Celtic Kingdoms of the British Isles

Dobunni

The territory of the Celtic Dobunni lay to the west of the Catuvellauni. To the north they were bordered by the Cornovii, to the west by the Silures, to the south by the Durotriges, and to the south-east by the Atrebates and Belgae. Their territory initially comprised northern Wiltshire and southern Gloucestershire, the locations of the earliest coin distributions. The tribe later expanded into western Oxfordshire, northern Gloucestershire, north-eastern Somerset, Avon, parts of Hereford and Worcester, and also South Warwickshire. They were a non-Belgic people who were organised around an impressive series of hill forts, mostly overlooking the Avon Gorge, but who were showing considerable signs of Belgic influence. Generally the people lived in small villages and farmed the fertile land.

It has been suggested that they were little more than a division of the Atrebates, and only gained independence during the reign of the pro-Roman Tincommius in the late first century BC. If this was the case then the Dobunni were probably an earlier people who had been pushed out of their territory by the arrival of the Atrebates, and subjugated along the way. The tribe's name is obscure, but a possible explanation also suggests an identity problem, either two tribes that formed a minor confederation or perhaps two halves of a single tribe. In the Dobunni name, 'do-' may be equivalent to 'du-' and would mean 'two', while 'bun' appears to be proto-Celtic for 'origin'. So, taking a rather large leap, the name might just indicate a tribe that was formed from two smaller ones.

By the first century (circa 35 BC) the Dobunni borders abutted those of the Atrebates, and the coinage of both tribes seems to be found in parallel, although the Dobunni ejected the Romanised Atrebatean coins which appeared from 30/20 BC. By AD 20/25, the Dobunni also found themselves bordering the powerful Catuvellauni, although they seem to have been on friendly terms with this powerful tribe.

(Additional information by Edward Dawson, and from The Oxford History of England: Roman Britain, Peter Salway, and Glevum - The Roman Origins of Gloucester, Nigel Spry.)

1st century BC

Throughout the course of the century there is evidence in the region of a new wave of settlements. Some earlier hill forts appear to fall out of use while others continue to be occupied. Some new settlements are enclosed, replacing older, open settlements, while other settlements remain unaltered. The pattern of change is uneven and occurs at different times across the century. The building of the grand enclosure settlement of The Bowsings which replaces The Park open settlement occurs at the start of the century, while the Duntisbourne enclosed settlement appears to be created towards the end of the century. The inference is of the slow in-filtering of a new people, probably Belgic people from the Atrebates or Belgae regions to the east.

c.5 BC - AD 7

If the Dobunni are indeed vassals of the Atrebates, or a constituent part of the tribe as is sometimes suggested, then it is in this period that they declare their independence. Atrebatean nobles, angered by the pro-Roman stance of Tincommius in direct opposition to the policy of his father and grandfather, seem either found the tribe of the Dobunni from an earlier subjugated people or liberate the westernmost Atrebateans (or possibly a combination of both).

These silver coins were issued by the Dobunni in the first half of the first century AD

Coinage exists in this period, which is issued from several sites up until the Roman invasion. There is no certainty that all (or any) of the issuers are overall kings of the Dobunni tribe. Instead, the tribe may regularly enjoy dual kingship, and perhaps even that level of unity may be beyond them. There is evidence of coinage being issued from Bagendon, Ditches, and possibly Salmonsbury during this period.

fl c.10 - c.30

Anted-

Name found on coinage only. King of north & south Dobunni.

fl c.30

Eisu-

Son? Name found on coinage only. King of north & south Dobunni.

Inam- / Inara-

Name found on coinage only. King of north & south Dobunni.

c.35 - 40

The Dobunni appear to fracture into northern and southern divisions, or else they are simply returning to the order that may have existed before unification under Anted.

Catti-

Name found on coinage only. King of the north.

Comux-

Name found on coinage only. King of the south.

43

By now the tribe has certainly divided in two. The north-eastern part, stretching from the southern side of the Stroud Valley to north-eastern Gloucestershire and western Oxfordshire, is issuing Romanised coinage. In Avon and southern Gloucestershire, the remnant is issuing coins of a native type. There are also two distinctive pottery styles in use which show a north-south split. The suggestion is that the Catuvellauni have gained some sort of control over the north-eastern section of the Dobunni by this stage. This would certainly be in line with their policy of subjugating neighbouring kingdoms.

fl 43

Bodvoc- / Boduoc- / Bodvoccus?

