[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

[image ALT: link to previous section]
Chapter I
This webpage reproduces a chapter of
Celtic Britain

by Nora K. Chadwick

published by Frederick A. Praeger
New York

The text and engravings are in the public domain.

[image ALT: link to next section]
Chapter III

p35 Chapter II

Celtic Rule in Britain

Native Celtic life had continued unbroken throughout the period of Roman rule in Britain, and we can watch its emergence during the closing years, and after the final departure of the Roman troops. Our evidence is perhaps fullest and clearest in the survival of the heathen Celtic religion. The most striking example of the survival of a native cult is that of the temple of Lydney,1 built on a precipitous site on the right bank of the Severn. Here, some time after 364, probably even after the Great Raid of 367, an elaborate temple was founded and dedicated to the god Nodens, or Nodons. The temple was apparently therapeutic, perhaps Aesculapian in character, and on the site were also offices, cells for the patients, and guest quarters for visitors. The buildings and coin hoards, the mosaics, the votive tablets and offerings, including a splendid little bronze wolf-hound, suggests that this elaborate establishment was for wealthy patrons. One of the inscriptions in a tessellated pavement, now destroyed, interpreted as stating that an officer in charge of the supply depôt of the fleet laid the pavement out of money offerings,2 is an interesting indication that there was still an official in charge of a naval depot in the Bristol Channel. The three dedicatory inscriptions give clear evidence of a close rapprochement with the native population, for there can be practically no doubt that the wealthy god Nodens is identical with the Irish god Núadu (Argatlám), 'Nuadu of the Silver Hand', and the Welsh Lludd Llau Ereint. The temple seems to have survived until the fifth century; but the quality of the repairs and the barbaric coinage of the latest phase are signs of decadence, and the substantial enclosing wall, now for the first time erected round the precincts, is an ominous witness to the unsettled times.

p36 The temple at Lydney had been constructed within the limits of an Iron Age hill-fort; and in the native hill-top city of Maiden Castle near Dorchester in Dorset another late Roman sanctuary had been built after 364, a simple building on the Romano-Celtic plan. We know of other examples of such late Roman temples built and richly furnished at a time when Christianity was making rapid headway in town and country, and it is thought that they may have been prompted by a heathen reaction, probably under the influence of Julian the Apostate.

The Roman type of sanctuary in Britain, as in Gaul, was a high square building surrounded by a portico, the actual architectural features being of Roman inspiration. At Silchester all the temples are of this Romano-native type and are decorated in Roman fashion. Even at Verulamium one of the oldest and most important temples was built on the characteristically Romano-Celtic, as opposed to the Mediterranean, plan.3

These temples, local in style, are not the only indications of the flourishing condition of Celtic religion under the Romans. The splendid group of round, steep-sided barrows at Bartlow in East Anglia, with their rich Roman furnishings, still stand proud witness to the wealth and importance of the British high aristocrats under Roman rule. On a humbler plane many wayside shrines have been found with Celtic associations in a Roman context, and Celtic spring and river gods are known from Roman sculpture and river names.

[image ALT: A drawing of two conical mounds, about 8 meters high and covered with vegetation. They are the Celtic barrows at Bartlow in East Anglia.]
Fig. 4. An impression of the two Bartlow barrows (cf. Plate 1)

In recent years the importance of the evidence of continuity is coming to be more fully recognised. Over and over again we find, in both temples and municipal buildings, that as the Roman buildings deteriorated in the late fourth and fifth centuries, repairs were effected, inferior in workmanship, and even new buildings were constantly erected on the ruins. In Verulamium the evidence of continuity is particularly interesting. Here, for example, a house built well after the middle of the fourth century had been enlarged later and furnished p37with lavish mosaics, and changes continued to be made in the structure down to 410. Still later, at least until 450, the technical skill necessary to maintain the city's aqueduct and to install a piped water supply was still available, and the civilised needs implicit in such a demand were still continuing.4

