The Origins of the British: The New Prehistory of Britain (34 page)

The real puzzle is finding a conservative Atlantic-coastal/Mesolithic genetic make-up (i.e. an undiluted founding hunter-gatherer
gene pool) specifically as a feature among Irishmen carrying Gaelic names. ‘Gaelic’ names should imply celtic language intrusions. But surely Gaelic did not arrive as the glaciers first melted? Some (e.g. archaeologist Colin Renfrew
¹⁰⁴
) have suggested that the branches of Indo-European languages spread from the Near East on the back of agriculture. But how could a new language arrive during the Neolithic without people? There are clear examples for language changing rapidly without major population replacement; the change from celtic Gaulish to French as a result of the Roman occupation is one. But there should be some traces of intrusion. Was 6% invasion enough to change culture and language? Probably yes – as we shall see in
Chapter 11
, 5.5% might have been enough to introduce Anglo-Saxon culture and language, although I keep an open mind on whether the Germanic language was already taking its first steps in Britain.

If the Irish Neolithic represented an invasion which brought celtic languages ultimately from the Balkans via Spain or France, why do so few Irish males carry intrusive Balkan or eastern Mediterranean Neolithic male gene lines? Most studies of male names and the Y-chromosome show strong associations. From the evidence associating Rory with male Gaelic names,
¹⁰⁵
it could be that these Gaelic names were adopted by non-Gaelic-speaking indigenous people. If this were the case, why is there a significantly
higher
frequency of specific Atlantic coastal Ruisko Y markers, such as Rory, among Irishmen with Gaelic names than among Irishmen with names of other linguistic derivations?
¹⁰⁶
A brief examination of this paradox shows that if the males with non-Gaelic names were more recent immigrants to Ireland, then it would not matter whether the Gaelic names
were inherited, adopted or given to individuals in the vernacular – they would still be associated with the older male lines.

As I have been suggesting, the uncommon Rory cluster, which accounts for one-third of Gaelic-named Irishmen, may well have moved north up the Atlantic Coast from Iberia into Ireland even before the Younger Dryas, and re-expanded afterwards. The rich Mesolithic lifestyle continued in Ireland until about 6,000–5,000 years ago, when much of the rest of Europe was already moving fully over to farming. This possibility was even allowed for by geneticist Emmeline Hill of Trinity College Dublin and her colleagues, in demonstrating the association of Ruisko generally with Gaelic names.
¹⁰⁷
Whether the Gaelic language branch actually moved up to Ireland at the same time is of course a matter of irresolvable speculation, but it again brings us back to the language problem I began the chapter with. That problem is of course the frustration of comparing language reconstruction with the archaeological record when stones and pots do not talk and most linguists ‘don’t do dates’.

Another simplistic approach to fixing the arrival of celtic languages might be first to define the period of greatest immigration from a genetic perspective and then to try to hook celtic languages to that time. However, when we add up all the genetic lines coming into the British Isles,
¹⁰⁸
we find that what is left over after the Neolithic is insufficient to call a real mass immigration, male or female, Celtic, Anglo-Saxon or Viking. The founder work by Martin Richards and colleagues is sufficiently precise to confirm that post-Neolithic Near Eastern maternal immigration accounts for no more than 5% of modern indigenous maternal DNA lines in north-west Europe.
¹⁰⁹
For the British Isles at the Atlantic fringe, this figure is likely to be even less. My analysis of
ages and numbers of post-Neolithic British Y-chromosome lines is similarly low (discussed in
Part 3
, and see Appendix C), but suggests an overlap in migration events to Britain between the Late Neolithic and Early Bronze Age.

The Neolithic period in Britain lasted nearly two millennia, from the first cultural arrivals over 6,000 years ago up to the elite Wessex Bronze Age cultures starting 4,000 years ago, well after the arrival of Bell Beakers (see
Chapter 2
and
Figure 5.12
). I would love to have been a fly-on-a-megalith and really know which of the four cultural sweeps that occurred during the British Neolithic (which I shall shortly summarize), or those even later during the Bronze and Iron Ages, brought celtic or any other languages.

