THE CELTIC MYTH.

A correspondent sent me a Prospect Magazine story by Stephen Oppenheimer about the ethnic origins of modern Britons. Now, I’m very suspicious of attempts to apply biology to linguistics, and I’ve bashed Oppenheimer before, but what the hell, I’ll toss an excerpt up and see if anyone has anything useful to add:

So who were the Britons inhabiting England at the time of the Roman invasion? The history of pre-Roman coins in southern Britain reveals an influence from Belgic Gaul. The tribes of England south of the Thames and along the south coast during Caesar’s time all had Belgic names or affiliations. Caesar tells us that these large intrusive settlements had replaced an earlier British population, which had retreated to the hinterland of southeast England. The latter may have been the large Celtic tribe, the Catuvellauni, situated in the home counties north of the Thames. 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. In support of this inference, there is some recent lexical (vocabulary) evidence analysed by Cambridge geneticist Peter Forster and continental colleagues. They found that the date of the split between old English and continental Germanic languages goes much further back than the dark ages, and that English may have been a separate, fourth branch of the Germanic language before the Roman invasion.
Apart from the Belgian connection in the south, my analysis of the genetic evidence also shows that there were major Scandinavian incursions into northern and eastern Britain, from Shetland to Anglia, during the Neolithic period and before the Romans. These are consistent with the intense cultural interchanges across the North sea during the Neolithic and bronze age. Early Anglian dialects, such as found in the old English saga Beowulf, owe much of their vocabulary to Scandinavian languages. This is consistent with the fact that Beowulf was set in Denmark and Sweden and that the cultural affiliations of the early Anglian kingdoms, such as found in the Sutton Hoo boat burial, derive from Scandinavia.
A picture thus emerges of the dark-ages invasions of England and northeastern Britain as less like replacements than minority elite additions, akin to earlier and larger Neolithic intrusions from the same places. There were battles for dominance between chieftains, all of Germanic origin, each invader sharing much culturally with their newly conquered indigenous subjects.

Oh, and he thinks Brits are really Basques: “But the English still derive most of their current gene pool from the same early Basque source as the Irish, Welsh and Scots.”
Thanks for the link, Paul!

Comments

  1. Interesting idea. Dubious. I think there is a slightly deeper split between English and Continental WG: as far back as you go in written records, Old English is still itself and doesn’t become just another dialect of the Old Frisian/Franconian/Saxon/High German blend, which do all start to look alike after a while. It looks most like Saxon, I thought the other day, looking at Old Saxon. But five hundred years (minimum) apart? Stretching plausibility.

  2. It’s an attractive hypothesis, but too bad there’s no actual evidence of Germanic language in pre-Roman Britain. Forster’s yet unpublished lexical analysis may say that English branched earlier than usually thought, but this doesn’t show that this branch actually came to England early.
    http://en.wikipedia.org/wiki/Belgae says:

    Whether the Belgae were Celts or Germanic tribes occupied 19th century and early 20th century historians. Caesar’s sources informed him “that the greater part of the Belgae were sprung from the Germans, and that, having crossed the Rhine at an early period, they had settled there, on account of the fertility of the country,” (De Bello Gallico 2.4) However most of the tribal and personal names recorded are identifiably Celtic. It seems likely that the Belgae had a mixture of Celtic and Germanic ancestry. Perhaps they were Germanic people ruled by a Celtic élite, or were a political alliance of Celtic and Germanic tribes, or, like the later Normans, were a formerly Germanic-speaking people who had adopted the language of the lands into which they migrated. In any case, the Romans were not precise in their ethnography of northern barbarians: by “German” Caesar may simply have meant “originating east of the Rhine”, with no distinction of language intended.

    And the paragraph from Tacitus mentions “German” separately from “Gaul”:

    Who were the original inhabitants of Britain, whether they were indigenous or foreign, is as usual among barbarians, little known. Their physical characteristics are various, and from these conclusions may be drawn. The red hair and large limbs of the inhabitants of Caledonia point clearly to a German origin. The dark complexion of the Silures, their usually curly hair, and the fact that Spain is the opposite shore to them, are an evidence that Iberians of a former date crossed over and occupied these parts. Those who are nearest to the Gauls are also like them, either from the permanent influence of original descent, or, because in countries which run out so far to meet each other, climate has produced similar physical qualities. But a general survey inclines me to believe that the Gauls established themselves in an island so near to them.