Intellectual continuity is no less clear. In the fifth century the Pelagian heresy was especially prevalent in Britain, 'the home of its birth', so we learn from Prosper of Aquitaine. Education was to some extent flourishing. A series of anonymous letters,5 still extant, are believed to have been written by a Briton travelling abroad to his father, a bishop in Britain, in which reference is made to the education of boys, suggestive of the Roman high school system similar to that on the Continent. If we can rely on the evidence of St Patrick's Confessio — and the fifth-century date of the text is not in doubt — the Orders of the Christian Church, bishop, presbyter, deacon, were still observed even in the west of Britain. The stone funerary inscriptions of the fifth and sixth centuries show that Latin was still understood p38and used for formal purposes. The quality of the lettering has not yet seriously deteriorated.

The Romano-British fusion, which had been achieved through the long period when Britain formed part of the Roman Empire, and the continuity of Britain native life were the basis on which Britain now had to build up her independent existence from within, and to resist disintegration and conquest from without. Henceforth the Celtic peoples of Britain were compelled to fall back on their native resources. The first need was to carry on some form of civil government for internal administration; the second, to look to her defences.

Britain had been divided by Severus into two provinces, distinguished as Upper and Lower Britain, though the nature of the division is not quite clear.6 Central and south-eastern Britain had contained most of the villas, chiefly occupied by native Romanised Britons, and also most of the towns, the important centres of civil administration. But into the Roman administrative system the earlier Celtic kingdoms, the tribal areas, had been absorbed almost intact, and each had its own tribal centre, its civitas, as in Gaul.

Commercial towns, such as London, and the coloniae, originally Roman civic urban settlements of retired troops, each doubtless had its ordo or permanent executive body. It seems clear that certain of these native civitates, or what we may call 'city states', had now become sufficiently Romanised to be able to carry on a form of organised civil life for some time, and a few at least of the civitates are known to have persisted.

Northern and western Britain were in general far less Romanised. In Wales, with the exception of the Silures, with their capital at Caerwent (Venta Silurum), we know of no self-governing system; but inscriptions to a cives of Venedotia ('a citizen of Gwynedd'), an ordous of north-central Wales, and a magistratus of Penmachno in Caernarvonshire,7 warn us against an incautious generalisation from negative evidence.

p39 It is evident that the Celtic peoples were not caught unready for defensive measures against external aggression. The Gaulish poet Claudian had been right in summing up the chief peril of Britain as arising from the fact that her enemies were now seeking entry on three fronts, the Irish in the west, the Picts in the north, the Saxons in the east. So far none had succeeded in invading on a large scale. The strong Pictish kingdoms were still contained north of the Antonine Wall by the British kingdoms stretching in a long unbroken arc from the Firth of Forth to Land's End. The north British princes would have no more wish than had the Romans to be conquered by their old enemies the Picts; and in the west the Welsh would be as concerned as the Romans had been to foil Irish attempts at penetration.

Owing to the careful preservation of the Celtic pedigrees, the North British princes, now left in control of the northern defences, are no strangers to us.8 They were represented in the fifth century by at least three important families. Of these the great Strathclyde dynasty with its stronghold at Dumbarton guarded the west end of the Antonine Wall and counted among its chief members a certain Ceredig Gwledig,9 a term of high distinction, probably the Coroticus reproached in St Patrick's famous Letter for carrying off some of his converts into slavery. His immediate ancestors bear names which look like corruptions of Roman names, and before them come three which have distinctly Pictish features. This would carry us back to a period shortly after the great raid of 369, and suggests that first, Pictish chiefs, and later, Romanised Britons had been employed to guard the western end of the Northern Wall as a result of the raid.

