Also this battle-list is the best evidence of what Arthur actually did, and, I will argue, convincing proof that he was not based in the south-west of England. Only one of the battle-sites could possibly have been located in the area where almost every historian places Arthur, and that is a highly conjectural thesis underpinned by circumstance with no toponymic evidence to back it. I want to show that Arthur’s battles locate him, place him in his centre of power, the area he wanted most dearly to defend, where the enemy were most powerful. Not in Glastonbury, South Cadbury Hill or anywhere near, but in the Scottish Border country, the place he came from and had his base.
After the end of the battle-list, Nennius adds, ‘in all these battles [Arthur] emerged forth as victor’ and then stops the narrative. But from the Easter Tables it is known that there was a thirteenth battle at a place called Camlann where Arthur perished and which happened at least fifteen years after the final victory at Mount Badon. That makes ten names of places for thirteen battles, and, aside from the references to Christ and the Virgin Mary and a hint of military tactics, that is essentially all there is: ten names.
The first of Arthur’s victories was won at the mouth of the River Glen, spelled Glein by Nennius. The meaning is straightforward: both P- and Q-Celtic in their respective versions of glyn and gleann mean a valley. Many valleys bear the name: Glencoe, Glenmore, Gleneagles but there are only two rivers called Glen. One rises near Grantham and flows for much of its course through the Fens of Lincolnshire before emptying into the River Welland near Spalding. In the fifth century much of this land was undrained and lay under seasonal water. Because it was flat – no glens. More likely as a derivation is the P-Celtic word glan for a bank or banking or riverbank. Subject to flooding, the Fen rivers are all heavily embanked to protect the fields lying around them. That spelling as well as the geography would disqualify the Lincolnshire River Glen from Nennius’s list. That leaves the Northumberland River Glen as more likely, although before this begins to sound too emphatic, I have only a reasonable reading of geography and good dictionaries to support me. And what I have seen for myself.
The Glen runs into the Till which in turn is a major tributary of the Tweed, at that time firmly in the territory of the Gododdin. It flows out of the Cheviot Hills in one of the most beautiful valleys in Britain running hard by the fortress of Yeavering Bell. One of the key strongholds of the Gododdin, it is a flat-topped hill with seventeen acres enclosed behind the remains of a long palisade, and the compacted stone bases of 130 huts have been identified. As important as Traprain Law and Eildon Hill North, Yeavering Bell held the southern frontiers of the Gododdin. It lay only fifteen miles from Bamburgh, already a haven for Anglian pirates in Arthur’s time, but it is much more likely that he fought a battle on the River Glen to protect his fortress from the embryonic kingdom of Deira. Based near York in eastern Yorkshire, the Angles were limited in expansion by their kinsmen’s success to the south and so, when their numbers allowed it, they raided north into the lands of the Gwyr Y Gogledd.
Although Arthur stopped them on the flat ground by the River Glen (horse trials are still held on the banks of the river and close by is an ancient standing stone, known as the Battle Stone), the Angles eventually captured Yeavering Bell and built an extensive royal palace at its foot.¹²⁸ This was excavated forty years ago to reveal a long rectangular hall, several smaller halls, a pagan temple (later converted into a church when Paulinus baptized King Edwin of Northumbria in the River Glen) and a remarkable open-air auditorium or meeting place known as the Grandstand. Set out like a section through a Roman amphitheatre, it had seating for several hundred but room at its base for only a small dais where, presumably, the Anglian king sat. Not designed, therefore, for debate but rather as a place where the monarch literally laid down the law to a pre-literate aristocracy who all witnessed the same thing being said to them by the same man at the same time. Remarkable.
Without labouring the point, I think it much more likely that Arthur would choose to fight near an important fort, big enough to secure his cavalry, in the path of Anglian war-bands pushing northwards, on good ground of his own choosing, rather than splash around in the clatch and mire of the Lincolnshire fen.
