Towns of the Damnonii, Selgovae and Votadini (straddling Hadrian’s Wall) and approximate tribal territories. The shaded area is the Debatable Land. For more detail, see figs 12 and 13.
The Kingdom of Selgovia
It took several weeks to reach this point in the atlas and I still had no idea what was waiting in the borderlands. The maps were certainly accurate enough to allow previously unidentifiable places to be identified.46 Since Ptolemy noted the tribes to which each town belonged, this had already led to a significant modification of the tribal map of Britain. I knew, too, that many of the towns shown on the maps had certain features in common. All of them stood at junctions of the Roman road network, almost one-third were inland ports and many, not surprisingly, lay in regions known to have been mined for precious minerals. It seemed likely that the Solway Firth and the inland port of Netherby/Castra Exploratorum would have been of particular interest to merchants and military commanders. But first there was the problem of the Caledonian map to be solved.
The peculiar rotation of Scotland to the east is easier to correct than to explain. The original map may have shown west at the top and Ptolemy reproduced the map as he found it, assuming north to be at the top. The other possibility is that this is an early example of southern meteorological prejudice. The Roman geographer Strabo had asserted that human life was impossible anywhere north of Hibernia (Ireland), and so Ptolemy ‘corrected’ the map which implausibly showed several tribes and towns in the supposedly frozen north by turning it ninety degrees to the right. In either case, the map simply has to be re-rotated ninety degrees anti-clockwise. Though the Caledonian map is less accurate than the northern England map – probably because it covers a much larger area – it can be read without much difficulty on a graticule of 4 by 1. The smaller number of Roman forts in Scotland makes identification relatively unproblematic since there are fewer candidates for each site (fig. 12).
The fort of Bremenium in the bottom-right corner of the restored map is known from the second-century Antonine Itinerary to be the fort of High Rochester which lies on the Roman road, Dere Street, just below Carter Bar on the Anglo-Scottish border. The place labelled ‘Alauna’ would have stood on or near Hadrian’s Wall. The most likely candidate is Corbridge – either the Roman fort or the neighbouring civilian settlement. Both these Northumbrian places are attributed by Ptolemy to the Votadini tribe, while the two Cumbrian forts to the west are attributed to the Selgovae.
In this sparse configuration of points, two other towns lie close together. Their names are Colanica and Curia. These are Celtic words, thought to be generic terms meaning ‘tribal centre’ or ‘meeting place’. Neither town has been identified until now.
Despite their proximity, each town is attributed to a different tribe – Colanica to the Damnonii of southern Scotland, Curia to the Votadini of northern England. This must account for the inclusion of two places so close together. Nowhere else on the British atlas are two towns shown in such proximity. The original map-maker evidently thought it important to indicate this tribal frontier. A merchant travelling from one to the other would have wanted to know with which tribe he would be dealing.
The accuracy of the restored map makes it possible to identify these neighbouring places with complete certainty. Thanks to the data gathered in Britain and digitized by Ptolemy, the cryptic coordinates of Colanica and Curia are as legible as the road signs on the A7 which crosses the border between Longtown and Langholm. Curia of the Votadini was the ‘ancient Citie’ whose ‘strange and great ruins’ lie under the fields of Netherby by the Esk, while Colanica of the Damnonii was the British town and Roman fort eight miles upstream at Broomholm, on the grassy platform at the end of the Roman road which forms the northern boundary of the Debatable Land.
*
Towns of the Damnonii, Selgovae and Votadini (straddling Hadrian’s Wall) and approximate tribal territories. The shaded area is the Debatable Land. For more detail, see figs 12 and 13.
It seemed (and was) an extraordinary stroke of luck to discover on a map of the second century AD the exact markers of the northern and southern boundaries of the Debatable Land. For some reason, this obscure corner of Outer Britannia was one of the most precisely mapped parts of the Roman Empire. It was strange to think that, in the second century and in 1552, when Henry Bullock created his ‘platt’ of the Debatable Land, this small region, which, even now, is missing from most guidebooks, was depicted on the most accurate maps of the time.