Bede implies the strength of the old British kingdom of Dumbarton ('Dún Breatann') when he refers to it (H. E. I.1) as a munissima civitas ('a very strongly fortified civitas') down to his own day.

p40 The family of Ceredig is the best known branch of a number of dynasties of south-western Scotland who trace their ancestry to a certain Dufnwal Hen, and whose family epithet is Hael. A tombstone at Yarrowkirk in Selkirkshire, once, like the Catstone, included in an extensive Christian cemetery, bears an inscription commemorating 'the most famous princes Nudus and Dumnogenus, two sons of Liberalis'.10 Liberalis is the Latin translation of Hael, 'munificent', 'wealthy'. These people had no coinage or organised trade. Can the wealth of the family have originated in Roman subsidies to enable them to defend their fort and the west of the Northern Wall? Did Ceredig and his soldiers find it convenient to support the family epithet by slave raiding and slave trading when Roman finances broke down? The silver treasure of the coeval kingdom on Traprain Law, at the eastern end of this Wall, was perhaps looted from Gaul under the same conditions.

The eastern counterpart of the kingdom of Ceredig in Strathclyde was Manau Gododdin around the head of the Firth of Forth. The name Gododdin (earlier Guotodin) seems to preserve that of the old tribe of the Votadini, which in Roman times stretched from Forth to Tweed.11 Nennius, making use of a much earlier written northern source, records a story of a chief named Cunedag (modern Welsh Cunedda) who migrated from this region to North Wales with eight of his nine sons, leaving the eldest behind in Manau, and adds that they 'drove out the Irish with immense slaughter'. According to Welsh genealogies, Cunedag's immediate forbears, like Ceredig's, bore Roman names for three generations, and his grandfather bears the epithet peis rut, 'the man with the red cloak'.12 Before these the names are Pictish, while his own name and those of his sons are normal British names. This family, like that of Ceredig, apparently grew to importance in Roman times, probably guarding in their turn the eastern end of the Northern Wall against the Picts.

p41 Another North British prince, a certain Coel Hen, founded one of the three great families ruling and founding dynasties on the northern border before the end of the fourth, and during the fifth and sixth centuries. These we shall meet later, fighting valiantly against the Anglian invaders in the sixth and seventh centuries.

Evidence for the defences of western Britain follows a similar pattern. In the Notitia Dignitatum no entries whatever appear for any western frontier garrisons south of Lancashire. Had the Welsh Border defences, like those of the Scottish Border, been entrusted to native local princes? The evidence of their genealogies and of stone inscriptions suggest that this was so.

In the first place, we have the well-known Christian tombstone in Carmarthenshire which bears in Latin lettering the inscription Memoria Voteporigis protictoris, and in addition in the native Celtic alphabet, known as ogam, the Irish form of the name, Votecorigas. The epithet protector was given in the later days of the Empire to barbarian princes honoured with foederati status as 'protectors' of the frontiers on behalf of the Romans. There is little doubt that Voteporius is identical with prince Vorteporius, the 'tyrant of Demetia' (Dyfed, Pembrokeshire) censured by Gildas in the sixth century among the rulers of Britain (cf. p42 below).

The title protector is applied to early members of the family of Magnus Maximus on the royal genealogy of Pembrokeshire. But the early steps on these Pembroke genealogies are too vague and corrupt to merit acceptance. They carry back the ancestry of (Magnus) Maximus to Constantine the Great, and are of course inadmissible. They probably owe their origin to the desire of a descendant of Rhodri Mawr of Gwynedd (cf. pp73 f. below) to trace his ancestry to a famous Roman General of Gwynedd. An alternative genealogy of the ruling family of Pembrokeshire,13 believed on doubtful evidence to date from the third century A.D., derives the dynasty from an Irish prince p42Eochaid Allmuir ('Eochaid from overseas'). Here, then, we have rival traditions, one Romano-British, one Irish, for the rulers of the kingdom of Pembrokeshire, the home of Voteporius Protector; and the title Protector, applied to members of the family of Maximus, suggests that his functions may have been comparable with those of Voteporius, 'Protector' of Dyfed. His two immediate forbears also bear Latin names, and these three generations are the first which agree in all our three versions of the Pembroke genealogy. We are hardly left in doubt that the Romans had given their support to an Irish family on the south-western frontier of Wales to protect an area, already bi-lingual and largely Irish in population, against further Irish encroachments — in fact the old story. And indeed Maximus in his Roman fortress in Caernarvon, whatever his precise title, must have had this function also.