There is another reason for believing that Glein is the River Glen in Northumberland. Because geography often dictates strategy and forces armies to meet at or near the same place, even separated by millennia, the case is strongly supported by a series of later conflicts. At the foot of Yeavering Bell in 1415 Robert Umfraville defeated Scottish raiders on their way south at what became known as the Battle of Geteryne (or Yeavering); and then in exactly the same place in 1465 a Yorkist force defeated Lancastrians in an important skirmish in the Wars of the Roses. Three miles away a disastrous battle at Flodden Field destroyed Scotland’s fighting capability for two generations in 1513. And a mile to the east of Yeavering, Harry Hotspur defeated the Earl of Douglas at Humbleton Hill in 1402, his story ultimately finding its way into Shakespeare’s Henry IV, Part I.
Two more points from this. Arthur was defending a Gododdin fortress at the Battle of Glein on ground that was flat and unobstructed, and, crucially, that was near a river. This last is a tactical theme which occurs throughout the Nennius list and it was the choice of a cavalry general commanding a small force of troopers who went against foot soldiers. Perhaps they surprised the Angles as they forded the river and splashing in on their horses turned them and rode them down along its banks or drove them into places where it was deep. A small mounted force could rout a much larger army of infantry in this way.
Parallels are available elsewhere. In Gaul military success against the Franks and Goths was due almost entirely to the brilliant use of small cavalry forces against overwhelming numbers of infantry. Around 471 an aristocrat called Ecdicius routed thousands of Goths near Clermont-Ferrand with, it was claimed in a contemporary source, only eighteen mounted warriors.¹²⁹ While the arithmetic is heroic, the principle is clear, and although Ecdicius seems to have fought only one campaign, the chronicler Sidonius described crowds dancing with joy at his spectacular success.
There is good evidence for sixth-century battles in Britain of this sort, and given the Gaulish experience of the fifth century, I think it highly unlikely that Arthur fought differently from his children and grandchildren or his neighbours.
The other germane point is to draw a clear distinction between raiding and invading. Since Gildas and the Annales Cambriae date Badon Hill in 499/500, then Glein is likely some time in the decade before that, too early to talk of Anglian settlement in the Gododdin territory. But raiding had gone on for a long time before and could be catastrophically destructive. When Arthur drove the Angles into the river, he was dealing not with an invading army but a war-band after plunder.
The next four battles in Nennius’s list were all fought in the same place; clearly a short campaign against the same enemy was prosecuted ‘beyond another river which is called Dubglas, in the region of Linnuis’. Now this is problematic despite a double map reference: the name of the place and the area where it is situated.
Dubglas is P-Celtic for ‘dark river’ and gave us the modern surname Douglas. Logically the Ordnance Survey should be scoured for Rivers Douglas and indeed there are several candidates. However at first blush there seem to be none located in a region with anything like the name Linnuis. This has, incidentally, not deterred several laureated historians from insisting that Nennius must mean the area of Lindsey, part of present-day Lincolnshire even though no river bears or bore a name remotely like Douglas – generally accepted in any case to be a Scottish surname.¹³⁰ They are so fixated with Arthur’s struggle to contain the Saxons in the south that they happily tolerate such an ill-fitting hypothesis.
However, if the campaign of four battles was not waged against invaders from across the North Sea but rather dealt with incursions from the north, from the Picts and the Scots, then the geography begins to fit the text much better. There is a River Douglas t
hat flows into the Clyde near Lanark, there is even a village named Douglas, and a Douglas Castle. And the infamous Douglas clan, at one time Scottish king-makers, and the opponents of Hotspur, originate from this bleak stretch of the Southern Uplands. Near the confluence of Douglas with Clyde there is the village of Drumalbin, a name which means the ridge of the Scots, perhaps a memory of a battle formation or a defensive position.
The value of this possibility is that it may identify a northern source for Arthur’s enemies correctly but it cannot, by any ingenuity, be made to sit in a region by the name of Linnuis.
For that it is necessary to go further north, and also to look for a strategic context. Most famously William Wallace and then Robert Bruce understood the importance of Stirling and its castle as the key to the whole of the north of Scotland. It is easy to forget this nowadays when we are able to move freely virtually anywhere. For the Dark Ages and a thousand years afterwards, Scotland was cut in two by the Firth of Forth to the east and the great and treacherous bog-land called Flanders Moss which stretched westwards from Stirling to the lake of Menteith and the foot of the mountains around Ben Vrackie. The only safe road from south to north was dominated by the towering rock of Stirling Castle, and the bridge over the Forth.