I suspect that if celtic languages did arrive during the Neolithic, they came rather early, and to Ireland and the British Atlantic fringe first. This would certainly be more consistent with the deeper estimated age of the celtic-language branch discussed in
Chapter 2
. As we have seen, the actual proportion of ‘fresh Neolithic’ lines in this western region was rather small, although they derived from the same general Iberian direction as did nearly all the older lines. As with the spread of new languages into the Pacific over the past few thousand years, an alternative view to their being part of a Neolithic juggernaut is that they were the new maritime trade-net language allowing people who already had ties of culture and genes to exchange prestige materials such as copper, gold and beaker pots more easily.

A less obvious problem is the fixation in all the academic literature on celtic languages to the exclusion of any others. While it
is generally accepted that there were other languages, probably non-Indo-European, in the British Isles before celtic, few have speculated as to what these may have been or as to whether there were non-Indo-European influences persisting from before the arrival of celtic in the Isles. Munich-based German linguist Theo Vennemann has addressed all of these issues and although his reconstruction is controversial, there are extraordinary resonances with the genetic picture.
¹¹⁰
First, Vennemann argues for an ancient post-glacial European language sub-stratum on the basis of river-names. He calls this language family Vasconic (i.e. linguistically like the Basque and as with their re-expansion, originating in the Basque refuge and spreading north, west and east). This sub-stratum was progressively overlaid from southeast Europe by Indo-European during the Neolithic starting from 7,500 years ago, moving through central Europe and reaching Scandinavia by 6,000 years ago.

A third language group which Vennemann calls ‘Atlantic’ (a non-Indo-European language and not to be conflated with anything Atlantean!) spread west along the Mediterranean, through the straits of Gibraltar and up the Atlantic coast to the British Isles and Scandinavia by 6,000 years ago. He argues the ‘Atlantic’ language was associated with the western European megalithic fringe (see below and Figure 5.8) and preceded the arrival of celtic in the British Isles. By studying the structural effects of this linguistic type both on Insular Celtic, and subsequently on English, Vennemann deduces that it was in fact a Semitic language. ‘Atlantic’ would have arrived, like Phoenician another Semitic language, from farther east in the Mediterranean, with its last relict surviving in now-extinct Pictish. The interesting point about Vennemann’s hypothetical Semitic language
‘Atlantic’ is that it provides a linguistic-geographic partner to male line J2 which does characterize Semitic-speaking peoples in the eastern Mediterranean ‘
Figure 5.8b
’ and, although following the same general coastal route as Balkan-derived E3b and I1b2 (
Figures 5.6
and
5.8a
), has a distribution distinct from the latter, favouring not Wales and Ireland, but Scotland, particularly Pitlochry, a Scottish town with a Pictish place-name. A similar Scottish distribution also applies to maternal line J1b1 which, in spite of its link to Norway (pp. 216–7), derives ultimately from the Near East via Iberia and along the Atlantic facade.
¹¹¹
Moving away from the influence of the Mediterranean we can see that, compared with the Atlantic fringe, England and northern Britain received a much larger influx of genes from north-west Europe across the North Sea during the Neolithic. Were they accompanied by new language? Or, in line with the Renfrew language–farming hypothesis, did yet another branch of Indo-European arrive with the Norfolk Neolithic? Apart from English and Brythonic and Vennemann’s hypothetical Vasconic and Semitic influences, there are no linguistic relicts left, Indo-European or otherwise, except perhaps in some incomprehensible place- and river-names, such as London, Kent, the Thames and the Severn, and the vexed Pictish question (see
Chapter 2
).

So it is difficult to speculate on what other languages might have been brought across the North Sea during the Neolithic. The orthodox view is, of course, that there is no need to look for other languages since insular celtic was universal by the time the Romans arrived (but see the discussion in
Chapter 7
). However, given that since the Early Neolithic there were consistent cultural and genetic differences between eastern Britain and the Atlantic fringe, not to mention the two different
sources of genetic entrants, the question of whether there was another Neolithic language in England does need asking.