The protecting functions of those in charge of the western peninsulas of Wales is comparable with those of Ceredig Gwledic14 and of Cunedag, guarding the northern frontiers against Picts and perhaps Frisians entering by way of the Firth of Forth, later known as the 'Frisian Sea'. Again, reference may be made to the story of the settlement of Cunedag and his sons in the maritime regions of North Wales to rid her of Irish settlers (p40 above). The Welsh pattern of federate protection probably helps to explain why the Notitia Dignitatum has no entries of forts or officials in the west. Here, however, still more momentous developments are recorded which changed the entire course of British history. At this point we leave all contact with Roman officials and titles, and henceforth we depend solely on British evidence, recorded for the most part in Latin, from oral tradition.

In the first half of the sixth century the author of the De Excidio Britanniae, whom we, in common with historians generally, will call Gildas, tells us of a powerful British ruler whom he refers to as a superbus tyrannus, and whom he evidently p43regards as responsible for the Saxon invasion and occupation of Britain.15 No name is given to him in the oldest and best text, but Bede, in his Chronica Majora, calls him Vertigernus, a form which he must have obtained from an early British source, whether Gildas or another. In Anglo-Saxon the form given to the name is Wyrtgeorn, and in later writers it appears as Vortigern, generally regarded as a proper name, though it means literally 'overlord' and may originally have been a title. In the Latin of this period the word tyrannus (lit. 'tyrant') generally signifies a usurper, and this is doubtless what Procopius implies when he speaks of Britain as largely in the hands of 'tyrants' after the departure of the Romans. A correct translation today of the superbus tyrannus would be 'absolute dictator'. In much later sources, Irish and British, Vortigern is styled rex Brittanorum.

Here again, this great Border prince is preceded on his genealogy by three generations of Romanised forbears, and his great-grandfather bears the eponymous name of Gloui (apparently from Gloucester), to which is added the epithet Gwalltir, 'of the long hair', perhaps from the long horse-hair of the Roman soldiers' helmets. Gildas tells us that at that time all the councillors, together with the supreme dictator (omnes consiliarii cum superbo tyranno) invited the formidable (ferocissimi) Saxons into the country. This places the responsibility for the invitation not so much on the superbus tyrannus as on the consiliarii, who were probably the governors of the 'city states', with the tyrannus at their head. Moreover, Gildas makes it quite clear that the Saxons were officially invited in as mercenaries to fight on behalf of the Britons. The invitation was in accordance with recognised practice, and Vortigern's action was a perfectly constitutional one. The story suggests that no anarchy followed the withdrawal of the Romans.

Who Vortigern was we do not precisely know, but he is the leading figure of the fifth century in most of our traditions p44of the west and the Welsh Border. His ancestral territory was probably in south central Wales, for the medieval territory of Gwerthrynion contains his name, and according to Nennius the later kings of Buellt (Buillt) and Gwerthrynion (in modern Brecknock and Radnorshire) traced their descent from Pascent, his third son, this territory having been bestowed on him after Vortigern's death by Ambrosius, 'who was king among all the flocks (greges) of the British nation'. According to an early reading made by Edward Lhuyd in 1696 of the pillar of Eliseg in the Vale of Llangollen,16 the kings of northern Powys traced their ancestry to a marriage of Vortigern with Sevira, the daughter of Magnus Maximus; but doubt has recently been cast on this reading.17

In later traditions, both Britain and Anglo-Saxon, Vortigern's power seems to have extended as far as Kent, and may indeed have included the whole of the Roman province. But he is never represented as fighting, or called gwledig or dux, or even protector, and Nennius tells us (ch. 43 ff.) that his army was commanded by his son Guorthemir. Our traditions on the whole suggest a Roman provincial governor, such as a vicarius, possibly the last provincial governor of Roman Britain. A large body of unfavourable legends gradually collected round his name, and he was universally execrated for his part in the adventus Saxonum; but these legends are obviously derived from saga recorded and cultivated in a hostile milieu, such as the fantastic account given by Nennius, writing nearly four centuries later, of his encounter with St Germanus. The same source tells us that he died non cum laude, and records three variant versions of his death. What seems clear, however, is that there was in fact a powerful Romano-British prince of the western border who played a leading part in the official entry of the Saxons into our history on a nation-wide scale.

