The Origins of the British: The New Prehistory of Britain (11 page)

Collis’s only constructive conclusion, in his excellent deconstruction, is grudgingly to acknowledge that if there were a historical-linguistic link between Celts and celtic languages, then their origins would include central and western France rather than south-west Germany.

So, we are left with objective evidence of the presence of Celts and celtic languages in France, Italy and Spain during Roman times. By combining inferences from early Greek writers with modern linguistic analysis, we can see that the presence of Celts and celtic languages in south-west Europe, and maybe even the western parts of the British Isles, stretches back to before the middle of the first millennium
BC
.

In my view, this perspective vindicates Lhuyd’s provisional naming of his clutch of non-English languages of the British Isles as Celtic with a big ‘C’, although I shall continue to use the small
‘c’ for language. We can now move on to look at them in more detail.

I have done my best to vindicate much of Pezron and Lhuyd’s broad linguistic claims, and, unlike Simon James, I feel that these claims justify continuing the study of modern insularceltic languages in the context of the classical Celts. But, there are several loose ends to address. The most important of these is their tacit assumption, still perpetuated, that
all
Ancient Britons were celtic-speaking (i.e. that the whole of the British Isles spoke celtic languages). While much of the western and northern British Isles was unarguably celtic-speaking, as far back as records go, the same cannot necessarily be claimed for England. There is fairly good evidence that over the past two thousand years celtic languages were spoken almost universally in those areas – Wales, Cornwall, Cumbria, Scotland and Ireland – that we now associate with a ‘Celtic heritage’. Indeed, hundreds of inscriptions on stone, made after the Romans had left, are ample evidence of how celtic languages thrived in Ireland, Scotland, Wales, Cumbria and Cornwall. However, England (apart from the West Country and Cumbria), although rich in stone inscriptions, is notably practically devoid of any in celtic dating from any period (
Figures 2.2
and
7.4
).
¹⁹

As far as the subjective visual impressions are concerned, Tacitus compared the southern Welsh exclusively to the Spanish in one of the few contemporary descriptions of the people of south-west Britain:

Figure 2.2
Gallo-Belgic coins and their British derivatives do not overlap later celtic stone inscriptions in the British Isles. That is possibly a coincidence, but it is consistent with a prior Belgic human (and/or cultural) migration to southern England, which could be inferred from Caesar and Tacitus. (For detail of Gallo-Belgic coin distribution, see
Figure 7.3
.)

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.
²⁰

The Silures inhabited south Wales. Along with the Ordovices in central Wales and the Deceangli in north-west Wales and Anglesey, they put up an extremely fierce and prolonged resistance to Roman occupation – a sense of independence that has lasted. Tacitus here makes a direct migratory connection between the Silures in Wales and the people of Spain, on the other side of the Bay of Biscay. He makes this comment based on physical appearances, in a throwaway remark that is reminiscent of similar casual subjective comments still made today about the Mediterranean appearance of Welsh people.

The Spanish connection also makes one think again of the ancient Greek and Punic references to the tin trade along the Atlantic coast from Spain up to the west coast of Britain. These references support the possibility that a ‘Celtic’ entry to the western side of the British Isles came from the south rather than from the nearby Continent on the other side of the Channel. What other information Tacitus may have based this remark on we shall never know, but it also emphasizes that there was a lack of similarity between the Welsh and the English even in his day. I shall discuss the Spanish connection shortly, but for the moment such comments mean that we cannot assume that the inhabitants
of Roman England had exactly the same origins as the Welsh, let alone being ‘Celtic’ or universally celtic-speaking.

I shall come back to the question of what other languages apart from celtic
were
spoken in Roman England in
Chapter 7
. For the time being, however, the patchy evidence in England tells us that we cannot assume that the Ancient Britons of England all spoke celtic, or that the English language is merely the result of complete physical and cultural replacement of Celts by the Anglo-Saxons starting from around
AD
400. In those parts of the British Isles that were clearly celtic-speaking throughout (
Figure 2.1b
), there is abundant evidence from inscriptions, literature and modern languages to chart their history of change and movement since Roman times.

Most of the known recent celtic languages of the island of Britain form a group known as
Brythonic
. As far as Wales and western England are concerned, it is fairly safe to assume that Brythonic celtic languages have been in continuous use since Caesar’s time, and most likely before. Apart from Welsh, Cornish is the only other living Brythonic language on the British mainland. Cornwall (or Kernow) was also known by Romans as Dumnonia and by the Saxons as the Kingdom of the West Welsh. Cornish (Kernowek) was spoken continuously until three hundred years ago. From then it declined, and just before the twentieth century it became extinct. The last monoglot Cornish speaker may have died in 1676, but there were still a number of fluent native speakers available for Lhuyd’s work, and he included the language in his study. As the result of work by a number of enthusiasts in the twentieth century, reconstructed
Cornish has now made a dramatic revival, with an estimated 3,500 speakers today. Cornish is now officially recognized by the United Kingdom Government as a minority language under the European Charter for such.

