h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

A recent publication is a book by Stephen Oppenheimer (author of "Out of Eden") entitled "The Origins of the British - A genetic detective story".

In it he traces the various and ancient migrations that have contributed to the population of Britain based on their genetic legacy.

The conclusions are interesting and controversial. Oppenheimer suggests that the Anglo-Saxon invasion of the Dark Ages did occur but had a minimal (genetic) impact - and the same is true of the later Viking/Danish invasions that established the Danelaw.

Further, Oppenheimer suggests an alternative view about why the Celtic language has had so small an impact on the English language. According to this theory it is because in Pre-Roman Britain Celtic may not have been the universal language. He suggests that "Saxon" may well have been spoken in certain areas well before the Romans arrived.

Oppenheimer is from Oxford University and specialised in the use of DNA to track migrations.

In simple terms Oppenheimer suggests the following:

o The original Celtic homeland was never Austria, but rather southern France and Northern Spain

o The majority of Britain's genetic inheritance derives from Northern Spain, as humanity expended northwards from 'refuges' during the ice ages (which made Britain uninhabitable). Lower sea levels meant that Britain (and Ireland) could be reached via land. This was the primary influx in the period 13,000BC to 5,500BC. This influx was specifically focussed on the western side of Britain.

o There was also an influx of genes at a later date from north-western Europe, but still in the 'Pre-Roman' period. This influx specifically influenced parts of eastern Britain. These came across the (restricted) land-bridge that existed at the time. Sea levels had risen but the "North Sea" was still a land bridge. This influx had specific influence on the eastern side of Britain.

o That the western side of Britain had a different set of inputs than did its eastern counterparts. This suggests that genetic differences between western and eastern Britain goes back millennia.

o Oppenheimer suggests that in some way the "North Sea" migrants may have brought a 'non-celtic' language with them. This could have been a Germanic language, and may have been the real origins for "Old English" (as opposed to a language introduced by the Anglo-Saxons a couple of millennium later).

o That parts of southern England (specifically from Hampshire eastwards) were later settled by 'Belgaeic' settlers - but still at a time before the Romans arrived - and that these settlers spoke an early Germanic (rather than Celtic) language. A number of the tribes that existed in southern England (Wessex) at the time may have been settled by Germanic influenced/non-Celtic parts of Gaul (such as the Belgae, and the Atrebates). This may may support why Caesar suggested that parts of Britain had a similar language and culture to its (Belgaeic) neighbours and why (ultimaltely) the "Saxon Shore" was so named.

o That (by reference to language mutations there is evidence that) 'Old English' shows signs of having diverged from other Germanic languages long before the post-Roman"Anglo-Saxon invasion". This suggests that English did not arrive in Britain with the post-Roman Saxon invasion. It existed already and was simply influenced by its newly arrived cousins.

o He also suggests that there is little evidence for Celtic languages being used in eastern/central England - given the lack of Celtic stone inscriptions which ARE evident in the West and North (eg in Cornwall, Devon, Wales, and Cumbria)

o That there is genetic evidence of the traditional view of post Roman 'Anglo-Saxon' (or more specifically 'Angle') immigration into Eastern England (and Southern Scotland) and of the slightly later "Danish" migration (that gave rise to the Danelaw). However genetic evidence suggests that these were not massive population shifts. The "Anglo-Saxon" migration might represent less than 4% of people in England today (reaching a maximum of 15% in parts of East Anglia and Mercia), and the Danes just over 4%.

The book questions a number of 'assumed truths' and 'traditional positions' and is well worth a read. I have yet to be convinced on some of his arguments - specifically I found there was little real evidence of languages other than Celtic being spoken in Britain in pre-Roman times.

What did Oppenheimer say about Devon? Genetically he said very little because he seems to have used the Capelli et al samples used for the "Blood of the Vikings' survey, and those sample sites missed Devon completely.

On a broader front he acknowledges that "there is good evidence that Brythonic Celtic languages, probably South-West Brythonic, were also spoken elsewhere in the West Country, in Devon."