In the passage of De Excidio immediately following the invitation to the Saxons and their later devastation, Gildas p45states (ch. 25) that a remnant took up arms and challenged their victors to battle under the leadership of a certain Ambrosius Aurelianus, a vir modestus, and a dux, whom he describes as 'almost the last of the Romans', and whose parents had worn the purple and had been killed in the Saxon invasions. Their descendants, with this single exception, had greatly degenerated from their ancestral nobleness. From now on, Gildas continues, sometimes the people were victorious, sometimes the enemy, till the last great slaughter of the Saxons in the year of the siege (obsessio) of Mount Badon, recorded in the Cambrian Annals for 516.

Elsewhere (ch. 28 f.) Gildas attacks by name five British princes who appear to have been his own contemporaries, and rulers of wide territories on the western seaboard; as they are apparently his seniors his picture is in some measure that of the closing years of the fifth century. He addresses them apparently in geographical order, beginning with Constantine, 'tyrant' of Dumnonia (the Devon-Cornwall peninsula), whom he calls Catulus. Then comes Aurelius Caninus, whose order in the list — to say nothing of the juxtaposition of 'cat' and 'dog' — suggests that his location is in south-east Wales. Next comes Vorteporius, seemingly of Pembrokeshire, followed by Cuneglasus, probably the Cinlas of west Wales, great-grandson of Cunedag, and finally Maglocunus, the great Maelgwn Gwynedd, also a great-grandson of Cunedag, and the ruler of North Wales and Anglesey. Gildas's survey has covered the western coastal kingdoms of Britain, from Devon and Cornwall to Lancashire, and would probably link up with the wide north-western kingdom of Rheged (cf. p63 below). These wide western kingdoms had probably served as the bulwarks of Britain against Irish encroachment.

In chapter 42 of the Historia Brittonum, Nennius records stories of a certain Emreis (l. Ambrosius) with the epithet Guletic, who is victorious over Vortigern in a contest of magic p46in the neighbourhood of Snowdon. Emreis claims to be the son of a Roman consul, and Vortigern is here stated to bestow on him a citadel18 and all the western side of Britain, while he himself moved northwards. In ch. 31 we had been told that Vortigern had been beset by fear of Ambrosius (timor Ambrosii), and after Vortigern's death Ambrosius bestows the regions of Gwerthrynion and Buellt on his third son, Pascent. This is not history, but these traditions suggest a picture of Ambrosius as a Romanised ruler in south-western Britain opposed to Vortigern, and engaged, as pictured by Gildas himself, in trying to maintain Roman authority against the invading Saxons.

If among the degenerate members of the family of Ambrosius Aurelianus there is one who was identical with any of the princes abused by Gildas, the most likely would be Aurelius Caninus. He is the only one of the five princes who has not been located with reasonable probability, and according to the order which Gildas appears to be following, his position would be in south-east Wales. The proper names increase the probability that they are identical. Neither Aurelius nor Aurelianus is a type of name found in early Welsh.

Some have sought to identify Ambrosius with King Arthur, who on various grounds is thought to have flourished in the late fifth century. The evidence for a historical Arthur is, one may say at the outset, at least highly unsatisfactory. He left no descendants, and we have no trustworthy genealogy of him, and no contemporary records whatever. Cumulatively the amount of tradition cannot be ignored. Professor Jackson concludes the most recent study of the subject: 'There may have been a supreme British commander of genius in the late fifth century who bore the Roman-derived name of Arthur.'19 But let us be candid: there is no reliable evidence that he held any sub-Roman office, despite the statement of Nennius (ch. 56) that he was dux bellorum; and no evidence whatever p47that he was the head of a mobile field force, or acted as comes Britanniarum; or owed his military success to Roman military tactics or horsemanship.