There is good evidence that Brythonic celtic languages, probably South-west Brythonic, were also spoken elsewhere in the West Country, in Devon. After the Romans left, the Kingdom of Cornwall persisted during the Dark Ages.

As may already be clear, Breton, the celtic language of Brittany, was no ancient Gaulish remnant stranded on the horn of the Continent (as was mistakenly thought by Pezron) but a Brythonic tongue, closely related to Cornish, which at some point over the past several thousand years moved across the short stretch of water between Britain and Brittany. Evidence for Old Breton, dating from the eighth to the eleventh century
AD
, can be found in lists and glosses in documents and as names in Latin texts. There is evidence for some borrowing from Gaulish into Breton, which is either an indirect sign of its longevity in Brittany or of the presence of Gaulish so far north and so late. Although not recognized officially by the French Government, Breton has half a million speakers, thus rivalling Welsh as the most flourishing of all the modern celtic languages.

Another Brythonic language, now extinct, can be fairly safely inferred: Cumbric, which was similar to Welsh. Pictish, formerly spoken in northern Scotland, is claimed to have been Brythonic, but whether this claim covers all languages present there in the first millennium
AD
, apart from Scottish Gaelic, is still disputed by a few (see below).
²¹

Except in Cornwall and Wales, there is little evidence from the spoken word remaining today to support the view that Brythonic languages were spoken throughout England in Roman times. However, even today, remnants can be found of a Brythonic-celtic language previously spoken in Cumbria and elsewhere in north-west England. When I was a child, the only celtic words spoken north of Wales were from dialects of a language closely related to Welsh, known generally as Cumbric. Even now, I can recall my brother being taught by a friend at school how to count from one to twenty in numbers used by North Country shepherds to count their sheep. As far as I remember it went something like this:
yan
,
tan
,
tether
,
mether
,
pimp
(5),
sether
,
hether
,
hother
,
dother
,
dick
(10),
yan dick
,
tan dick
,
tether dick
,
mether dick
,
bumfit
(15)
yana bumfit
,
tana bumfit
,
tetherer bumfit
,
metherer bumfit
,
giggot
(20). Many distinct variants of this sheep-counting system are recorded from the Lake District (Cumbria) and the Pennine Hills, north from the west coast right through to Ayrshire in the western lowlands of Scotland: in Keswick, Westmoreland, Eskdale, Millom, High Furness, Wasdale, Teesdale, Swaledale, Wensleydale and Ayrshire.
²²

This method of counting things in four tallies of five up to twenty was known as ‘scoring’, hence the term ‘score’ for twenty. Counting in fives reflects the use of fingers, if not toes. After twenty, a scratch or score was made on a piece of stone or wood and the scoring recommenced. The word ‘tally’ comes from the Latin
talea
, meaning literally a stick with notches.
²³
Cumbric scores have even been found in a number of places in the USA. In Cincinnati we find
een
,
teen
,
tother
,
feather
,
fib
,
soter
,
oter
,
poter
,
debber
,
dick
; and from Vermont there is
eeni
,
teni
,
tudheri
,
fedheri
,
fip
,
saidher
,
taidher
,
koadher
,
daidher
,
dik
.
Cumbrian settlers presumably took these celtic counting-words to the New World. Children’s counting rhymes also retain relics of Cumbric scores, although the counting is in fours, presumably to suit the rhyme: from Edinburgh, for example, ‘Inty, tinty, tethery, methery; bank for over, dover, ding …’ and the universal ‘Eeny, meeny, miney, moe …’
²⁴

This numeric digression from celtic remnants in England took us briefly to Welsh, which I shall come back to in slightly more detail later. Linguistic reconstruction, a well-recognized method based on knowledge of systematic sound changes, indicates that both Welsh and Cumbric are related but distinct languages sharing a common ancestor, with Kernowek, in Brythonic Celtic. This branch has descendants only in mainland Britain and in Brittany, and is distinct from another group of insular-celtic languages, found in Ireland and Scotland and known collectively as Gaelic or Goidelic (see below).

When we look at what Roman writers have to say about the tribes of northern England, we find that they mention one group which occupied all the areas where the Cumbrian celtic-language relicts are still found. They were known as the Carvetii, but that is about all that is known about them from the texts. Their territory included all of Cumbria (the Lake District) and parts of north Lancashire and south-west Durham (the Pennines), and south-east Dumfries and Galloway in Scotland (
Figure 2.3
). They were probably mainly hill farmers, with few large settlements apart from ones associated with the numerous Roman forts in the area. They did not make their own coins. For some obscure reason their entire county and its towns were left out by Claudius Ptolemy from his famous second-century
Geographia
(
chapter 2
‘Geography of Albion Island of Britannia’). This omission was probably a block clerical error, since there is none of the concordant detail that should be expected, for instance cross-classification of Carvetian towns with their neighbours, the Brigantes. Their Roman provincial capital was Luguvalium (Carlisle). Further north into western Scotland, the area occupied by the Damnonii would have corresponded geographically to where today’s remnant Brythonic counting numbers are found in Ayrshire.