And Fulup will be pleased to note that he considers Cornwall to be separate from England (although he does seem to confuse Cornwall and Dumnonia somewhat.)

Nevertheless - an interesting read.

As Ozzie Exile has stated, Oppenheimer indicates an initial migration from Northern Iberia after the Ice Age, which can still be detected in the fact that throughout the whole of Britain the descendents of these people still constitute the majority. However, later (Neolithic, Bronze, and Iron Age) migrations reveal an east/west split in England, between mostly further migrations from Iberia in the west, and mostly migrations from North West Europe in the east. As Oppenheimer has primarily used the Cappeli et al (“A Y Chromosome Census of the British Isles”) data, which had no data from Devon, he could not be precise as to where the split line occurred. Other (non-genetic) evidence that he presents (archaeological and linguistic) suggests that the whole of what was Dumnonia lies to the west of this line.

The Capelli et al sampling of Y-Chromosome data has been criticised by some, due to the large geographical data gaps that were left throughout Britain. Fortunately, a more recently reported survey of both Y-Chromosomes and mitochondrial DNA (mtDNA) was carried out by Professor Bryan Sykes (also of Oxford University). Sykes’s survey sampled over 10,000 men and women from all over Britain, with no significant geographic gaps. He and his team are in the process of constructing a genetic atlas of Britain, and the early results from this project have been outlined in Sykes’s recent book “Blood of the Isles – Exploring the genetic roots of our tribal history”. In this book (which is aimed more at a less academic readership than the Oppenheimer book) Sykes divides Britain up into geographical regions (Highland, Grampian, Argyll, Tayside and Fife, Borders, Northumbria, Northern England, Central England, East Anglia, London, Southern England, South West England, North Wales, Mid Wales, and South Wales). Interestingly, Sykes’s Mid and South Wales regions extend eastward into the English border counties of Gloucestershire, Herefordshire and Shropshire, and the extent of South West England is defined as Cornwall, Devon and West Somerset, i.e. Dumnonia. Sykes has successfully filled in the Y-Chromosome ‘gaps’ left by Capelli.

Sykes has provided a very brief summary of the results obtained from the data in the appendix to his book, but he has also referenced an ‘Oxford Genetic Atlas Project’ website on which the whole of the Y-Chromosome database can be found. Sykes collected a total of 238 Y-Chromosome samples from all over the South West (including many from Devon), and the comparisons with other regions of England, Wales and Scotland (Appendix in Sykes’s book) are illuminating. South West England has by far the highest percentage of Iberian derived Y-Chromosomes in England, with 78.2%. This figure compares well with North Wales (78.5%), Borders (78.1%), Tayside and Fife (78.9%) and Highland (75.9%), and contrasts markedly with the 57.7% figure for neighbouring Southern England. Only Mid Wales (86.4%), South Wales (84.2%), Argyll (81.1%) and Grampian (83.5%) had perceptably higher percentages of men of paternal Iberian descent. The region of England nearest to the South West in terms of Iberian descent was Northumbria at 68.3%, which notably includes the former Celtic kingdom of Cumbria, whereas the region with the lowest percentage was (perhaps not surprisingly) East Anglia with 51.2%. Also the 78.2% overall figure for the South West compares well with the 79% figure obtained by Capelli et al for Penzance, so it would appear that this high percentage was fairly consistent throughout the region. The source for the above data is acknowledged as “Sykes, B.C., Blood of the Isles, Bantam Press, London (2006)”.

The much more extensive Sykes data set reinforces the conclusions arrived at by Oppenheimer (using the Capelli data set), and strongly suggests that Oppenheimer’s divide line should be located at the eastern border of Dumnonia, and to the east of the English counties adjacent to South and Mid Wales. Further north than this, its location becomes more problematical. Should Northumbria be located to the west or east of the line, and were the Cumbian results significantly more ‘Iberian’ than the rest of the results from Sykes’s Northumbrian region?