    (The paragraph also shows the debate about British origins hasn’t changed much in 2000 years…)

  3. Caffeind,
    By “actual evidence” do you mean reference in texts of the time? That evidence may indeed be absent, but that is not evidence of absence. What interest would the Romans have hasd in the question, beyond the military intelligence value of the information? The same kind of situation obtains in China, where actual references are very sparse in the historical record, but other evidence happens to survive – a “Hunnish” song transcibed into characters that appears to resemble Ket, Hmongic substratum in the dialects in what used to be Chu and so on – but here the question is itself what is the substrate and what is the main language. It can’t be the easiest job to determine with absolute clarity even what is Celtic and what is Germanic anywhere, to begin with, considering their geographical proximity over many centuries.
    Here’s another question – what would have prevented Germans from settling in Anglia and all that low swampy country in the centuries before the Romans showed up? Is that narrow, little body of water really such a barrier? It didn’t suddenly get narrow and easy to navigate in the fifth century.

  4. There’s plenty of physical evidence of intercourse across the North Sea for thousands of years back, but no information about language.
    Don’t forget the parts of the Continent closest to Britain were neither Celtic nor Germanic, but Nordwestblock, until just before the Roman period.

  5. I think that the most interesting thing in the article is the claim that language change didn’t involve a lot of population change (genetically). We’ve seen that in Ireland in historical times: the same people gradually ended up speaking mostly English and not much Erse.

  6. Yes, and it happened in just a couple of generations when the economic pull of English was at its height. The same may have been true of Romanization, or Arabization.
    If elite dominance can change language so effectively, the corollary may be that genetic history tells us almost nothing about language history. But then, there are also lots of cases of language replacement via population replacement.

  7. The most appealing explanation for the receding late Roman Britons is that they did exactly what today’s Britons are increasingly doing in an integrated Europe: move south to better climates.

  8. I was taught that the britian/iberia link went the other way around, that the celrs invaded Spain some time before the Visigoths showed up.

  9. “The corollary may be that genetic history tells us almost nothing about language history.”
    I think that people are less hopeful that it will. It’s still possible to trace physical migration if you have distinct genetic markers, but a second problem is that Celts, Germans, and Slavs aren’t that differtent.

  10. Richard Hershberger says:

    Am I safe is assuming that this lexical evidence analyzed by a geneticist is just glottochronology rearing its head again? Did I miss some reason why anyone should take it seriously?

  11. Just thought I’d give another example. The Manichaean liturgy in Central Asia went from Persian to Sogdian to Turkish over the course of centuries (about 5 centuries, I think). The society was stable and continuous, with no major population displacement. Persian was probably a purely-liturgical language to begin with, replaced by the spoken language Sogdian, and then replaced with Turkish as the spoken language changed again.
    C. Asia has been bilingual Persian/Turkish for perhaps a millenium. The movement has been toward Turkish, but plenty of Tajiks (Persian dialect speakers) remain, and there’s pervasive bilingualism.
    Source: Gnosis on the Silk Road, Klimkeit (sp?).

  12. Sure, you could also say Europe has been bilingual Romance/Germanic for two millenia, and that Christian liturgy etc. changed languages more than once. Then the next question is why those linguistic areas have been relatively stable for a long time now, and why much more change occurred in the Migrations Period.

  13. Another take on this whole controversy is here:
    http://www.gnxp.com/MT2/archives/000648.html by a historian taking a look at the original genetic survey information and bringing in evidence from written sources, language, place names, etc. Interestingly, he mentions that the genetic study sampled folks around Britain, and compared it with samples from “Norway, Denmark, North Germany (Schleswig-Holstein), Friesland (Netherlands), and the Basque region of Spain”. This seems pretty odd since you have to leap over a lot of territory to get from that stretch of northern Europe to the Basque region. The study would seem to ignore France and the rest of Spain as possible influences, skewing results toward Oppenheimer’s theory.