King of the north-eastern pro-Roman Dobunni.

fl 43

Corio-

King of the south-western native Dobunni.

43

With the defeat of the Catuvellauni by Roman Governor Aulus Plautius and Emperor Claudius, part of the Dobunni appear to surrender themselves to Plautius by means of envoys. It seems likely that the surrender is made by Bodvoc in defiance of his Catuvellaunian overlords. It cannot be long before the rest of the Dobunni are conquered, and within a year a fort has been established at Corinium (modern Cirencester).

43 - 47?

There is reason to believe that the ex-Catuvellaunian king, and High King, Caratacus, shelters with the remaining anti-Roman section of the Dobunni. By AD 47 the area is almost certainly included in the occupied or supervised Roman territory so perhaps this forces him to join the Silures. The collective Dobunni remain a Roman client tribe until about AD 97.

c.49

The earliest phase of building in the city of Glevum (Gloucester) is begun. A Roman legionary fort is founded at what is now Kingsholm, close to an Iron Age settlement, overlooking the River Severn (although its course has since changed). The fort probably houses the Twentieth Legion (although a sizable body of opinion prefers the Second Legion). It is located in the (possibly friendlier) northern region of Dobunni territory, commanding the mouth of the River Severn (Sabrina Fluvius) near the Fosse Way, and is probably close to the southern borders of Cornovii territory.

61

During the Iceni-led revolt in the east, the Silures, Ordovices, Dobunni, and perhaps the Durotriges are probably pinned down by the Roman Second Legion and are unable to join Boudicca. The presence of the legion, under Poenius Postumus, is perhaps due more to fortune than planning. When Governor Suetonius marches back from Wales to reassemble the scattered Roman forces at a location in the Midlands, Postumus refuses to move. Possibly he is influenced by memories of the death of the praefectus castrorum at the hands of the Silures during the governorship of Ostorius. When he hears of Suetonius' victory against Boudicca, Postumus kills himself and his legion joins the governor in the field.

c.66

The Kingsholm fortress is prone to flooding so a new and larger fortress is built on the higher ground one kilometre to the south, at what becomes Gloucester Cross. It is around this fort that a civilian settlement grows up, forming the early city. Troops are based here in the build up to the invasion of Wales, with the first strike being against the Silures and Demetae. However, this is apparently delayed by the events of AD 69, the 'Year of Four Emperors'.

mid-70s

The Roman fort at Corinium (modern Cirencester) is vacated by legionaries who are probably off to fight the Silures and Demetae. In their place, the Dobunni tribal civitas of Corinium Dobunnorum is founded, perhaps as a replacement for the tribe's possible client status. The Ala Gallorum Indiana auxiliary cavalry unit is (or remains) based here, as evidenced by a tombstone dated to this period on stylistic grounds. Over the next couple of decades the city's street grid is laid out and slowly filled, mostly by stone public buildings, private dwellings and shops. The forum and basilica are the largest in Britain apart from those on Londinium.

87

Following its campaigns in Wales and then in Pictland with Agricola, the Twentieth Legion returns to Glevum. The fortress undergoes major changes, with wood being replaced with stone, and new buildings with outer stone walls and timber-framed internal partitions. The general defences are also greatly enhanced, making them amongst the most impressive in Britain to date.

96 - 98

Roman Emperor Nerva designates the fort and settlement of Glevum a colonia - Colonia Nervia Glevensium - a self-governing citizen colony regulated by a council and four magistrates. This certainly ends whatever client tribal status the Dobunni may have experienced since the Conquest. Legionaries settle there as a kind of demobilised reserve, officially retired but liable to be recalled as an auxiliary force should trouble arise in the sometimes fractious western country. A basilica which houses the administrative body, and a forum which acts as a market place are also added. Within fifteen years, civilian buildings replace the original fortress.

