at the early end of the Anglo-Saxon story, and I am working on that first (also at Saxons).--Andrew Lancaster (talk) 13:26, 19 June 2024 (UTC)

Thanks for starting this discussion, @Andrew Lancaster -- and for being willing to work on such a challenging article.

I agree that finding ways to trim this massive article would be good. It's hugely challenging though, and I absolutely agree with @Urselius that the article should try to provide an up-to-date mapping of where debates on this subject are at while recognising that we can't provide a definitive history. I've often thought that it would be good to have a day-long academic symposium about this article, bringing together experts on the various different aspects of this topic, at which people could hammer out what the status quo is in different disciplines and work out how best to present current thought. But I've never tried to organise it!

I agree that reorganising the article around key questions/debates rather than types of evidence would be good (though it wouldn't be easy!).

For what it's worth, when I worked on the "Linguistic evidence" section a few years ago, I decided that it was best to write the separate article now called Celtic language decline in England, and include a much briefer account of that material in Anglo-Saxon settlement of Britain, and I agree that spinning out material into other (existing or new) articles in a similar way would be a good way to slim down Anglo-Saxon settlement of Britain. For example, as archaeology has grown more useful and important, the portrayal of the settlement in our textual sources has become relatively unimportant to current thinking. The sections on "King-lists and the Anglo-Saxon Chronicle" and "Old English poetry", which would have been major parts of this article if it had existed a hundred years ago, are really just saying "no-one thinks these are reliable sources any more" and belong more in an article like Anglo-Saxon migration myths and/or @Johnbod's suggested Historiography of the Anglo-Saxon settlement of Britain. The stuff on Bede could go into an Anglo-Saxon migration myths article too, though a separate article on Bede's portrayal of the Anglo-Saxon settlement of Britain would certainly count as notable. The stuff on Gildas could be adapted to be integrated into De Excidio et Conquestu Britanniae (though since Bede makes a lot of use of it then it might also feed into coverage of his account).

I'm less confident about how the "Archaeological evidence" section might best be spun out and slimmed down because I don't know that material so well, but "Settler evidence" could certainly be rephrased more as "Role of Germanic-speaking foederati", perhaps with some material spun out to Foederati or to a stub like Foederati in late Roman Britain; "Re-use of earlier monuments" could certainly be the basis for an independently notable Reuse of monuments in Anglo-Saxon England; and the material in "Cemetary evidence" is increasingly being superceded by archaeogenetic evidence and might belong more in a historiographical article?

We might eventually produce regional sub-articles as well: Anglo-Saxon settlement of the Thames Valley, Anglo-Saxon settlement of East Anglia, and the like. There would be no problem demonstrating the notability of those topics, and those articles might make it easier to produce a more regionally nuanced account in the main article. Alarichall (talk) 12:19, 20 June 2024 (UTC)

Thanks for the feedback. Perhaps you will already with me that spin off articles are for now just a possible plan B. They might not be necessary and if they are then their specific topics are not yet clear because we should first improve the articles we have so that we can see what we have, and what is missing.--Andrew Lancaster (talk) 13:15, 20 June 2024 (UTC)

Yikes, @Andrew Lancaster: I've just started looking in detail at your edits. While I absolutely recognise that you're doing a lot of diligent work, and that the previous version of this article has a lot of problems, I'm not convinced that your edits are an improvement. Just commenting on the lead: I don't think it's at all appropriate to make Gildas's account as prominent as you do. Historians and archaeologists have done an awful lot of work over the last couple of centuries to show how problematic it is for us to rely on a source whose provenance we know so little about. The previous lead, which focused on spelling out what the current and recent scholarly understanding of the article's topic is, was much more in the right direction, I think. Likewise, in the opening of the section "Late Roman Britain and the Saxons", you seem to be trying to weave your own account of the past by bringing together our scanty and problematic primary sources, together with a few secondary sources. But I think that, with this topic, our job is to recognise that current research tells us that we can't put a coherent story together from the written sources in this way. I'm sure on some points of detail you've made improvements and I wouldn't want to just roll back your edits, but so far overall I don't think your approach is working well I'm afraid. As I suggested above, sources like Bede and Gildas belong more to the historiography of this topic than to the current state of knowledge.

