by Mike Ashley
1. The first battle was at the mouth of the river which is called Glein The name Glein is derived from the Celtic glan, meaning “pure” or “clean.” No river is called Glein today, but two are called Glen, in Lincolnshire and Northumberland. The Lincolnshire Glen flows through the Fens and today joins the River Welland near Spalding, but in the fifth century Spalding was on a hard ridge of land virtually on the shoreline of the Wash, which then reached further inland. Interestingly, the origin of the name Welland is uncertain, but it is also a Celtic word and could mean “good” or “holy” stream, thus the names Glen and Welland may be connected and the mouth of the Glen may, at one time, have been at the mouth of the Welland. This is a possible site, because the area to the north, in Lindsey, was one of the first to be settled by the Angles. The Fens do not lend themselves to major battles but, as Hereward the Wake proved five centuries later, they are suited to a covert guerrilla operation in territory which would be known by the British but highly dangerous to the unwary invader. There is, however, no significant base nearby from which Arthur could have launched his attack. There may be another appropriate site in Lincolnshire at Brigg, originally Glanford Bridge, which I discussed in the last chapter. The river, now called the Ancholme, was a major estuary, before drainage works reduced much of the surrounding marsh.
The Glen in Northumbria also flows into another river, meeting the Till near Doddington. This confluence is close to Yea-vering Bell, the largest Iron Age hill fort in Northumberland. It was a significant site of over a hundred dwellings, and a major archaeological dig in 1960 showed that it had been reoccupied after the Roman period. Yeavering has an unspoilt view down to Bamburgh and Lindisfarne and would have been a major defensive site against the early Germanic invaders in the fifth century. The site was of such importance that after the conquest of the area by the Angles, Edwin of Northumbria established his own palace here at the foot of the hill. If Nennius’s list is in chronological order, Yeavering Bell is also a suitable location for the first conflict. However, considering how important this site would have been to the British and Saxons, it is surprising the battle list refers to the river and not to the fort.
There are other suggestions. In 1867, in Chronicles of the Picts and Scots, W.F. Skene suggested Glen Water, a small stream running down to the river Irvine at Darvel in Strathclyde, near Kilmarnock. His suggestion was based on local legend, which even supplies a date for the battle, 542AD. However, since that was the year that Geoffrey of Monmouth said Arthur of Badon died, it would seem an unlikely date for the first of his battles.
Josephus Stevenson suggested, in notes to his 1838 translation of Nennius, either the River Lune, in Westmorland, or the Leven in Cumberland. The Lune is of special interest. The name is derived from glein, likewise meaning “pure and healthy”. The river has a major estuary at Lancaster where there was a small Roman fort, strengthened in the 340s as a coastal defence against the Irish. Of all the glein rivers it is the only one with a mouth to the sea and a significant fortification.
Other suggestions include the River Glyme in Oxfordshire, Glynch Brook near Bewdley, and Gleiniant near Llanidloes, in west Wales. The Glyme is an intriguing possibility. The name means “bright one”, so is not immediately related to glein, but its confluence, where it joins the River Dorn at Wootton, north of Oxford, is at the southern end of the little known British enclave of Calchvynydd, which ran up through the Chilterns between Oxford and Northampton. Nearby is Ambrosden, a town which is suggestive of Ambrosius Aurelianus, and a likely spot for one of his battles.
Gleiniant has the distinction of retaining the name glein. The stream at Gleiniant meets the Trannon at Trefeglwys in present day Powys, close to the old borders with Gwynedd and Ceredigion. It is also close to one of the suggested sites for Camlann. Gleiniant would suit an internal struggle, but is far out of the conflict zone for the Saxons.
One final possibility is the Glynde Reach in Sussex, one I’m not aware has previously been suggested. This small stream was originally the Glynde Bourne – indeed it flows right below the famous Glyndebourne Opera House – and Glynde is derived, according to some etymologies, from the Celtic for valley, glen. Others say it comes from the Saxon glind, for enclosure. Either way it has a striking similarity to the first in Nennius’s list, made all the more intriguing as this could be the site of one of Aelle’s battles listed in the ASC as happening at Mearcrædes burnam in 485, exactly when I have suggested that Arthur’s battle campaign may have started.
Of all the suggestions the best possibilities are the Northumbrian Glein, the Cumbrian Lune and the Sussex Glynde.
2–5. The second, third, fourth, and fifth battles were above another river which is called Dubglas and is in the region of Linnuis.