But was it really so astonishing to see the Debatable Land rise from the night of ages? In the land on the other side of the Liddel, two tribes, each controlling a vast area, had faced one another across a frontier. Apart from the rivers themselves and the Cheviot watershed, this is the earliest sign of the future Anglo-Scottish border and one of the oldest clues to any political division in pre-Roman Britain. The emptiness revealed by the archaeological record was not misleading. The medieval documents had been right to call that oddly resilient realm ‘ancient’. Neither Scottish nor English in the Middle Ages, in the late Iron Age it had been neither Damnonian nor Votadinian.
Ptolemy’s maligned atlas of ancient Britain also reveals a crucial difference between the frontier which still exists and its Celtic ancestor. The inter-tribal zone was not just a buffer between two states. A short distance to the south, the map shows a town or fort called Uxellum, attributed to the Selgovae tribe. This is Uxellodunum, where Hadrian’s Wall reaches Carlisle on the bluff above the river Eden.
It is easy to think of the tribes of ancient Britain as either proto-English or proto-Scottish, but the territory of the Selgovae belongs to a different era. Ptolemy’s map proves that the Selgovae inhabited both sides of the Solway Firth, which now divides England from Scotland. Their kingdom was a Mediterranean of the north: it stretched from the highest Pennines in the east to the middle of the Galloway peninsula in the west. The broad arm of the Solway looks like the permanent marker of an ancient division, but two thousand years ago, when a Selgovian stood where King Edward I gazed at Scotland for the last time, he would have been looking at his own country, and when he walked out over the sands at low tide, he would have crossed the water without leaving the motherland.
*
The kingdom of the Selgovae would have met those of the Damnonii and the Votadini in the region of the Debatable Land. Though the Selgovae possessed the Cumbrian Plain and both shores of the Solway, the vital inland port at Netherby belonged to the Votadini, while the Damnonii controlled the northern end of the Esk corridor at Broomholm, where roads converged from north, west and east, tapping the rich mining and agricultural areas of lowland Scotland. This tripartite division centred on a march is typical of the ancient Celts. A very similar situation existed in Gaul, where the great tribal federations of the Aedui, the Arverni and the Bituriges formed a broad frontier zone, later called the Marche, which is still recognizable as a linguistic and cultural watershed. The arrangement is rare if not non-existent in the modern world, despite its obvious geopolitical value.
With shared boundaries in the flatlands on the edges of the Debatable Land, all three nations had access to the trade routes of the Irish Sea. This inter-tribal zone still looks like a no man’s land. The landscape has been transformed by roads and railways, wind farms and pylons, but the physical and even the human geography is unaltered. At Longtown, the farmers of Cumbria and Dumfriesshire meet at Britain’s biggest sheep market. In 2001, the area’s centrality was demonstrated by an infected sheep sold at Longtown Mart, from where foot-and-mouth disease spread rapidly to the rest of the country.
The proto-Debatable Land may have arisen as a buffer zone in the aftermath of a destructive war, or it may have been created by consensus or divided inheritance when the region was settled by Celtic tribes. These tripartite divisions cushioned by a broad frontier could be as sturdy as the arch of a bridge sustained by the force of compression. Each tribe had an interest in preserving the mutually beneficial arrangement. The closest examples today are antagonistic rather than cooperative. The buffer zone in Cyprus, which separates Greek and Turkish Cypriots, is administered by a third party, the United Nations. Large parts of the ‘Dead Zone’ have been abandoned to nature and are useless to the people on either side. To judge by the long absence of any settlement, the Celtic solution, which was still being applied by the Anglo-Scottish borderers five hundred years ago, was to make the buffer zone ‘batable’ rather than debatable. The livestock of opposing tribes or nations maintained the unoccupied zone as pasture and prevented it from turning into wasteland.
The oldest known trysting place in the borderlands was still in use when the Debatable Land was partitioned in 1552. The Lochmaben Stone at the foot of the Sark, on the south-western tip of the Debatable Land, where English and Scottish officials discussed international matters, bears the name of a local Celtic god, Maponus (here). It has often been suggested that the sacred stone was once a meeting place of Celtic tribes. This now looks more than likely. The monolith on the ocean’s edge, where the river-borne rubbish of two nations gathers in the mud before being carried out to the Irish Sea, stands in the area where the three Celtic nations of the future borderlands came together.