As for linguistic differences, in what is now England, between the West and the East in pre-Roman times, this if far less certain than the observed genetic differences. West of Oppenheimer’s genetic line, Brythonic languages definitely prevailed by the time that the Romans arrived, but the evidence is far less clear to the east of the line. Only a third of the known pre-Roman place-names in the East can be definitely identified as being Brythonic in origin, and such names are very sparse in the South East, where migrants from the Belgic areas of Gaul and adjacent regions are reputed to have settled. Also, as the more recent pre-Roman migrations into the East had been from Germanic speaking areas, it is perhaps quite possible that Germanic derived languages had gained a foothold in the South East. Another possibility is that a British form of Latin had become the common language of the heavily Romanised region of Britain prior to the Anglo-Saxon invasions (as discussed here previously).

I found this summary of Oppenheimer’s book from Prospect magazine.

http://www.prospect-magazine.co.uk/article_details.php?id=7817

It certainly is a tempting theory, and it is refreshing to see some new ideas about this subject emerging.

I have been interested in the idea that the English language is older than previously thought for some time now. But I think it’s more likely Germanic speech was brought to Britain in Roman times, by the Germanic mercenaries that we have discussed before.

Archaeological evidence shows these people had settled on the east coast of England and along the Thames all the way into Oxfordshire by the end of the 4th century.

I agree with Oppenheimer that there is not much evidence for Celtic culture or speech in either Roman or post-Roman central and southern England. But there is no evidence for pre-Roman Germanic speech either. The place name evidence across England suggests that the Romans found a people speaking a Celtic language living a rudimentary, non-urban agrarian existence. The few settlement or regional names that existed were taken into the Latin language. Eboracum, modern day York, is a Celtic word adapted into Latin. The Celtic name Kent seems to have survived the same way. Also a large number of topographical names are indisputably of Celtic origin, such as the Darenth Valley in Kent. No truly ancient---as in pre-Roman---Germanic names are apparent, although there are some, like Thames and Severn, which some believe pre-date the Celts.

As Plymouth Exile says, I have thought for some time that the British language gave way to Latin in these areas.

The changeover to a Germanic language is better accounted for by plague, depopulation, and abandonment of cities in the 5th and 6th centuries---all of which are described by contemporary sources---combined with the low social status of the remaining Brythonic speakers who, for the most part, had probably been used as a convenient supply of slave labour by the Roman government for the previous 300 years.

This scattered and probably somewhat demoralised Briton population may have adopted Germanic out of economic necessity. As I have suggested before, it is possible the new rulers offered a greater degree of social mobility to the Britons than the Romans, and this formed a strong motivation to learn the German language. And given the long-term presence of the Germanic mercenaries in Roman Britain, it is quite possible some Britons already knew these languages.

Outside of the genetic field, I find some of Oppenheimer’s assumptions in the Prospect article a bit tenuous. Typically he criticises Gildas, as seems to be the fashion lately, describing him along with Bede as “clerical historians”---as if there was any other kind of historian in Britain at the time---when the accounts of these early British writers are actually better than anything we have surviving from the Roman period itself. I agree there is probably exaggeration in these texts, but that is no reason to dismiss what they so clearly say---that there was pestilence, famine, civil war, anarchy and conflict in this troubled period. Gildas’s observations also correlate with other writers of the time. I cannot believe they were all making this stuff up.

“Tacitus reported that between Britain and Gaul ‘the language differs but little’

”The common language referred to by Tacitus was probably not Celtic, but was similar to that spoken by the Belgae, who may have been a Germanic people, as implied by Caesar. In other words, a Germanic-type language could already have been indigenous to England at the time of the Roman invasion.”

There is no real evidence on offer here in this statement, and Oppenheimer seems far more comfortable with early Roman accounts that may be just as confused and inaccurate as the later texts he dismisses. He cannot even say for sure whether the Belgae were Germanic.