Dubglas is the original of the name Douglas. It is usually translated as meaning “black water”, but a more strict interpretation is “black-blue” or even “black-green” (dub+glas). Glas means that blue-green colour seen in glass – the name Glasgow means “green hollow”. So we’re really looking for a dark, probably deep, river that reflected blue-black, or green-black. Unfortunately, that could apply to many, not helped by the fact that the name Douglas survives as one English river, two Scottish rivers and twelve called Dulas in Wales. Doubtless there would have been plenty more called Dulas in England, which changed their name under Saxon domination to such variants as Dawlish or even Blackwater (there’s a tempting site in Hampshire that I discuss separately on page 162). It was probably because of this abundance of names that the original chronicler qualified the description by adding that it was in the region of Linnuis.
In 1945, Kenneth Jackson, in an article in Modern Philology, determined that Linnuis derives from Lindenses, meaning “the people of Lindum”, or Lincoln. This at first seems promising, because we know that the area around Lincoln, which was Lindsey, was one of the earliest areas settled by the Angles. Unfortunately, no river in that area has a name remotely like Douglas. The primary river is the Witham, and some have suggested that the Witham might originally have been called the Douglas, on the assumption that Witham is a Saxon name, derived from “Witta’s ham.” However, Kenneth Cameron, in English Place Names, states that the Witham is probably one of a group of rivers the names of which go back before Celtic times into unrecorded history, so it was probably never known as the Dubglas.
There is another candidate for Linnuis. The Roman geographer Ptolemy used that word to describe the area now known as Lennox, covering the territory north of the Clyde and Firth around Loch Lomond. Just east of Loch Lomond is Glen Douglas, where the Douglas Water gushes down through the glen to enter the loch at Inverbeg. Beyond, across Loch Long, but still clearly visible from Glen Douglas, is the strangely shaped peak of Ben Arthur, which may well be connected with the Dál Riatan king’s son, Artúir mac Aedan. The old road from the Dál Riatan capital at Dunadd, in Argyll, skirts the southern fells of Ben Arthur before descending into Glen Douglas. There could certainly have been a battle here involving Artúir mac Aedan, probably against the Picts. Otherwise it is far too distant for a battle of a southern or even a northern British Arthur against the Saxons.
Leslie Alcock has suggested that Linnuis may have been copied wrongly and that the original word was Lininuis, which would have derived from the peoples known as the Lindinienses, who lived in Dorset and parts of Wiltshire, Somerset and Hampshire, the area that later became Wessex. The Roman name for Ilchester was Lindinis. Here the river Divelish runs from Bulbarrow Hill at Woolland, to Sturminster Newton in Dorset. Just south of Bulbarrow Hill is the Devil’s Brook, running south to Burleston where it enters the River Piddle. En route it passes through Dewlish, a village which also means “dark stream”. Although these two watercourses are minor, they do form a north-south barrier. Bulbarrow Hill is the site of a Celtic hill fort, and the rivers run through a triangle formed by Cadbury Castle, the Badbury Rings and the Cerne Giant, all significant Celtic landmarks. This could certainly be a location for a confrontation b
etween the British and the West Saxons.
August Hunt on the Vortigern Studies website draws attention to the Devil’s Water, a stream in Northumberland that passes through Linnel Wood and joins the River Tyne near Corbridge, at Dilston. Linnel is probably derived from llyn-elin (“lake-elbow”). It is an interesting combination of the two names in an area that would have been rich for conflict during the fifth and sixth centuries.
Of the many Dulas rivers in Wales, Steve Blake and Scott Lloyd in Pendragon suggest the Dulas that flows into Liverpool Bay at Llandulas, just east of Colwyn Bay. Another Dulas worth noting is now called Dulas Brook, and runs parallel to the Golden Valley in northern Ergyng, eventually joining the River Dore at Ewyas Harold. This is in the same area as part of Arthur’s hunt of the giant boar Twrch Trwyth, as told in the story of Culhwch and Olwen (see Chapter 8). That story, as we shall see, may well represent a series of battles conducted by Arthur across southern Wales and it is possible that at least some of the battles in Nennius’s list equate to it. We shall encounter another later.
One other Dulas, or Dulais, worthy of note flows through Pontarddulais in Glamorgan, where it joins the River Neath. Near its source it flows through Cwm Dulais, above which is Craig y Bedw, or Bedwyr’s Crag. This area was known for its groves and bushes. The Welsh for grove is llwyn, and there are places called Llwyngwenno, Llwynadam, Llwyn-y-domen, and so on. The area might have been known locally as the land of groves, or Llwyni, which might have evolved into Linnuis.