To me, the obvious default candidate is an ancestral Germanic branch. As discussed in this and the previous chapter, if Renfrew’s paradigm of Indo-European branches carried by different Neolithic spreads from Anatolia and the Balkans is correct, then the LBK culture may have carried the ancestral Germanic branch of Isidore Dyen’s Meso-European cluster (see
Chapter 2
) up the Danube to Germany along with Y-chromosome group Ian. I shall discuss Forster’s minority view, arguing for the antiquity of English, later in this book, but I quote here an excerpt from his conclusions, which allow the possibility of English diverging as a separate branch of Germanic during the Bronze Age or even earlier:

The … analysis reveals a Scandinavian influence on English and apparently a pre-Scandinavian archaic component in Old English. All Germanic lexica spoken today appear to converge in the network on an ancestral Common Germanic lexicon spoken at an unknown time, but constrained to before
AD
350 and probably after 3600
BC
.
¹¹²

Forster’s view is consistent with Dyen’s lexico-statistical analysis, in which English separates early from Dutch, German and Scandinavian languages and forms a deep branch from core Germanic on a par with the Brythonic/Goidelic split.
¹¹³

Before grasping the nettle of language again and trying to blame it on Neolithic farmers (see
Chapter 6
), I should like to put these speculations on the British Neolithic in a more cultural context, since I have so far only sketched the dates of its commencement in the British Isles. There are two more
‘cultural ages’ prior to the historic period which could have been associated with gene flow, and even language flow, into the British Isles and need to be considered. These are the Bronze Age and the Iron Age; in the orthodox view of course, the latter (discussed in
Chapter 1
) sees the arrival of celtic languages into the British Isles.

The first Neolithic cultural development happened in the west on the
Atlantic
side of the British Isles, and took its initial form in a rapidly evolving monumental attention to the dead. From around 5,800 years ago, megalithic ‘portal tombs’, literally blind gateways to the other world, began to appear on the coastlines of Ireland and Wales. Today these tombs can be recognized on the landscape, shorn of their earth covering, as dolmens. Dolmens typically have at least four stones, two large uprights representing the gate, a smaller upright at the back, and these three propping up a large flat slab on top. Dolmens continued to be built for up to six hundred years.

Contemporary with these, but outlasting them in fashion, was a more elaborate industrial type of megalithic grave suited to repeated communal burials: the passage tomb, lined with massive, flat slabs. Northern Ireland saw a proliferatation of local variants named court tombs (or cairns) after the courtyard opening into the passage. Similar designs appeared across the short sea-gap from Antrim in western Scotland, where they are known as Clyde cairns (
Figure 5.9
). Although most of the numerous passage graves were relatively simple, several such as those at Knowth, Dowth and Newgrange in the Boyne Valley,
Ireland, were built on a truly monumental scale: 85 metres in diameter and 11 metres high, with extraordinary attention to detail of celestial alignment and including decorations such as petroglyphs and quartz stone (Plates 6).
¹¹⁴

Megalithic chamber tombs were fashionable along a vast coastal belt during the Early Neolithic, stretching from Corsica and Sardinia in the Mediterranean, along the south and west coasts of Iberia, through Brittany and Normandy to the south coast of England and the Atlantic fringe of the British Isles (Plate 7).

But chamber tombs also appeared to the
east
of Britain, as far north as Frisia, north-west Germany and Denmark on the Continental coast, where between 5,600 and 5,200 years ago they went through their own megalithic evolution, from simple then polygonal dolmens, through passage graves, to complex chambered and gallery graves. These were associated with so-called
funnel-necked-beaker
(in German
Trichterbecherkultur
or TRB pottery), derived ultimately from LBK. North-west European Dolmen burials may have evolved locally from earlier long barrows, a tradition which had, in turn, seeped into eastern England during its earliest Neolithic phase (
Figure 5.9
).
¹¹⁵