This is where invaders had to funnel through if they wanted to gain the fertile south. And this is where Arthur halted them in a brilliant campaign. Names once again contain secrets.
Earlier I noted that the Gododdin’s only modern onomastic survival was the Q-Celtic or Gaelic name for the Firth of Forth, Linn Giudain. Bede offers some help. Early in his Ecclesiastical History of the English People he notes how narrow the waist of Scotland is, lying between Alcluith in the west and Urbs Guidi in the east. Alcluith is Dumbarton Rock and the focal fortress of Strathclyde, and Urbs Guidi is Stirling, a western outpost of the Gododdin. Pronunciation of the first element of Linn Guidain begins to unlock this. In its Q-Celtic meaning of ‘firth’ ‘Linn’ is spoken with a ‘u’ suffix so that it sound approximately like ‘Leenu’. I think that ‘in the region of Linnuis’ means in the area of the Firth of Forth. Dubglas or more correctly in P-Celtic Dubhglais for ‘dark stream’ is again made clearer in the version used in its cousin language, Gaelic. Because it meanders through a trackless area of peaty bog-land and takes a deep brown colour from it, the Gaels call the River Forth ‘an Abhainn Dubh’ or the ‘Dark River’.¹³¹
Remembering how constant a threat they were to the south – the raids on London in the late fourth century showed how far they could reach – Arthur’s war against the Picts was not only a local matter, an attempt to protect the Gododdin kingdom of Manau. It was also part of a clear policy designed to protect all of Britain from its enemies. Even in 540 Gildas describes the Picts as ‘transmarini’ able to sail to whatever point of plunder looked promising. All the sources for the fifth century repeat that it was the Picts and not the Anglo-Saxons who posed the greater threat to Britannia. Bede, the Anglo-Saxon Chronicle, Gildas and such European commentaries as exist, often note the threat of the Picts. And when the Vortigern found himself in power in the south, he had immediately to deal with the problem of Pictish incursion and, fatally, he brought in Saxon mercenaries to guard against them. Even though the Angles, Saxons and other Germanic settlers posed increasing difficulties as the fifth century wore on, the fearsome reputation of the Picts, built up in Britannia over centuries, did not abate. In the 490s Arthur had to counter these ancient enemies and in four battles by the Dark River in the district of the Firth he quietened them.
And in the same way as at Glein, many battles have been fought in that place for similar strategic reasons. The Irish annals reported a Battle of Manau won by Aedan MacGabrain in 582 against the Picts, while in 711 the kings of Northumbria were pushing their power north and they defeated the Picts in the Battle of Manaw. Both Stirling Bridge in 1298 and Bannockburn in 1314 were fought nearby for the same compelling geographical reasons which brought Arthur’s cavalry to the shores of the Firth of Forth.
In attempting to decipher Nennius’s list I have followed another assumption, not, I hope, a fanciful one: that the bardic or aural source for it originated with the Gwyr Y Gogledd, the composers of ‘The Gododdin’ and the focus of P-Celtic or Old Welsh language culture. If that is so, then it is more likely to be a recital of battles won where they lived, in southern Scotland. However that may be, it does not apply to the sixth victory, won ‘beyond the river which is called Bassus’.
Neither the map nor any of the text offers much help with Bassus but my firm belief is that it was not in southern Scotland.¹³² The only ‘Bas’ names are in England, such as Basingstoke, Bassenthwaite, or Basildon and all of these are early English, relating to personal names. In P- and Q-Celtic the prefix ‘Bas’ does exist and is a borrowing from Greek via Latin for basilica for a church. The name of the town of Paisley near Glasgow is P-Celtic and has this origin. However, there is nowhere on a map of Britannia in 490 to 500 which has both the prefix and any strategic significance. In Staffordshire there are three villages called Basford which all lie on the Hammerwich Water. That last is an English name and it may have replaced a River Bassus now remembered only in the villages. Nearby is Lichfield and the Roman settlement of Letocetum which was a crossing of two Roman roads, one running north-west to south-east, the other south-west to north-east. Both Saxon raiding parties and British cavalry used Roman roads, and even by the end of the fifth century they remained in good repair.