  14. I think that the undelying idea is that some, but not all, NW Europeans are the genetic descendants of the pre-Indo-European peoples, even though all but the Basques speak IE languages by now. The contrastive populations would be Mediterraneans, who did replace the original inhabitants in their area.
    These would presumably be Gimbutas’ “Old Europeans”, and the people who built Stonehenge.
    You hear me right. Stonehenge was bult neither by hippies nor by Celts. Ain’t history fascinating?

  15. Wow, he can tell all that from the DNA? Can he tell what language my grandmother spoke by looking at my DNA? :-) :-) :-) Of course not. If the Brits, Irish, Welsh, and Scots are all Basques from way back then for all I know she could have been speaking Welsh or Basque to me instead of Irish.
    Just a nit on caffeind’s response: “Yes, and it happened in just a couple of generations …”
    to The New Yorker’s point: “We’ve seen that in Ireland in historical times: the same people gradually ended up speaking mostly English and not much Erse.” The modern transition to English in Ireland proably started with the Statutes of Kilkenny in 1367 forbidding the Brits to speak Irish rather than with the Penal Laws of 1695, which would make it more than a couple of generations.

  16. The Brits? In 1367?
    Anyway, the decline of erse must be partly the responsibilty of the Catholic Church with its insistence on Latin services. The Church of England in Wales used Welsh and Welsh survived pretty well as a consequence.

  17. No, it was the Methodists who used Welsh and saved it! And it was reading the Bible that was significant, not the ritual part of services.
    Yes, Irish’s transition from a robust language to a dying one took only a few decades. Existence of an English-speaking elite does not by itself imply language death. By that criterion, English has already conquered half the world.

  18. Indeed. What the Statutes of Kilkenny say is that the English in Ireland had fears for the survival of English in Ireland; they had no effect whatever on the use of Irish. They said nothing about Irish people learning Irish.
    As for the question about the correlation of DNA and language, the simple answer is that people tend to raise thier own offspring.
    “Then the next question is why those linguistic areas have been relatively stable for a long time now, and why much more change occurred in the Migrations Period.”
    Well before that question, there is still the question of how much linguistic change really occurred. Mario Alinei says their is no archeological evidence for a change of population in Bavaria, for instance, and no evidence whatever for language change, since no one can say what language was spoken there. All anyone can say is that the Romans called the tribes Celtic. They could have called them that for any niumber of reasons, but maybe simply because the rulers spoke a Celtic language. The same may have been the case in the low-lying areas of Britain just opposite the low-lying areas of Northern Germany. we don’t know because no one was out there doing descriptive work.
    And on the other hand, there is no evidence for population change in Southern Mexico, but Spanish has gained a lot of ground in only a few decades. Soon to be followed by English, probably, when people start coming home with all their new money.

  19. “No, it was the Methodists who used Welsh and saved it!” Nope, you’re wrong. See “A History of Wales” by John Davies (Penguin, 1993), especially pp 242-3.

  20. Very interesting — Amazon.com‘s “Search inside the book” feature allowed me to investigate dearieme’s reference:
    “The myth concerning the Protestantism of the Celtic Church was given prominence in Epistol at y Cembru, the introduction to the Welsh translation of the New Testament published in 1567… Thereafter, Welsh would be the language of the services in those parishes where the language was in general use…
    “The Welsh bible which resulted from the labours of Salesbury, Morgan, Davies and the others was as central to the experience of the Welsh as was Luther’s Bible to that of the Germans or the Authorized Version to that of the English. Indeed, it could be argued that it was more central, for as German and English were languages of state they had secular means to maintain their unity, purity and dignity…
    “It is sometimes suggested that the Welsh Bible ‘saved the language’. The claim has little substance in view of the fact that most of the non-state languages of Europe were fairly secure as spoken languages for centuries after 1588. What it did ensure was the continuance of the Welsh language as something more than a spoken language.”