In contrast Gildas references surviving Celtic place names in Britain, and he also clearly describes Latin as ‘lingua nostra’---meaning ‘our language’.

I think if Oppenheimer wants his theory to gain ground, it needs more supporting archaeological and cultural evidence to go with his genetic conclusions.

ExeValleyBoy,

I think one has to be careful not to overplay the depopulation card when trying to explain the changeover to a Germanic language in Dark Age England, as a massive depopulation among the native Britons is not supported by the genetic data. Both Oppenheimer and Sykes estimate an Anglo-Saxon incursion representing about 4-5% of the population at that time, rising to about 15-20% in East Anglia, so the incoming population would have still been small in comparison with the existing inhabitants.

Oppenheimer’s use of quotes from both Caesar and Tacitus in support of his thesis of pre-Roman Germanic speech in southeastern parts of Britain was far from being the only evidence he cited, as he devoted two whole chapters to the topic. Having said that, I felt that this was probably his weakest argument in the book. Surprisingly, I felt he underplayed what was potentially the strongest of his cards in this respect, i.e. that there was significant migration into Eastern England from North West Germany during the late Neolithic and Bronze Age periods, when one might have expected Germanic to have become the predominant language in this continental region. Of course that does not mean that the migrants would have continued to use their Germanic tongue, as they might well have adopted the native (Celtic?) speech of the native British majority in Eastern Britain.

Oppenheimers contention that there was probably a pre-Roman split between the east and the west of Britain does not rely on there being a language difference, as most of his evidence for this is genetic and archaeological. He roughly locates the split line as running from Wessex to the Grampian region. Certainly Sykes’s data indicates a significant genetic step between the South West of England (effectively Dumnonia) and Wessex in the Y-Chromosome record.

Plymouth Exile,

I have only read the article in Prospect so far, so was only commenting on how Stephen Oppenheimer presents or argues his theory in that.

I get the impression from the article that he is trying to sneak yet another ‘Teutonic’ race paradigm in through the back door. The old Anglo-Saxon genocide explanation having faltered and the elite-dominance or sexual apartheid approach being too flimsy, Oppenheimer is now going to back to prehistory to say, essentially, that there never really were any Celts, and the ‘English’ were Germanic, or part-Germanic, right from the start.

Without the ancient German language theory, which pops up in the article more than once, I cannot see what is that new or particularly distinctive in Oppenheimer’s argument.

I find the reasoning a bit odd on occasions;

“But the presence in Roman England of some Celtic personal and place-names doesn't mean that all ancient Britons were Celts or Celtic-speaking.”

But aside from a few river and landscape names of great antiquity and which don’t appear to be Germanic anyway, what evidence is there for non-Celtic languages in Britain either directly before or during the Roman period? There is vastly more evidence for widespread use of Latin than any kind of early Germanic speech, but Oppenheimer---in this article at least, I cannot say for the book---does not mention Latin.

Oppenheimer then goes on to say;

“The orthodox view is that the entire population of the British Isles, including England, was Celtic-speaking when Caesar invaded.”

And follows with;

“But if that were the case, a modest Anglo-Saxon invasion is unlikely to have swept away all traces of Celtic language from the pre-existing population of England. Yet there are only half a dozen Celtic words in English, the rest being mainly Germanic, Norman or medieval Latin.”

What does Oppenheimer think had been going on in Britain between Caesar’s invasion and the arrival of the Anglo-Saxons? The Romans changed the country out of recognition and introduced a new language that became vital for ordinary trade, commerce and everyday life in the new cities and towns they founded.

“One explanation is that England was not mainly Celtic-speaking before the Anglo-Saxons.”

I agree, that is a possible explanation. But Oppenheimer does not seem to consider the possibility that it was the Romans who changed the language, and that as in other parts of the Roman Empire, the indigenous Celtic (or otherwise) tongue had become mainly the preserve of country dwellers living far removed from Latin-speaking urban areas.