Do you know Robin Fleming, The Material Fall of Roman Britain, 300-525 CE (Philadelphia: University of Pennsylvania Press, 2021)? I was able to download it easily from Z-library. It's quite up-to-date (except that it doesn't really tackle the new archaeogenetic work that is going to really change our understanding of migration in late antique/early medieval Britain) and indicates how differently people now are conceptualising the (so-called) 'Anglo-Saxon settlement' from the old Bede-based narratives. If you haven't seen it you might find it useful. Alarichall (talk) 09:42, 21 June 2024 (UTC)

I'll have a look for Fleming. Please be a little bit more clear about what your new concern is? I've been using Gildas a little bit to structure recent versions the lead and I'll reconsider that, but I've been using Halsall and Springer as secondary sources, because both of them give critical perspective on the overall debates and sources. The remarks in the lead about what Gildas reported, as the only near-contemporary source, are attributed. Secondary sources all seem to do something similar? In terms of WP style keep in mind that the lead should summarize the body, and the body contains a discussion about the account of Gildas (as it did previously). Also just to repeat I think the previous version of this article, let's say 31 May or earlier [1], were not work as a coherent article, because they were (I think) apparently attempting to be so neutral that it looked more like a listing of topics for possible discussion. The old lede said for example The few literary sources tell of hostility between incomers and natives. They describe violence, destruction, massacre, and the flight of the Romano-British population. I don't think we need to be quite so vague? It also distorts the source (not a plural), which despite its dramatism clearly reports that the British pushed back, won, and then experienced a generation of relative peace, during the lifetime of the writer himself. I think what scholars find questionable concerning his own period is perhaps that he understates/ignores any continuing importance of the people we now call Anglo-Saxons, who the archaeologists and linguists presume to have taken hold of large parts of the country by now, but his text does seem to narrow down the possibilities of what kinds of conflict were happening. --Andrew Lancaster (talk) 11:03, 21 June 2024 (UTC)

Thanks for the useful source. It focuses very much upon material cultures which is not what I was planning to work on at this time. I can see that this work is very much into the listing of doubts, and in that respect I think it is helpful but not necessary a "template" we can use - at least for the lead or the sections about textual evidence, which I have been working on.--Andrew Lancaster (talk) 12:11, 21 June 2024 (UTC)

What I'm trying to make more clear in recent edits is indeed that Bede-based narratives are the basis of a lot of old ideas that are now doubtful. Perhaps we read Gildas differently, but I don't think he gives quite the same problems. He does not say mythical-looking things (no King Arthur, not even Vortigern) and he is a near contemporary. He just doesn't say much about our topic, because he is interested in other things. That's also how I read the secondary material on him. Just for discussion, I doubt he would call someone a Saxon, which to him implied a violent lifestyle, just because of language or fashion, which is what people have been doing with archaeology. He'd probably be more worried about whether they are pagan (as would Bede probably, who uses the term Saxons mostly for the early pagans as well). So as Fleming says near the end, in his time the Anglo Saxons of Bede's time were still being invented. In many parts of the country people might not yet have been conscious of being in such a category. From our retrospective point of view there might have been many local "proto Anglo Saxons" around Gildas, but he would not necessarily equate them all to the Saxons involved in the violence in his grandparents' time.--Andrew Lancaster (talk) 14:48, 21 June 2024 (UTC)

I agree you've made some improvements to the lead -- thanks! And I agree that the article before you started working on it was really problematic. And I think agree with what you're saying in your most recent comment, though I haven't had a chance to look at your actual edits yet. Going back to the prominence you've given Gildas in the lead, though, which I am still alarmed by: I think you have access to Nicholas J. Higham and Martin J. Ryan, The Anglo-Saxon World? Pages 57-62 have a handy essay by Higham explaining why we can't use the De excidio to provide a framework for a history of fifth-century events. The De excidio is an interesting source for the sixth century for the accidental information it conveys: that a Briton was good at Latin, that he knew about money, etc. But that isn't what you're trying to use that text for. I really think this needs fixing. Alarichall (talk) 17:22, 21 June 2024 (UTC)

I am of the same opinion, Gildas was writing a religious polemic, a very long sermon, not a history, his work cannot be used to create a narrative history of the period. Much as Pilgrim's Progress would make a poor history of its time. Besides, Wikipedia frowns on using primary sources for that sort of thing, reliance on secondary sources, preferably recent, is much the better approach. This stricture, concerning synthesis from primary sources, would also apply to Bede, for similar reasons. Urselius (talk) 18:49, 21 June 2024 (UTC)