  21. And on p245 “Welsh was the only non-state language of Protestant Europe to become the medium of a published Bible within a century of the Reformation. This consideration is one of the most important keys to an understanding of the difference between the fate of Welsh and the fate of other non-state languages – Irish and
    Gaelic in particular…..”

  22. In 1567 Welsh Protestants wouldn’t be Methodists.

  23. I have Davies. p. 343:
    the Methodists were obliged to promote the Welsh language as a medium of evangelical zeal. They used the language in a less self-conscious way than did the scholars and the patriots, and, if the culture they created was narrow and pietistic, its medium was Welsh and its productions were prolific.
    p. 360:
    the upper ranks of the Anglican minstry in Wales had for generations been totally anglicized. Although there were a number of Welsh enthusiasts among the clergy, man parsons considered that their primary duty was to be chaplain to the local landowning families, families which had long been wholly English in speech. In the chapel, there were at least three Welsh services on Sunday and through the chapel the monoglot Welsh could participate in a host of other activities held in the only language they understood.

  24. The revivial of Irish – or its political importance at home – is such that it will become an official language of the EU next Jan,1, with romansian and Bulgarian.
    Bbut there is a serious problem of lack of qualified interpreters and translators. To such an extent that there is a suggestion of setting up a special school for Irish interpreters/translators into various of the many EU offical languages. It is said to take five years to qualify.
    The Eurosceptic British press is loving this one. One report said that life is so nice in the areas of western Ireland where there are the most Irish speakers that even qualified people don’t want to move to crowded Brussels. Having been there, I understand…
    See a good wikipedia piece on the whole EU language question, including the role of non-state lnaguages such as Galician:
    http://en.wikipedia.org/wiki/Languages_of_the_European_Union

  25. “In 1567 Welsh Protestants wouldn’t be Methodists”
    No, but having the liturgy and Scripture in “a language understanded of the people” was an Article of Faith in the first Anglican Book of Common Prayer, whatever that language might be, by that wording. Wales went with Protestant with most of England. The shift to Methodism came later, naturally.

  26. Quite, Jim. It’s not an either/or situation here; it’s a historical process. Without the Welsh prayer book and Bible and the approved use of Welsh in churches, it’s unlikely that the Welsh would have become Protestant enough to welcome Methodism (and all the other sects; you can walk round any Welsh town and see all the competing chapels…). The Welsh-language religious revivals of the 18th and 19th centuries saw the language through some massive social and cultural changes that could have killed it off – industrialisation and mass influxes of English speaking workers, for example – but they weren’t starting from scratch. And the Methodists (et al) didn’t just use Welsh in preaching, but also in literature and hymns, and they both built on and reinvigorated the language’s printed culture. One of the key factors in the survival of Welsh is that point made by John Davies: from the 16th century onwards Welsh was not *only* a spoken language: it was a print language too.

  27. Re lack of qualified interpreters and translators of Irish in EU:
    There has been the same problem with each new language ever added by the EU, most notably the 8 most recent. There has also been resistance to every previous addition, great or small, even Spanish back in 1986.
    Re ‘Erse’:
    “The archaic term Erse (from Erisch), originally a Scots form of the word Irish, is no longer used and in most current contexts is considered derogatory”, according to Wikipedia, which must be wrong on that one.
    Re the Statutes of Kilkenny and English in Ireland:
    It all depends what you mean by ‘English’. The ‘English’ apparently saw no contradiction in enacting this, for example (Statute III):
    Item ordine est et establie que chescun Engleys use la lang Engleis

  28. cm,
    Tha’s hilarious!

  29. Norman language – Wikipedia, the free encyclopedia claims:

    In particular a distinctive variety of “Norman French” remained alive in the barony of Forth and Bargy in Wexford until very recent times.

    Is there any basis to this or is someone confused?

  30. Good question. Their footnote says:
    “http://www.ucd.ie/irishfolklore/ The Department of Irish Folklore in University College Dublin possesses recordings of these individuals.”
    But a cursory examination of the site didn’t show me how to access those recordings, and a site search on “Norman” didn’t turn up anything useful. Bah.