2nd century

The region is blessed with particularly good soil for Roman farming methods, and there is a significantly high rate of villa-building. The Chedworth and Woodchester villas are amongst their number, both of which possess high quality mosaics. Chedworth is one of the largest villas in the entire country, built facing east (which is unusual), in a sheltered spot overlooking the River Coln. Glevum itself is at its height, and has reached its largest extent. Both it and Corinium are fortified towards the end of the century.

165 - 180

Plague enters Rome from the east, brought back by returning legionaries. It quickly spreads throughout the empire and is generally known as the Antonine Plague. When it arrives in Britain it strikes hard. In 2004, archaeologists uncover the remains of ninety-one men, women and children dumped haphazardly into a mass grave at Glevum. The bones are dated to this period, and are unusual as the Romans are typically very careful about interring their dead. The situation must be fairly dire.

The Antonine Plague that killed these people is thought to have been smallpox

3rd century

Archaeologists later unearth a tombstone dating to this century which bears an inscription of someone who has served in the Twentieth Legion. Although the legion itself had supposedly left Glevum by the end of the first century AD, there seem to be a remaining link, perhaps due to Glevum being created a colonia for the legion in AD 97.

c.300 - 306

Around the very start of the fourth century, changes take place at Glevum. The second century wall is replaced by one that is stronger and higher, and with deeper foundations in places. Similar refortification takes place at Caerwent (in the Ewyas district) and Caerleon (in the Cernyw district), as preparations to face a possible threat from the River Severn. The threat is probably presented by a sudden increase in Scotti raids from Ireland, but whether the defences are ever put to the test or not is unknown. Perhaps linked to this threat, and others, in 305-306, the Diocese of the Britains is sub-divided into four provinces. Glevum probably serves as the capital of Britannia Prima, and perhaps even has its own mint.

c.325

Around this date the beautiful Orpheus mosaic is laid down in Woodchester villa, not far from Glevum, which remains inhabited until the end of the century.

mid-5th century

The region which had belonged to the Dobunni, with extra territory to the north and south apparently now included, appears to emerge as a new territory in its own right. Caer Gloui is unnamed by any contemporary accounts, but the city which bears that name (modern Gloucester) would be a highly likely candidate to be a regional capital. It, rather than the Dobunni civitas of Corinium Dobunnorum (modern Cirencester), has a commanding position overlooking the Severn, while the stronghold of Amesbury further to the east of Corinium guards that side of the territory.

Caer Gloui

The Romano-British city of Caer Gloui had been founded by the Romans as Glevum (modern Gloucester). It was first settled around AD 49 as a legionary fort, and a city grew up around it. During the fourth century it probably served as the capital of the province of Britannia Prima within the Diocese of the Britains, and it seems to have retained its importance into the fifth century. While the later name of Caer Gloui is used here, the name of the territory itself has not survived, and in the fifth century the city may instead have been known as Glouvia.

Central administration of Britain appears to have broken down in the mid-fifth century, to the extent that the regions began to establish partially or wholly independent districts or kingdoms. The administration at Caer Gloui found itself in command of much of the land around the mouth of the Severn, which also encompassed the cities of Caer Baddan and Caer Ceri to the south, essentially making it a successor to the former Dobunni tribal territory. In the sixth century, the Romanised district evolved into a kingdom, and its fall is noted in the Anglo-Saxon Chronicle. The incoming Hwicce who took control of the area maintained the city's name, mispronouncing it in their very individual Teutonic language, so that it survives today. Romano-British Glou became Saxon gloe added to -cester from the Latin castrum (fort), emerging as Gleawanceaster (Gloucester).

Just about all the (extremely scanty) information we have about the post-Roman city and the events of the sixth century comes from Gildas' De Excidio Britannia, Nennius' Historia Brittonum, and Geoffrey of Monmouth's Historia Regum Britanniae, plus a little from Bede. While Gildas and Bede can be said to be reliable, Nennius seems to be less so, and Geoffrey is prone to wild flights of fancy while still retaining a distinct foothold in events that must have been recorded by sources earlier than him.

Vortigern has a claim on the region as a power base until his fall in the mid-fifth century. After that, it seems highly likely that Caer Gloui was one of the centres of operations for Ambrosius Aurelianus during his battles against the Saxons in the south. It is possible that his father also called this territory home. Descendants of the two men seem to have based their claim on the later kingdom on this, if Ambrosius himself didn't specify their continued rule in the city, which is a possibility.