Thanks to both of you. I'll look at Higham and Ryan, but I hope there is no synthesis. As mentioned, secondary sources I have seen take the certain parts of Gildas seriously. (Obviously sources like the Fleming book about material evidence and deliberately avoid discussing texts can't be used to give a full report in an article like this.) Gildas gives our best written report of what happened, and historians may not and do not ignore surviving contemporary sources because to do so would be cherry picking. As stated, I also try to make sure his reports are attributed and not stated as facts. Remember there is no other alternative textual source from this period. Bede, who is the basis of many respected secondary sources, and will guide this article from behind the scenes if we do not discuss Gildas, is surely the bigger concern. As Urselius says, Gildas made something like a sermon and that is obviously part of the reason why there are very few parts of the work which are relevant, but there are specific parts of the sermon which give some "unpoetic" information. I also don't think our secondary sources allow us to ignore these parts of Gildas, and I am not sure what is meant by it being used as a "framework". Perhaps please look at my actual edits and try to explain this in a more concrete way? --Andrew Lancaster (talk) 20:23, 21 June 2024 (UTC)

By the way, I am suspecting that like me in the past you might be misunderstanding how Gildas writes. It is nothing at all like Pilgrim's progress! (And I've never seen a specialized secondary source treat him this way.) I've been looking into it more as I've been editing (which means also the more specialized secondary sources). For your reading pleasure here is a bi-lingual edition which we have not yet linked to. You might be surprised. https://archive.org/details/gtu_32400006703627 I think in the past negative attitudes about Gildas sometimes came from scholars who supported the Bede narrative, which Gildas undercuts.--Andrew Lancaster (talk) 20:36, 21 June 2024 (UTC)

Gildas and Bunyan both produced works of primarily religious content with the overt intent of improving the morals of others. Those are two definite similarities. Urselius (talk) 05:28, 22 June 2024 (UTC)

Andrew, I also agree with Alarichall's concerns about overreliance on Gildas. I have had a look at Halsall 2013: he has speculatively identified Gildas' proud tyrant with Magnus Maximus and the source page I checked (218, added here) repeats words like "could" and "hypothesize", but our article doesn't.

Also Robin Fleming, The Material Fall of Roman Britain, 300-525 CE (Philadelphia: University of Pennsylvania Press, 2021) seems to be available in chapters via the Wikipedia Library and de Gruyter. TSventon (talk) 20:57, 21 June 2024 (UTC)

TSventon I have not mentioned anything about Halsall's speculative proposals. I've already mentioned that I found Fleming's work as well. It is about material cultures and avoids discussion of texts. I am wondering if the concerns about overreliance upon Gildas are based on anything I have actually put in the current version of the article or just a general theoretical concern?--Andrew Lancaster (talk) 21:12, 21 June 2024 (UTC)

Alarichall I'm looking at Higham and Ryan and not seeing anything relevant to the way I've used Gildas, but tell me if I am missing something. I think what I have is comparable to p.62 which is their summing up. I've not used Gildas for dates for example, and I have deliberately de-emphasized the common but criticized idea that the invitation of the proud tyrant was (as in Bede) some sort of special large (or first) turning point migration. (Gildas really does not imply that at all as many secondary sources point out. Higham and Ryan apparently don't realize this idea is from Bede, not Gildas. Halsall helped me see that Gildas is sometimes read in the light of the later writer who he influenced. In any case I hope I've avoided this problem.) The minimal useful information which Gildas does really give, such as the existence of the letter to Aëtius (which he directly quotes from) and the fact that there was subsequently an invitation to Saxons (at some point, which he does not date) and the subsequent period of conflict,and subsequently a period of peace, are however hard to ignore! If we remove these from the article would that not be very controversial indeed?! If this is a "narrative history" or "framework" then so be it, but Higham and Ryan also seem to accept that much?--Andrew Lancaster (talk) 21:12, 21 June 2024 (UTC)

Hi there, @Andrew Lancaster.