It is possible that Arthur fought a raiding party near a river in Staffordshire whose name has been lost. He had the mobility to strike quickly over long distances but if Nennius has the sequence of battles right, then it seems an unlikely thing. Cavalry can operate only in the spring, summer and early autumn and to move so far from his base and use up so much of a campaigning season in doing so, Arthur would have needed a good reason. Glein had a logic, as did the campaign against the Picts in Manau. The battle at the River Bassus was not remembered by any other historian and the landscape forgot it too. Either it was a skirmish fought on the way to something more important, or Nennius misplaces it. Or likely both.
By contrast the location of the seventh battle is much clearer. Unusually in the vast historiography of Arthur no historian, no matter how fixed their gaze is on events in the south, disagrees that ‘the wood of Celidon, that is, Cat Coit Celidon’ is in southern Scotland.¹³³ In the near-contemporary story of Myrddin, he fled from the Battle of Arthuret near Carlisle in 573 into the Wood of Caledon. It is the place now called the Ettrick Forest, to the west of the town of Selkirk.
The question here is not the location, but the enemy. Who fought against Arthur in that place? Not the Angles who were too few and too far east at that time. Of Gildas’s list of the three enemies of Britannia, it can only have been the Picts. The Wood of Caledon is the heartland of the ancient tribe of the Selgovae. And remembering the scatter of Pictish remains that lie south of the Pentland Hills near Edinburgh it seems very likely that he had to go into battle against the old enemies of the Gododdin in combination with a Pictish force on the fringes of their territory.
Added to this conjecture is a rare piece of Dark Ages archaeology which definitely locates the site of a battle around 500. At Yarrow Kirk, eight miles west of Selkirk, hard by the river of the same place there is a standing stone bearing a remarkable description. Very weathered now and difficult to read, it carries these words:
Hic memoria perpetua
in loco insignisimi principes Nudi Dumnogeni.
Hic iacent in tumulo duo filii Liberalis.
Which translates as:
This is the everlasting memorial
In this place lie the most famous princes Nudus and Dumnogenus.
In this tomb lie the two sons of Liberalis.¹³⁴
The immediate area around the Yarrow Stone also remembers an ancient battle very clearly. To the north-west there is a marshy area of haughland known as the Dead Lake and tradition holds that it was a mass
grave for warriors killed in the same action that claimed Nudus and Dumnogenus. Near at hand is an old cottage which still bears the name Warriors’ Rest. But two more standing stones in the immediate vicinity are much too old to relate to the battle and it may be that the Yarrow Stone was a third, commandeered by the victor to commemorate his distinguished compatriots. A few hundred yards to the south-west is a concentrated group of four place-names with Nennius’s ‘Cat’ in them: Catcraig looks down on the battlefield while Catslackburn, Catslack Knowe and Cat Holes lie on the lower ground nearby.
The terrain once again suits the operation of cavalry against foot soldiers. There is flat ground, well drained, close to a river where there is an old ford, no longer used now. Just as at Glein, a surprise attack on a group of infantry could have caused panic, allowed little concerted action and turned the enemy into fatal flight. But if Arthur fought the Picts here then it did not all go well for his side. There was a cavalry and chariot tradition among the Picts which may have stiffened their resistance. Two P-Celtic princes, who sound like Damnonians from Strathclyde, fell in the fray and warriors were buried in a place that people remembered. None the less only victors erect memorials and Arthur’s expedition into the heartlands of the Selgovae achieved its objective.
The Battle of Celidon Wood is the only location in the Nennius list that no one disputes. Just as in ‘The Gododdin’, an epic poem written in Edinburgh, Arthur has his earliest literary reference. Facts are extremely rare in all this and these may be the only two bits of documentary evidence that are universally accepted. And they put Arthur in his right place: a leader of battles, a cavalry general who led the coalition army of the P-Celtic kingdoms of the Old North, the Cymry.
Arthur and the Lost Kingdoms Page 21