Neither Gildas nor Bede mentions him, and such evidence as we possess from later sources is not calculated to inspire confidence. His name occurs in the unsatisfactory Pembrokeshire genealogies, as well as in a later version by Geoffrey of Monmouth. He is mentioned in the Annales Cambriae for 516 as the successful leader of the British at Bellum Badonis — a statement supported by Nennius; and again in annal 537 where it is stated that Arthur and Medraut (the Modred of later tradition) fell at the battle of Camlann. These two notices are of unknown derivation.

Our longest account of Arthur is that of Nennius. He tells us in ch. 56 that when the Saxon leaders were gaining ground in Britain, Arthur — who has not been previously mentioned — was accustomed to fight against them along with the kings of the Britons. But the twelve victorious battles20 here credited to him are likely to have been of originally independent origin, for such lists were a common oral convention. The last battle mentioned, however, that of Badonicus Mons, has at least the support of the Annales Cambriae.

Much later tradition, represented by Geoffrey of Monmouth, connects Arthur with the royal line of the princes of Devon, naming one, Uther Pendragon, as his father; and poetical tradition perhaps lends some support to this.21 Geoffrey further represents him as son of Custennin (Constantine), prince of Devon, and brings him into connection with prince Gereint of the same kingdom, and this is in line with later literary and hagiographical tradition; but on the whole none of this late tradition can be accepted as valid as it stands. The most that one can say with confidence is that there was probably a Celtic chief called Arthur, who met his death in battle against the Saxons. It may perhaps be added that St Columba's p48biographer Adamnán, writing before 704, records an Arthur among the sons of St Columba's king, Aedán mac Gabráin, a king of Argyll, whose family has other Welsh connections; and Adamnán states that this Arthur was killed in battle in south-eastern Scotland against enemies who included the Miathi and Saxons. He is mentioned by Adamnán (I.IX) as first among the sons of Aedán who, had he lived, would have been his successor.

However we view the historicity of Arthur, a growing body of opinion now regards his traditions as originating in North Britain, and believes that this limited Cycle of Arthur moved southwards, along with a vast body of North British traditions, such as those of Myrddin (Merlin) and other heroes of the sixth and seventh centuries, to be freshly localised in Wales and Cornwall. A line in the North British corpus of poems known as the Gododdin (cf. pp103 ff. below), probably dating from c. 600, refers to Arthur as representing the highest standard of valour. While this may be a later interpolation, it is one of the most convincing pieces of evidence, if not for a historical Arthur, at least for a lively North British tradition of an Arthur who was at this early date the standard pattern of a great hero and valorous man. If we could get behind the massive superstructure which southern Celtic tradition of later times has erected round a prince of Britain who perished in battle against her enemies, the present writer is of the opinion that we might regard him as a member of the ruling dynasty of Argyll, which seems to have had strong British ties by intermarriage.

Separation from the Continent has given to the British Isles its personality, and a continuity which has survived invasions and occupations. These are the great incidents of our history, but they do not give a balanced picture of the part that the sea has played in the normal life of the country; and in recent p49years archaeology has completely revolutionised our ideas of the earliest Saxon settlements. The evidence of the Saxon cemeteries and the pottery which they contain suggests that the Saxons had perhaps been granted settlement rights, and had been brought into the country as foederati long before 400, and more than half a century before the traditional date of the Adventus; and that even at the end of the third century Saxon settlements here were en rapport with both their Continental neighbours and with the Romans in this country. Whatever the explanation of the precise nature of this early Saxon peaceful life in an unequivocally Roman context in eastern Britain, early Saxon communications were evidently taking place normally across the North Sea.22