• Re "Gildas gives our best written report of what happened, and historians may not and do not ignore surviving contemporary sources because to do so would be cherry picking". I'd suggest that what (modern) historians do is evaluate all their sources, and if the best one (which may indeed be the De excidio) isn't reliable, then they recognise that they can't write a reliable history. Historians don't just pick the least bad source and suppose that because it's the least bad source it's a good source. In the case of the topic of this article, archaeological evidence is these days far more informative than the De excidio anyway (and highlights the problems with its narrative), and we could structure the lead around that.
• I see the source on Gildas that you've linked to is from 1899; the edition and translation people currently use is Gildas: The Ruin of Britain and Other Works, ed. and trans. by Michael Winterbottom, Arthurian Period Sources, 7 (London 1978). If you'd like to get a sense of what recent scholarship on Gildas is getting up to, I'd recommend Stephen J. Joyce, The Legacy of Gildas: Constructions of Authority in the Early Medieval West, Studies in Celtic History, 43 (Woodbridge: Boydell, 2022) (which I see can be downloaded from libgen). You'll approve of Joyce's argument that modern historians have privileged Bede over Gildas in ways that are weird, given that Gildas is the earlier source, and you might be quite excited by his argument that Gildas was writing in the fifth century—but you'll also see that Joyce is more interested in using Gildas as evidence for the British church and politics in Gildas's own time and after rather than for the settlement of Britain.
• I don't think Urselius thinks that the De excidio is actually like Pilgrim's Progress in content: I think the point is that both texts construct narratives about the past not in order to give later historians reliable evidence but to achieve a moral/religious/political purpose. In a minute I'll try and pull out some of the quotations from the Higham essay that I suggested to indicate what I see as his warning signs about making much use of the De excidio as a source for the migration by Germanic-speakers :-)

Alarichall (talk) 22:13, 21 June 2024 (UTC)

I think Higham's key point on p. 62 is (with my emphases) "whilst Gildas cannot be precisely dated or located, his writings nevertheless offer important evidence regarding what one man of this generation thought about his own past and his is a valuable guide to insular culture in his day". Higham's point is the one I've been trying to articulate above, that we can learn a lot from the De excidio about interesting things like how (some) ?sixth-century Britons saw their past, how intimately they knew the Bible, how good their Latin was, or what they thought a good king should be—but not how Germanic-speakers' migrations to Britain or their political expansion actually happened.

Here are some of Higham's key warnings against using Gildas in that way: "Despite Gildas’s own focus, however, scholars down the ages have paid more attention to the schematic review of Britain’s history down to the year of Gildas’s birth which prefaces his main complaint – the so-called ‘historical’ section which forms chapters 4–26 of this 110-chapter work. These passages provide the only extant near contemporary insular account of the ending of Roman Britain and the settlement of the Anglo-Saxons. But Gildas never set out to write history as we would understand it today; he offered no dates and his story is so hemmed around by his condemnation of the British leaders as to be close to incoherent. His historical section was designed to establish that there was an inescapable relationship throughout history between sin and divine punishment" (p. 57). "This may be the only history that we have from the period, but the historical section is clearly mistaken or ill-informed in important respects ... It is when, in the fifth century, Gildas is our only guide that these problems threaten to overwhelm us" (pp. 58-59). "... the work itself clearly presents considerable problems, which are compounded by our relative ignorance about the author" (p. 59). Alarichall (talk) 22:34, 21 June 2024 (UTC)

@Alarichall: I think all this is fine in principle, and this has been helpful, but after the chat about those principles can I ask if you actually see anything in what I have written which involves synthesis or undue reliance upon Gildas? I've been trying to keep it minimal, and I've been trying to avoid over-reliance on Bede. Secondly, as to what Higham would do, actually there is a Gildas-connected speculation of his still in the article which I am thinking of removing, and that is the proposal that after the war mentioned by Gildas (which he clearly accepts) the Britons were paying some sort of tribute to the same group of Saxons. I am not sure if we are giving this undue weight? I am not personally against such speculations but there are many, such as those of Halsall, and I am not sure this article needs to summarize them all. Furthermore, coming back to the idea of being cautious about the old Bede narrative, I think this speculation follows Bede in the sense that it sees a continuity between the specific group of invited Saxons, and Bede's own Anglo-Saxons. Bede's idea is that they must have kept fighting and eventually beaten the Britons. I think that one implication of what many authors (such as Halsall, Fleming and others) are saying more recently is that we can't assume (like Bede, or apparently Higham) that this one invited group was the starting point, or the end point, for the speaking of Old English in Britain.--Andrew Lancaster (talk) 07:01, 22 June 2024 (UTC)