In your reply, you said;

“I think one has to be careful not to overplay the depopulation card when trying to explain the changeover to a Germanic language in Dark Age England, as a massive depopulation among the native Britons is not supported by the genetic data.”

The kind of depopulation we are talking about is important here. I am not suggesting everyone died or fled the country, only that a certain kind of population vanished during this period.

This was the urbanized, Latin-speaking population of the Roman towns and cities. What happened in Canterbury is a good example.

Archaeologists excavating ancient Canterbury have found a thick layer of black composted matter on top of the Roman city, up to 20 centimetres thick in places. This layer is made up the decayed remains of collapsed buildings. The only conclusion is that this large settlement was abruptly abandoned and became completely deserted. Other evidence shows that when the Saxons came they did not go near the place, and chose to settle well outside the ruined city. It appears that Canterbury was only reoccupied at the beginning of the 7th century.

The population living in these cities formed only a small part of the British total, but it was a very important segment of society. When these people disappeared---for whatever reason---so did the spoken Latin language and much of the country’s Roman-era cultural heritage.

There would have been a good number of Britons left, but they were scattered in small settlements widely dispersed across the country. The cities and the villas were lost, and with them the economic advantages and pleasures of civilized life. The Germanic settlers in Britain had no incentive to learn Latin and settle in down in a post-Roman culture, as happened in Gaul.

The Britons had lost their cultural centres, and with them, the ability to affect the language and culture of their new rulers. Nevertheless, Latin was the greatest influence on Old English, and I think a lot more it survived in the common speech than is thought.

We assume Old English---as in the Anglo-Saxon chronicles---was the common language during Saxon times, but as we have touched on here before, it may have been more a literary or court language. The Normans and the medieval church are held to be responsible for the massive amount of Latin vocabulary in English, but I think it is possible there were elements of Latin surviving in common, unwritten speech that made its 11th century reintroduction to England much easier.

ExeValleyBoy,

Having read Oppenheimer’s book, I certainly didn’t get the impression that he was trying to contend that there were no Celts in Britain; quite the opposite in fact. However, he certainly doesn’t believe that there was a large Iron Age migration from a supposed Celtic heartland in central Europe. In fact he doubts whether such a Celtic heartland ever even existed. His hypothesis is that the Celtic language came to Britain via one of the later smaller scale migrations from Northern Iberia into Western Britain. In fact most historians and linguists now believe that ‘Celtic’ is essentially a linguistic label, rather than a racial label as the Victorians believed.

What is new (to a certain extent) about Oppenheimer’s study is that whereas in previous population genetics studies, the assumption had been made that where evidence of Germanic influx was to be found in the genetic record, that the whole of the influx must have been due to the 5th century Anglo-Saxon invasion and the later Danish invasions/migrations. The possibility of earlier influxes from the region of North West Germany/Denmark had not even been considered. Oppenheimer used the newer technique of Phylogeography to determine which haplotypes arrived approximately when and from where, and discovered that most of the influx from this region arrived in Eastern Britain well before the Dark Ages. He therefore concluded that the Anglo-Saxon and Danish influxes of the 5th century and later must have been real, but minor events in terms of overall numbers.

According to a number of present day archaeologists, the abandonment of the cities following Roman withdrawal was not as complete as was once supposed. Philip Barker found evidence for post–Roman large-scale timber buildings at Wroxeter. The buildings themselves had rotted, but the postholes remained. In York, Mark Whyman has shown that a crude form of hand-made pottery was still being made in the city well into the 5th century, so the city must have been occupied to some extent at that time. Modern archaeological techniques are uncovering new evidence for continued, but reduced, habitation of the Roman cities well into the Dark Ages. The main reason that historians had assumed that occupation had suddenly ceased at the end of the Roman era was that the Romans had largely used stone construction techniques, which provided good archaeological remains, but the later wooden constructions had rotted into dark layers and been built over (and therefore obscured) by constructions of the 7th century and later. In fact many historians and Archaeologists are now convinced that the decline of the cities started well before the Romans departed, probably at the beginning of the 4th century, and was a more gradual process.