In the fifth and sixth centuries Britain became the centre of a large-scale maritime activity in which she herself took an important share. The Irish Sea and the waterways to the north appear in our traditions as a busy intersection of sea-routes, the northward extremity of the Atlantic coastal route of prehistoric times. The Celtic peoples were essentially habitual seafarers, with simple but fully adequate sea-going vessels, known as curachs,23 scaphae, naves longae, etc.; and throughout the historical period the narrow and stormy seas between north-eastern Ireland and south-western Scotland were traversed constantly, while the Irish Sea, a great land-locked lake, has been, more than the land areas, the true unit of Celtic civilisation. We have seen that the Irish expansion on to the west coast of Wales had already been extensive in the Roman period, and from the fourth century onwards reached a pressure second only to that of the Teutonic peoples on the east and south of Britain, so that Irish immigrants of the fourth century A.D. may have been welcomed by friends and relatives.24 Tradition, inscriptions, archaeology and place-names alike place it beyond doubt that continuous Irish settlement was taking place in the Cornish peninsula, along the shores of the p50Severn Sea, and in South and West Wales to the Caernarvonshire peninsula,25 and that it had even made its influence felt in Brecon (cf. p69 below).

[image ALT: A drawing of two canoe-type fishing boats, each manned by three men; in each a single pair of unusually long oars can be seen. A seventh man stands on shore helping with a net. They are curachs, a type of Irish boat.]
Fig. 10.º Modern native Irish curachs, Blasket Island, Co. Kerry (after Mason)

Under these conditions it is in no way surprising that in the fifth and sixth centuries, and indeed doubtless even earlier, a migration from south-western Britain into Armorica, the westernmost peninsula of Roman Gaul, established a British- (Welsh- and Cornish-) speaking population overseas, the nucleus of modern Brittany, and our first colony.26 The reason is generally believed to have been the westward progress of the Saxons, chiefly on the authority of Gildas, who in ch. 25 refers to his countryman being forced to leave their native land; and though he never mentions the name there can be no doubt that he has Armorica in mind. But the Irish encroachments on the west must have offered a far stronger motive for the colonists to leave home than the attacks of the Saxons on the east. The majority of the immigrants appear on linguistic evidence to have come from Cornwall, while literary tradition hints that the leaders came from Wales, far from the areas as yet affected by hostile Saxons. Moreover, the contacts between Armorica and Britain were not new. Caesar tells us that the Veneti, the most powerful of the Armorican tribes, were perfectly familiar with the sea-routes, having a large fleet with which they were accustomed to sail to Britain.

The Britons seem to have moved into the north and west of Armorica with little opposition — on the whole a peaceful penetration. Consistent traditions suggest that the true pioneers of the migration were British ecclesiastics, who negotiated the establishment and rights with the legal authorities. They no doubt attended to the religious needs of the new colonies, but there is no suggestion of missionary activities. Indeed, the Armoricans were members of the Roman Church in Gaul. The ecclesiastics seem to have been generally members of the princely families of east Wales, Cardiganshire, and the Severn p51Valley, and perhaps of Brecon. In some traditions princely leaders accompanied and supported the priests, and then became the rulers of the new colony.

An interesting glimpse of the seafaring activities of the Celtic people of this period is a little Celtic settlement and the establishment of a Celtic Church and monastery in Galicia. The early history of this settlement in north-west Spain is unknown. The area is an ancient Celtic one, in which the Teutonic tribe of the Severi had established themselves early in the Migration Period. Even after its destruction by the Arian Visigoths in 585 it retained much of its independence, and as it was a Catholic state its monasticism remained conservative. In particular the Celtic monastery of Santa Maria de Bretoña near Mondonhedo, included in the episcopate of Britonia in a list dating from Suevic times, must have been a link forged with the Celtic world to the north. The Celtic signature of their first bishop is attested as Mailoc. Was this little church and monastery a new creation, in response to a stimulus from Celtic lands in the Age of the Saints? Was it an Irish or a British one, founded direct, or from Brittany itself? None of Mailoc's successors can be clearly identified by their names as Celtic. The problem remains. It may be added, however, that Orosius speaks already in the early fifth century of a city which he calls Brigantia, and which he regards as having some kind of relations with Ireland.27 º

The Author's Notes:

1 Wheeler, Lydney.

2 As Professor Richmond reminds us R. B., 64, 140.

3 Richmond, R. B., 192.

4 Frere, A. J. XL (1960), 20.

5 Caspari, Briefe, 15.

6 See Dio Cassius, LV.XIII; and cf. Collingwood and Myres, 172; Rivet, T. C. R. B., 136.

7 Nash-Williams, E. C. I. W., p14.

8 Jackson, Antiquity XXIX (1955); cf. H. M. Chadwick, E. S., ch. X; cf. also Hunter Blair, O. N., 36.

9 I have discussed this term in B. B. C. S. XIX (1961), 225.

10 Jackson, loc. cit.

11 See the study of the Votadini by Hogg in Grimes, A. A. B., 100.

12 The scarlet tunic was one of the insignia of Roman military officers, including those of the federate states.

13 For their early history see O'Rahilly, E. I. H. M., 64; Jackson, L. H. E. B., 155. Two versions of the Irish saga are extant, both dating from the eighth century, and both contained in manuscripts in the Bodleian Library at Oxford. The version contained in the Oxford MS. known as Rawlinson B 502 has been edited and translated into English by the late Kuno Meyer in Y. Cymmrodor XIII (1900), and later in Ériu III (1907), 135. See more recently Spender, in Essays and Studies presented to Professor Tadg ua Donnchada (Féilscribhinn Torna (Cork, 1947), 211).

14 For a discussion of the possible relationship between the terms gwledic and protector, see E. K. Chambers, Arthur of Britain (London, 1927), 176; cf. Stevens, É. C. III (1938), 86 ff.; and my note in B. B. C. S. XXIX (1961), 225.

15 The chief early references are cited in a recent study of Vortigern by Radford, Antiquity XXXII (1958), 19; cf. further H. M. Chadwick, The Origin of the English Nation (Cambridge, 1907), 37, and later ibid., S. E. B. H., 21.

16 Nash-Williams, E. C. I. W., no. 182, Plates XXXV, XXXVI.

17 Bu'lock, Antiquity XXXIV (1960), 49.

18 The name has been associated by antiquarian speculation or tradition with Dinas Emrys ('the citadel of Ambrosius'), in Snowdonia, at least from the twelfth century. Dr Savory's recent excavations have proved that the site was occupied in the late Roman and sub-Roman periods, probably in the late fourth and fifth centuries A.D. and again in the sixth to eighth centuries A.D. See his important article, in A. C. IX (1960), 13.

19 Jackson, 'The Arthur of History', and 'Arthur in Early Welsh Verse', in R. S. Loomis, A. L., 1, 12; also Thomas Jones, B. B. C. S. XVII (1960), 235.

20 See Crawford, Antiquity IX (1935), 289; Jackson, M. P. XLIII (1945). Count Nikolai Tolstoy, in B. B. C. S. XIX (1961), 118.

21 For valuable notes on the literary evidence relating to Arthur, see Bromwich, Triads, 358, 521.

22 Collingwood and Myres, R. B. E. S.; Clarke, E. A., 129; Dauncey, Antiquity XVI (1942); Myres, ibid., and in Harden, D. A. B., 37; for a full recent summary, Myres, C. B. A., Report no. 11 (1961), 40. Cf. Richmond, R. B., 65; for coins, see Sutherland, in Harden, D. A. B., 3; Allen, in Frere, U. L. I. A.

23 For a study of the curach, see Marcus, S. G. S. VII (1953), 106.

24 C. O'Rahilly, I. W., ch. 2; Wheeler, P. R. W., 234, 292; Stevens, Mélanges XLII, 671.

25 See Wrenn, T. H. S. C. (1959), 38; Melville Richards, J. R. S. A. I. XC (1960), 145.

26 J. Loth, L'Émigration.

27 Full references are given by the present writer in The Age of the Saints in the Early Celtic Church (Oxford, 1961), 58.

[image ALT: Valid HTML 4.01.]

Page updated: 24 Nov 12