(Additional information by Edward Dawson.)

c.410 - c.446

Aurelius Ambrosius (the Elder)

Roman senator and pro-Roman. Killed by plague.

c.410 - 418

Aurelius Ambrosius is the official representative of Roman Emperor Honorius to the British provincial council and is claimed as a prince, marking him out as a member of the Romano-British nobility. The country is reorganising at this time, following the expulsion of Roman administration, but links with Rome are clearly being maintained.

c.418 - 425

This period is said to witness the increasing influence and power of Vortigern of the Pagenses, culminating in his high-kingship. The provincial council decides, and external factors dictate, the need for strong, central, leadership in the country, and the impression is that Aurelius Ambrosius is not strong enough to offer a viable alternative. It seems that he and Vortigern form the figureheads for opposing parties, but for the moment it is the latter who has dominance.

c.432 - 436

Aurelius Ambrosius is apparently a leader of a British council, which presumably answers to Vortigern. It is his decision to confirm the Irish Deisi as commanders of the Demetia area of the west coast to counter the threat of Irish raiders. Vortigern acquiesces and assigns Ambrosius 'Dinas Emrys and all the western lands', suggesting that Ambrosius becomes the architect for the defence of these western areas. This is motivated by the council's reluctance to depend entirely on Saxon mercenaries, with their constant demands for increased provisions, especially in an area were they would be lightly supervised. The Deisi have already been settled for some time and would be self-supporting.

With discord building in the country between about 420-450, many Romano-Britons left in a hurry, burying their wealth in the hope that they could return in better times to collect it

c.437/438

According to Gildas and Nennius when referring to either Aurelius Ambrosius or his son, this family represents the Romanised nobility in Britain. It is possible that by this time, as elsewhere, a magistrate is in charge of the governance of Caer Gloui (and seemingly Caer Baddan and Caer Ceri too, given that the three cities are closely linked). Given the later role of Ambrosius Aurelianus in this region, it seems entirely possible (although hypothetical), that his father now fills this position.

Ambrosius (the Elder) has long been Vortigern's main rival, with it seeming likely that they not only head two opposing factions in the country, but also opposing ideologies, with Ambrosius retaining his Romanised, Roman Church background while Vortigern is leader of the Pelasgian pro-Celtic party. Around this year, internecine warfare breaks out between the two rival factions, resulting in the Battle of Guolloppum (Cat Guolph, Wallop in Hampshire). The result is uncertain, but it is probably followed by a period of civil strife in eastern and southern Britain.

c.440 - 443

In the early 440s the Saxon foederati and laeti revolt, causing widespread chaos and temporarily controlling swathes of the country. Soon after this, the defences of both Caer Baddan and Caer Ceri are repaired. In the latter, flood prevention work is carried out on the Verulamium gate.

446

Serious plague hits southern Britain and unburied bodies are to be found in the streets of Caer Ceri. The town contracts to some wooden huts inside the amphitheatre.

It is this point at which Ambrosius the Elder, who must be an old man in his sixties, also dies, "in these same broils", ie. the Saxon revolt, although according to tradition it is the plague which actually claims him. Ambrosius' surviving family is in hiding by now (traditionally in Armorica), avoiding the vengeful clutches of Vortigern. An archaeological excavation at a site in modern Gloucester produced an early fifth-century secondary burial in a Roman funerary building with indications that the man had been of high rank. Could this have been Ambrosius the Elder?

446 - 455

At the same time as the Anglo-Saxon mercenaries in the east revolt (in 455), the entrance to Caer Gloui's amphitheatre is reduced in size to make it easier to defend, and life continues, as evidenced by fifth and sixth century pottery.

There is a gap in what can be pieced together of the story at this point, with Ambrosius Aurelianus, the son of Ambrosius the Elder, not emerging into British affairs until perhaps 455 or 460. Given the period at which he appears to be at his most influential, his date of birth is probably around 430, making him too young to succeed his father immediately as a possible magistrate of Caer Gloui, so it seems likely that someone else, a possible deputy or one of Vortigern's supporters, assumes command.

c.446 - c.455?

?