6. The sixth battle was above the river which is called Bassas.
This is one of the more baffling locations and even the most dedicated Arthurians have declared it impossible to identify. The more intrepid have suggested sites as far afield as Bass Rock off North Berwick in the Firth of Forth, and the River Loddon in Old Basing in Hampshire, suggesting that the Loddon was once known as the river of Basa’s people. The etymology for Basing is Saxon, and though this makes it unlikely to appear in what was originally a Celtic battle song, it may refer to an area so long occupied by the Saxons that their name had superseded the original.
The same problem affects Basford, the name of three places in Cheshire, Staffordshire and Nottinghamshire. All seem to be derived from the Angle Basa’s ford. Blake and Lloyd’s suggestion, Basingwerk, in Shropshire, is also of Saxon origin (“Basa’s stronghold”), and its Celtic name was Maesglas (“Green field”). Equally frustrating is Bassingbourne in Cambridgeshire, for although of Angle origin, it does at least mean “Bassa’s stream”. It is, however, in an area long occupied by the Saxons.
In the 1860s, Skene suggested Dunipace, the site of two hillocks at Falkirk in Scotland, near the Roman fort of Camelon. He proposed that the name was originally Duni-Bass, meaning “two mounds.” However the origins of Dunipace are not clear, with suggestions that it came from Dun-y-pax (“hills of peace”) or duin-na-bais (“hills of death”). John Stuart Glennie, writing in Arthurian Localities, whilst recognising this as a possibility felt there was an even better site across the river where a huge rock precipice may be the bass (or rock). Neither of the rivers in the area (the Bonny and the Carron) is called Bassas, but there is a ford across the Carron, and the Celtic name for ford, or shallows, is bais. But all this seems to be clutching at straws. The same concept of bais for shallows would work even better at the Fords of Frew on the Tribruit, discussed under battle 10.
The most likely suggestion is Baschurch in Shropshire, put forward by Graham Phillips and Martin Keatman in King Arthur, The True Story. The name derives from the churches of Bassa, mentioned in a poem by Taliesin as the burial place of the kings of Powys. Situated near the Welsh Marches, Baschurch could well have been the site of forays by the West Saxons. It is within a day’s march of the Gewisse territory to the south, and close to the site for Badon (Caer Faddon) given in the Mabinogion story The Dream of Rhonabwy (see Chapter 8). The nearest river is the Perry, but this name may be of Norman origin, derived from the Peveril family who controlled the area after the Norman conquest. We do not know the original Celtic name.
One other possibility concerns the Roman cognomen Bassus. One of the consuls at the time of Julius Caesar was Ventidius Bassus, and there were two noted poets at the time of Nero, Caesius Bassus and Saleius Bassus. Several inscriptions have been found in Britain, mostly in the north, bearing the name Bassus. One at Black Carts, halfway along Hadrian’s Wall, notes that part of the wall here was built by Nas . . . Ba[ssus] of the First Cohort. Most significantly, at the fort of Alavana at Kendal in Cumbria, there is a tombstone to the centurion Publius Aelius Sergius Bassus Mursa of the Twentieth Legion. Alavana stood on what is now the River Kent, though the original name for this river is not known. Might Publius Aelius Bassus have earned such a reputation that the area around his burial would be remembered by his name?
7. The seventh battle was in the forest of Celidon, that is Cat Coit Celidon.
Unlike Bassas, most scholars pounce on this site as straightforward and unchallenged. Leslie Alcock says it is the battle “about which we can have the most confidence”, adding that “there is full agreement that this was in Scotland.” But if there was full agreement when he wrote that, there isn’t now.
Cat Coit Celidon means “battle in the forest of Celidon”, which Nennius had already said in Latin. It seems strange that he should restate it in Celtic unless he wished to emphasize the original Welsh name as something specific, rather than another Forest of Celidon which had become better known in the intervening years. If so, then it suggests there are at least two Celidons, which immediately complicates the matter.