  31. Re Norman French in Wexford:
    Someone is confused. What survived until recent times in Forth and Bargy (which I believe are two baronies not one) was Yola, a dialect not of French but of English.

  32. There are quite a few words in Irish that derive from French. Examples: seomra/chambre, garsún/garçon. In particular, the replacement of the Irish legal system by a Norman-French legal system brought in a number of new words.
    The late Seán Ó Tuama did a remarkable investigation of the relationship between the forms of medieval French poetry and traditional Irish song.
    However, this cultural exchange seems to have ended quite a long time ago.
    Let me throw out something else I’ve read somewhere. The language of the pre-Indo-European British, the people who built Stonehenge, was related to Berber. This conclusion supposedly was based on analysis of a small number of pre-Celtic placenames.
    Didn’t the Guanches, the original inhabitants of the Canary Islands, speak a language related to Berber? This would indicate some skill in seafaring. I’ve also read that Guanche teeth were very similar to Cro-Magnon teeth.
    Megalithic structures are found as far south as Malta. However, a similar culture does not necessarily mean a similar language.
    I must admit to buying those Dover Press books. Could someone update me on the current status of this theory?

  33. Oppenheimer’s idea is that the Belgic tribes living in SE England at the time of the Caesarian and Claudian invasions of the island were not Celtic speakers, but rather speakers of a Germanic language, essentially the ancient ancestor of modern English; the English language did not arrive here at the time of the mythical Hengist and Horsa, around the time of the rule of Vortigern and the withdrawal of the Roman legions in the fifth century, but much earlier, before the arrival of the Romans indeed. This thesis is argued with considerable cogency in his recent book, “The Origins of the British”, referring to a mass of sholarship in archaeology, classical history, ethnology and linguistics. His own expertise is in the field of genetic studies, and one must suppose that in his argument that the peoples of SE England are genetically most closely related to the peoples of North Gaul and North Germany must be taken seriously.
    This does not mean, however, that the Belgic British were speakers of a Germanic language – essentially proto Old English. What suggests the Belgic tribes of Britain at least must have been Celtic speakers are the following facts:
    1 All the place names known in SE Britain in roman times were of Celtic derivation: Londinium, Verulamium, Camulodunum, Durobrivae, Calleva, Venta, Durnovaria etc.
    2 There is not a single reference to an indisputably Germanic place name, tribal name, or personal name recorded anywhere in Roman Britain other than near Hadrian’s wall where soldiers from Germany were quartered.
    3 Personal names of SE British tribal leaders such as Caratacus, Boudicca, Cartimandua, and Commius are all Celtic.
    4 Contrary to what Oppenheimer claims in his book, there are numbers of Celtic place names in Southern and Eastern England, presumably surviving from Romano British times. In the area of Buckinghamshire where I live, ther is Chetwodwe (coed is Welsh for wood, anciently *ceto), Brill (bre/*briga – hill), Penn, and others. If the Belgae were English speakers, where did such celtic names come from?
    Oppenheimer is surely correct in assuming that, after the withdrawal of the legions, ther was not necessarily a mass evacuation of Britons to Wales, Cornwall and Brittany in the face of the English invasions from Friesland, Schleswig and Holstein (including Angeln, ther district of N Germany preserving the name of the Angli, a tribe recorded by Tacitus in the first century). and nor was there a great massacre of the Britons, as suggested by Gildas. What seems more likely is that there was a withdrawal of the upper class Romano British elite, who were replaced perhapss by English tribal chiefs, and perhaps an intermarrying of the Briton peasantry with the English lower orders, who in any case shared a common genetic ancestry.

  34. Thanks, Mike, that’s a sensible and convincing comment.

  35. MIke Hickox says:

    Mike G’s critcisms of Stephen Oppenheimer aretrenchant. Hopwever, he is still faced by the mystery which Stephen Oppemheimer attempts to solve-why so few Celtic loan words in English ?Even more mysterious given there seems to have been no massacre of the Britons.Perhaps the flight of a Romanised celtic elite could explain it. The language came to be associated with low social staus.

Speak Your Mind

*