I would tend to agree with you that a significant number of Latin words passed into Old English from the Britons, centuries before the Norman influence.

As stated in my earlier posting, I also have some reservations about Oppenheimer's suggestion that some form (or forms) of Celtic language was not spoken throughout Britian in pre-Roman times.

The problem is Oppenheimer's lack of evidence on that supposed alternate language.

Having said that, this is the evidence he presents:

1) An almost complete lack of Celtic inscriptions outside of the western side of Britain. Incidentally he he cites a number of inscriptions (about ten) for Devon and slightly more for Cornwall, but none (that I can see) in Somerset or Dorset. There are only 1 or 2 sites in England outside of Cornwall, Devon, Cumbria and the Welsh borders.

2) Indications that Belgaeic tribes had occupied south/south-eastern England in pre-Roman times - and had similar spoken language and culture to their European contemporary equivalents.

2) Old English (even in its earliest forms, such as evidenced in the poem 'Boewulf' or in Alfred's Will) is considerably different from the equivalent Anglo-Saxon of the day. Based on a comparison with other Germanic languages it is well removed from German (either high or low), Frisian, or Dutch. In fact it seems roughly equidistant between those languages and the Scandinavian languages of Sweden and Denmark. Based on the differences in vocabulary it sits somewhere between them, and yet out on a branch of its own.

Based (I assume) on the rate of change in languages, the changes are far more diverse than could have occured in the brief period since the "Anglo-Saxon" invasion. Oppenheimer quotes Forster as saying that the break up of the (Germanic languages) can be no younger than 350AD, and possibly as much as 4,000 years old.

This latter point suggests a number of things (even if proven true), and not all of which require a Germanic language in pre-Roman Britain.

One option might be that England was invaded by a group who were somewhat insular in language and hence had 'diverged' from traditional sources well before they arrived in England.

A second is that the language was subject to a strong external influence in its early years - from Celtic or Latin?

A third option may be that the language arrived with the Romans. The later suggested date of 350AD is consistent with EVB's suggestion of early Roman 'mercenaries' bringing the language with them - although the earlier date of 4,000 years ago (2,000BC) is not.

As I stated earlier, Oppenheimer's suggestions are interesting, but not all are wholly convincing.

>>Oppenheimer used the newer technique of Phylogeography to determine which haplotypes arrived approximately when and from where, and discovered that most of the influx from this region arrived in Eastern Britain well before the Dark Ages.<<

This is very interesting. For quite some time I have, as stated here in the past, been of the opinion that a majority of "Germanic" genes in England are probably derived from Pre-Celtic migrations rather than Post-Roman ones. This was based partly on the fact that for numerous reasons, I didn't think the Anglo-Saxon invasions were that profound; and it was also based on anthropological evidence.

I think a likely period for this would be the Mesolithic, when the Maglemosian culture spread from Denmark into England and parts of Northern France. The Bronze Age also saw an immigration of people from the Rhineland and Belgium, whose culture was a blend of the Corded/Battle-Axe (who are considered the most likely candidates for bringing Germanic languages to Europe) people with that of the "Bell Beaker" folk, originally from Spain, and who in my opinion were probably Proto-Celts.

As for whether their were "Germanic Britons", Caesar also said that the Belgae and Celtae were similar. I've read elsewhere that Romans considered Britons, Gauls and the Irish to be closely related. So the relations of the Britons were clearly Celts rather than Germans, although Caesar did say that the Belgae claimed to be partly Germanic (which might have dated from the Bronze Age mixture of Corded and Beaker cultures).

Just to add some further comments, it seems to me that Oppenheimer should be viewed as a scientist, not a historian. I find his speculations about the linguistic/cultural affiliations of Britons, Gauls and Germanic people way off the mark; it's almost as if he's joking.

Either he's pushing some agenda (or just making a good story), as ExeValleyBoy suggests, or simply that history is very much a foreign field to him.