The usual interpretation is that this refers to the Caledonian Forest, which is a very widespread location. Caledonia was the Roman name for Scotland. Sometimes it was used to specify the Highlands, north of the Antonine Wall, but generally it applied to the whole country. Stories about the bard Myrddin, which we will explore later, record that after the battle of Arderydd he fled into Coed Celyddon, where he ran wild and went mad. Arderydd is the modern Arthuret, a few miles north of Carlisle. Just beyond, up Liddesdale, is the start of the present day Border Forest, which runs through to the Kielder Forest in Northumbria. There is no specific spot within this forest called Celidon, so if this is the same forest where Arthur’s seventh battle occurred it could have happened almost anywhere across the north-west, perhaps as far as High Rochester, where the Roman fort of Bremenium stood. Nikolai Tolstoy, following clues in the Scottish Arthurian romance Fergus of Galloway, has determined that the battle probably took place near Hart Fell.
There is an ancient inscribed stone here, near the village of Yarrow. Dating from the early sixth century, it commemorates the burial of two princes, Nudus and Dumnogenus (Nudd and Dyfnwal), sons of Liberalis. Liberalis may be a Roman cognomen but it is as likely an epithet suggesting he was generous, a nickname that appears as Hael in British. It is tempting to think this refers to Nudd Hael, the grandson of Dyfynwal Hen (see Table 3.4), but he lived in the late sixth century, too late for this inscription. Nudd/Nudus was involved in a raid on Anglesey against Rhun ap Maelgwyn to avenge the death of Elidir the Wealthy. Rhun retaliated with a march across Britain to York, and up as far as the Clyde, so he would have passed through this area. Possibly Nudd was killed in this show offeree, which might have been the real Battle of Celidon.
The Caledonian Forest is too generalized a description to pinpoint Nennius’s battle, and yet he seems to be trying to be specific. He does not say the “Caledonian Forest” but the “Forest of Celidon”, which may be something different, even personalised. In the Mabinogion story Culhwch and Olwen, we learn that Culhwch is the grandson of Celyddon Wledig, an important local chieftain. The story is set chiefly in Gwent. Celyddon is not otherwise identifiable, so we cannot verify his territory, but it would not be far removed from Gwent. This is the area of Arthur’s capital, Gelliwig, as we will explore later. There are several towns and localities in this area bearing the prefix gelli-, derived from celli for a woodland grove, includin
g Gellideg, Gelligaer and Gellinudd. Though no Gelliddon survives, the other names are testament to a special wood around Caerphilly, and it is quite possible that the Forest of Celyddon was once there. Blake and Lloyd have used similar logic but different etymological trails to fix Celidon in North Wales, between the rivers Clwyd and Conway.
Frank Reno follows a different route, reminding us that coed is a contraction of Argoed, the proper Brythonic word for forest, and that the phrase “Men of Argoed” was a phrase used to describe the Men of Powys. The main forest in Powys is the Clun, in present day Shropshire, and is relatively close to Baschurch and Caer Faddon. Intriguingly the derivation of Clun is the same as for Glein.
Earlier in his History Geoffrey refers to the Forest of Calaterium, where one of his pre-Arthurian kings, Archgallo, wanders dejectedly after being deposed. Archgallo is almost certainly based upon Arthwys ap Mar (Arthur of the Pennines), and I’m convinced that Geoffrey had access to a Northern Chronicle which covered the exploits of Arthwys, some of which he may have confused with Arthur of Badon’s. Calaterium is sufficiently similar to Celidon to cause possible confusion. Some experts, including J.A. Giles, have suggested that Calaterium was the old Royal Forest of Galtres, north of York, around Sutton-on-the-Forest and Easingwold. This was a rich area much treasured by the later kings of Northumbria. We will see in Chapter 9 that, according to Geoffrey, Arthur pursued the Saxons from Lichfield to the Forest of Caledon. It is over 300km to the Caledonian Forest, but about half that to the Forest of Galtres.
8. The eighth battle was at the fortress of Guinnion, in which Arthur carried the image of holy Mary, the everlasting virgin, on his shoulders [shield]; and the pagans were put to flight on that day.
This is the only battle that carries a description and is not unlike the Welsh Annals reference to Badon, suggesting that the reference to the Virgin Mary must be significant. Why did Arthur carry the image of the Virgin Mary at this battle rather than any of the earlier ones? The usual answer is that the battle occurred at a church or other holy place, and that Arthur may have been protecting a church from the heathen invaders. We should not overlook the fact that the Celts were Christians whilst the Saxons and Angles were pagans. One legend attached to Arthur, but linking him to the Crusades, tells that Arthur brought back with him from Jerusalem a splinter of the Holy Cross, which was kept at Wedale. Wedale is in the Scottish Borders and the main town is Stow. Stow is the Saxon for “holy place” and the church there is dedicated to St. Mary.