Name unknown, possibly a deputy of Aurelius Ambrosius.

c.455 - c.480?

Ambrosius Aurelianus

Son of Aurelius Ambrosius. Magistrate? High King?

c.455 - c.496

Ambrosius (and perhaps the elusive Arthur after him) seems to base himself in the territory of Caer Gloui. Amesbury (which in a Saxon charter of about 880 is spelled Ambresbyrig, 'the stronghold of Ambrosius'), located on the territory's eastern borders, is perfectly suited to be the focus of Ambrosius' military operations. He probably governs the territory as a Roman magistrate rather than as a princeps or king (although he is claimed as the first king by later chroniclers).

It seems likely that the Wansdyke is constructed around this time, possibly in response to further Saxon incursions to the east. Groups from the Thames Valley appear to force their way into the western end of neighbouring Cynwidion while further groups from the Middil Engle push through the Vale of Aylesbury to complete the encirclement of that kingdom, exposing Caer Ceri's eastern border in the process. There is the possibility that during this period Glevum's residents leave in some numbers to head to Cernyw, on the other side of the Severn, although the connection is tenuous apart from the change in that kingdom's name to Glywyssing around 470-480.

All building and repair work on major new defensive works probably comes to an end with the British victory of Mons Badonicus around 496, with the siege possibly being fought outside Caer Baddan.

?

Name unknown.

c.500

The city shows modest re-growth now that peace has been won, and later archaeology shows that a new north gate is created in the city's walls at the beginning of the sixth century. The old gate is now ruined and blocked. However, by this time, the old city is in a very run-down state, and new building work is only in wood. The focus of settlement seems to be nearer the river, away from the Roman city which has suffered from assault and plague during the previous century.

?

Name unknown.

c.540

The three cities, Caer Gloui, Caer Baddan and Caer Ceri, still apparently form a single kingdom (called Guenet by Nennius). This seems to be partially borne out when Gildas infers that Aurelius Caninus is ruling his kingdom as a single political entity instead of one of three minor states.

fl 540

Aurelius Caninus

High King. Named by Gildas. Still ruled the united three cities?

c.550

Around this time, either upon the death of Aurelius or his successor, the unnamed fifth king, the single kingdom based at Caer Gloui divides into Caer Baddan, Caer Ceri and Caer Gloui. This is probably a result of the kingdom being divided between sons, an act that is based on traditional Celtic practise. The act suggests that a true kingship is being practised by this time, rather than the previous Romanised role of magistrate.

Despite the focus of settlement now being away from the old fort, Glevum's Roman walls were still very much in use in the sixth century

?

Name unknown.

c.570s

The name of the last king is rather remarkable in that it breaks down as 'Con' meaning 'dog' and 'mail/fael' meaning servant. Speculatively speaking, this 'dog servant' may have links to the kings of Glastenning to the south. The king there, Cyndrwyn Glas, is also king of Dogfeilion, which name means 'servant of [the god] Dagda'.

However, a much more likely reason is the sense of humour sometimes exhibited by the Welsh (even today) in naming their offspring. Conmail's grandfather, Aurelius Caninus, may well have been alive at the time of his birth. What better jest than to poke gentle fun at the mostly pagan naming convention of 'Cuno-' ('cyn') added to this or that god than by naming someone just plain 'dog' (Aurelius Caninus, the latter being the Latin for dog), and then in his grandson combining it with the mostly Christian convention of mal/mail/mael?

? - 577

Conmail / Cynfael

Killed fighting the West Seaxe.

577

Caer Gloui, together with Caer Baddan and Caer Ceri, falls to the West Seaxe following the Battle of Deorham or Dyrham (an event which is rather obtusely doubted by some but which would be entirely in keeping with the pattern of Saxon advance to the west). With this collapse, the territory of Caer Celemion to the east is now totally isolated, and Dumnonia is cut off from any overland contact with other surviving British territories. Gwent and Pengwern now form the western frontier against further Saxon advances.

The Hwicce take over the territory and eventually push its borders north into Worcestershire, at the expense of Pengwern. However, rather than simply sweep away all that is British, they appear to form a new top layer of aristocracy over a largely British population that retains much of what it had before, possibly even down to its church organisation.