Certainly the objections he raises to ideas such as Britain being Celtic-speaking in Roman times are easily explained without resorting to such strange theories.

But his suggestion, based on genetic evidence, that the proportion of Anglo-Saxons to the earlier population is only five percent is very much in the vicinity that I would have expected, judging from historical and other evidence.

Einion,

While I would agree that Oppenheimer’s hypothesis that a Germanic language in Britain could have preceded the Anglo-Saxons by hundreds or maybe thousands of years is probably the weak point in his book, we are still face with the fact that pre-Anglo-Saxon migrations of people from what is now North West Germany did settle in Eastern Britain at a time when Germanic languages were established in that part of continental Europe. It is therefore highly probable that they came speaking a Germanic tongue. Whether they continued to do so, or not, is open to conjecture. The fact remains that we do not know for sure what the common speech among the Britons of the heavily Romanised lowland part of Britain was.

>>we are still face with the fact that pre-Anglo-Saxon migrations of people from what is now North West Germany did settle in Eastern Britain at a time when Germanic languages were established in that part of continental Europe.<<

Is this the Bronze Age "Beaker Folk" from the Rhinelands you're referring to?

>>The fact remains that we do not know for sure what the common speech among the Britons of the heavily Romanised lowland part of Britain was.<<

It's difficult to be certain, however I feel it goes against the evidence to suggest a Germanic language was spoken in parts of Britain in Roman times.

Tacitus thought that the Britons were probably of Gaulish descent, based on similarities in language, customs and appearance. He is apparently referring to the Gauls as a whole, not just the Belgae, which seems to be Oppenheimer's suggestion.

Einion,

The Bronze Age migration that I was referring to was that identified by Oppenheimer as originating from Lower Saxony and Schleswig-Holstein (i.e. from the exact same region that the much later Anglo-Saxon invasion/migration emanated from). It consisted of a mixture of I1a-7a and R1a1-3c Y-chromosome haplotypes, and accounts for 7-8% of the current male populations in the East Anglia and East Yorkshire regions (Oppenheimer, Fig.5.14b). This migration is dated at about 4,000 to 3,700 years ago, at a time when early Germanic languages would almost certainly have been spoken in these continental regions.

Concerning the question of whether a Germanic language was in use in parts of Roman Britain, one should perhaps bear in mind that a number of accounts refer to the German ‘foederati’, who would almost certainly have been Germanic speaking. This is not to say that the whole of the heavily Romanised part of Britain was Germanic speaking when the Romans departed, but there could possibly have been a small but significant element who were. When the Anglo-Saxons arrived in what is now Eastern England, any existing Germanic speech in that area would have strengthened their foothold and given what was to become known as Old English a significant boost.

>>The Bronze Age migration that I was referring to was that identified by Oppenheimer as originating from Lower Saxony and Schleswig-Holstein (i.e. from the exact same region that the much later Anglo-Saxon invasion/migration emanated from).<<

So has he concluded from genetic evidence alone that this invasion occured (and at the specified time), or is he combining the genetic and archaeological evidence to come up with the conclusion?

It would certainly fit in well with archaeology except that he thinks the invasion came from a little further north-east than that of the Rhineland "Beaker Folk". But it's perfectly conceivable (and I think I may have read something to this effect) that the Corded Ware people who formed an element in this mix had originally come from further north-east.

Einion,

“This is very interesting. For quite some time I have, as stated here in the past, been of the opinion that a majority of "Germanic" genes in England are probably derived from Pre-Celtic migrations rather than Post-Roman ones. This was based partly on the fact that for numerous reasons, I didn't think the Anglo-Saxon invasions were that profound; and it was also based on anthropological evidence.”

However, a pre-Roman Germanic presence does not explain what happened to the Latin and Celtic languages in England after the Romans left. Neither does it explain the lack of Germanic language evidence in pre-Roman era topographical names, or the lack of recognizably Germanic cultural elements in pre-Christian Britain; such as Romanized shrines incorporating known Germanic deities. Early on in Roman times, British culture generally appears to be Celtic, as at Bath with the cult of Sulis.