The worn stone of Maponus now has nothing to tell us, and the trail revealed by the second-century map ends like the ghostly footprints on the stairs of Netherby Hall. The history of these tribal and national divisions can be traced only in the centuries-long emptiness of the Debatable Land. Perhaps the sanctity of the buffer zone was preserved by a religious veto, later materialized in the chapels which stood on its borders. The only other clue is a name traced faintly on the map of 1552, which identifies the land by the Esk at the end of the Scots’ Dike as ‘Dymisdale’. A document of the same period calls it ‘Dimmisdaill, as the common people say’. The name disappeared from maps and local memory long ago. Three other Dymisdales or ‘Doomsdales’ in Britain are associated with justice and execution; the Dymisdale of Inverness, for example, was the way that led to Gallows Hill. Did the ancient Celts and their Dark Age successors picture that sacred enclave, which lay deserted and silent after sundown, as the other world, inhabited only by the spirits of the dead?
The zone’s political significance is easier to grasp. Its nearest equivalent in the Roman Empire is the tribal frontier zone ‘of doubtful ownership’ between the Rhine and the Danube. This Debatable Land of Germania had a similar history. It was eventually invaded by ‘worthless vagabonds’ from Gaul who, like the Armstrongs and Grahams, involuntarily acted as the spearhead of a greater power. Its integrity destroyed, it was swallowed by the Roman Empire, primarily so that its inhabitants could be taxed. In both cases, borders established for the purposes of trade and cooperation became barriers to be exploited for financial and political ends.
* * *
As the day of the Scottish Referendum drew near, I often looked at the map of the kingdom of Selgovia. The Solway Firth, where one country seems to pull away from the other, was not an ancient border after all. It had been no more a boundary to the people of the Iron Age than it was to the oystercatchers and the barnacle geese. After the Romans, it had belonged to the kingdom of Strathclyde, which straddled the future border. Only much later was this kingdom which embraced both sides of the Solway divided. That division, which now seems such an essential feature of British history, had lasted only five hundred years, and for much of that time, England and Scotland had been officially at peace. The frontier zone shared by three powerful tribes had continued to serve its purpose, and when the two modern nations had united, its spirit had lived on. Now, there was a real possibility that they would be separated again.
The Celtic tribes of the borderlands proved that independent nations could form a stable partnership. Ptolemy’s map and the history of the Debatable Land could be used to support rational arguments on either side of the independence debate. But the tone of the debate was aggressive and the references to Anglo-Scottish history were becoming increasingly dubious. The Border ballads were read as evidence of fierce national pride. Even some academic historians seemed to guide the hand of history towards a ‘Yes’ vote. One scholar claimed in a study of the Scottish Middle March that cross-border marriages had been a myth invented by the English: ‘Reports of them were partly the result of English scaremongering.’ The absence of marriage contracts was held to prove that interbreeding had not been prevalent – though it is hard to imagine illiterate reivers postponing their wedding feasts so that the proper documents could be obtained from the relevant authority.
The view which prevailed in this part of the borderlands was more in keeping with recorded history. What was traditional and ancient in the Debatable Land was not division but agreement. In metropolitan Scotland, the borderers’ reluctance to see their cross-border community disrupted was characterized as ‘rural’, as though people who stand in the rain without umbrellas and who recognize the seasons within seasons must be out of touch with important realities.
On a last visit to the Lochmaben Stone, a few feet from the national border, I turned down the lane which leads to the sea in the part of Gretna called Old Graitney. In a field bordered by houses, a farmer was trying to separate two bucking cows. As I cycled past, he was nearly knocked off his feet by the cows and shouted, ‘Gie ower!’ (‘Stop it!’) The words are identical in the two dialects of English spoken in the region – Southern Scots and Cumbrian – and it was as hard to tell whether the man was English or Scottish as it might have been to know on which side of the Solway a Selgovian farmer had been born.