There is good evidence to suggest that there was a period of significant mid to late Roman era Germanic migration into Britain, in which large numbers of Saxon mercenaries or foederati—as Plymouth Exile points out—settled in Britain over a period of a couple of hundred years, particularly in the south east.

There were large numbers of Germanic people settled in Britain long before the end of Roman rule. I would think it more likely that Germanic language became first entrenched in Britain during this late Roman period through the foederati recruitment system rather than having invisibly weathered 400 years of Roman rule only to re-emerge as native speech as soon as the legions departed.

As Ozzie Exile mentions, Forster—quoted by Oppenheimer—says the cut off point for Germanic language divergence is about the year 350, which would tie in with the Roman era foederati settlements.

In this situation, we see Germanic foederati settling in Britain from the 4th century onwards and possibly earlier—there is some evidence for foederati coming to Britain as early as the 1st and 2nd centuries—and bringing with them an earlier form of Germanic language that then diverged from other Germanic tongues because of Britain’s geographic isolation and from exposure to the Latin speech of the southern Romano-Britons. Celtic speech may have influenced early English too, but there is less evidence for it.

ExeValleyBoy,

I actually agree that the evidence points clearly toward Brythonic being dominant in Britain in Roman and earlier times. I think that much of the genetic similarity between England and Northern Germany is likely to date from both Pre-Celtic and Pre-Germanic times, so the people who entered were not Germanic speaking, although the Bronze Age may have brought a small number who were; at any rate the survival of this language throughout the Celtic-dominated millenia (indeed the Iron Age probably saw a further invasion of Celts from Gaul) is highly unlikely.

As you mention, there must have been foederati in Britain perhaps as early as the third century, however I'm inclined to doubt that it would have much do with the relative lack of Celtic influence in early English, or on the fact that it, rather than Latin or Celtic, became the language of England.

I think there is another likely explanation for both, which I've written somewhere on this forum (I can't remember which thread).

In short, I think the reasons Old English being dominant may be:

1. Latin was widely spoken as a second language in Britain during Roman times, but failed to displace Celtic.

2. Many of the invaders (including the leading men) had a knowledge of Latin (which could have been reinforced by the fact that there were foederati already in Britain, who must have been familiar with Latin), due to the centuries of contact they'd had, particularly around the English channel.

3. When the invaders took over the British kingdoms, Latin and Old English (and not Brythonic) were used as the languages of government, the army (with Latin of secondary importance, simply as a medium) and trade, because the latter was the native language of the conquerors who had formed a new administration and army, and the former was a lingua franca known to both groups.

4. The Britons, becoming allies of their conquerors and joining their administration and army, gradually became familiar with English, until the medium of Latin became unnecessary, since it was not the native language of either group. So it eventually dropped out of use, leaving only Old English in a position of dominance, and there was no reason for Brythonic to re-emerge into dominance, since the Anglo-Saxons couldn't speak Brythonic, and any Britons in frequent contact with them could now speak English.

My meaning would probably be clearer if I compare this situation with what happened in Gaul, which I'll do later.

So the situation in Gaul was (I think):

1. Latin had established itself as the native language of at least the upper sections of society.

2. The invaders, as in Britain, had a knowledge of Latin.

3. As in Britain, the Germanic language was established as the primary one for the government and army, but with Latin also used, for communication between the two groups.

4. The Gaulish people (those in frequent contact with the invaders) became familiar with the Germanic language. However, the difference would be that since Latin was the native language of these Gauls, and one the invaders could speak, it didn't simply drop out of use as an irrelevant extra language, but continued to be spoken in the army and government.

And because the native Latin-speakers outnumbered the Germanic, usage of Latin increased at the expense of the latter, until the latter was no longer used in government etc., but became a local language of villages and homes, and eventually (perhaps after centuries) died out completely.