Wansdyke home I What's New I Sitemap I Bibliography I Vortigern Studies l Vortigern I POLLS I LINKS l Sitemaster I FAQs
search l about Vortigern Studies l Messageboard I Games I Arthurian Collection I View Guestbook I Sign Guestbook

  Vortigern Studies > Wansdyke > Articles > Nurse2

Guest Author:
Keith NurseKeith Conrad Nurse

Keith Nurse was for many years the Daily Telegraph’s Arts Correspondent in Fleet Street, covering a range of subjects, including archaeology, theatre and heritage issues. He became a freelance in 1986 and later worked for Thames TV and Carlton Television. He is the author of Torments Ancient and Modern, which is an account of working-class life in Wales in the '40s and 50s, and of Footsteps to the Past, a family Footsteps to the Past, a book by Keith Nurse
history account set against the uniquely rich social and historical landscape of north-east Wales.

Vortigern Studies Index


.Wansdyke Project 21
is part of
Vortigern Studies



Articles about Wansdyke
click here

  Wat’s In A Name?
Keith Nurse

Recent archaeological findings have focused attention on Wat’s Dyke - a little known frontier earthwork in the northern Welsh borderlands. Until recent times, this linear boundary line has been a neglected landmark, not widely known outside academic circles.But the developing debate about the political and military context that led its construction - and whether this background lay in the post-Roman period or later in early medieval times - is likely to engage scholars in a challenging exchange for some time to come. It is a debate that touches on complex issues with wide-ranging implications for the study of the most difficult to understand periods of early British history. What follows is a discussion based on some of the many-sided arguments that are now emerging.


On one issue if nothing else, historians and archaeologists are in agreement: Wat’s Dyke is an archaeological and historical enigma. A rampart and deep ditch earthwork, it runs for some 40 miles through the northern sector of the Welsh borderlands.  

View to the Welsh west from the line of Wat's Dyke, near Wrexham.
View to the Welsh west from the line of Wat's Dyke, near Wrexham.
Click the image to enlarge.

From the ground, it is often visually disappointing, especially when its profile is compared with that of its parallel near neighbour, the famous Offa’s Dyke, as the latter follows an impressive course across higher land, to the west. The linear landscape feature bearing the name of Wat looks much like any other field boundary: a high standing line of hedgerow, with its once west-facing ditch mostly invisible, long filled in or ploughed out. Sometimes it simply appears in the distance as nothing more than a ‘crop mark.’ But present-day views are misleading. Strategically, Wat’s Dyke is considered to be a more sophisticated construction than its near counterpart to the west. The consensus view: it is a defensive boundary frontier line, skilfully designed to take advantage of the available natural west-facing features to protect the agriculturally productive territory, just below the foothills of the Welsh uplands. This tactical approach suggests that those who built the dyke possessed an intimate knowledge of the local terrain, and knew just how best to define and protect the productive land to the lowland east that the boundary line defined.   

Aerial photography has highlighted its distinct profile, both as a crop mark and a surviving above-ground feature, and its line is at its most striking where it links the two major border towns of the region, Oswestry and Wrexham. Here it follows a distinct hedge-line course through a rural landscape that has remained largely unaltered for centuries. Given its frontier role, it is not surprisingly to learn that the dyke bears all the hallmarks of the work of a man trained in the military manner. [1]

But recent archaeological-based evidence has given rise to new speculation on its possible political genesis. This is a radical and inevitably controversial new reading of the background, because it suggests that the earthwork was built 300 years earlier than previously thought. This would mean that the dyke was constructed in the post-Roman turmoil of the mid 5th century. But the scenario is a highly plausible one, and raises the possibility that this politically enduring frontier line was inspired by one of the most controversial figures in British history – Vortigern, the ‘superbo tyranno’ as depicted by the British monk Gildas, in his celebrated 6th century work, De Excidio.

The historical constructions or models that emerge from this view of the period are not as fanciful as might first seem. Firstly, the dyke could have formed part of the late Roman strategies designed to counter the ‘barbarian’ threat – in this case sea-borne Pictish and Irish incursions into what is present-day North Wales. It is equally conceivable that the earthwork emerged somewhat later, as the result of a schism within the kingdom of Powys, or as a result of the ceding of territory as part of a peaceful or highly controlled English take over in the mid 7th century. It has been argued that Penda (d.655), the Mercian king bearing a Welsh name, possessed both the resources and authority to have built the dyke. Alternatively, its construction may have resulted from the direct rule exercised over northern Mercia by the Northumbrian king, Oswui (fl.642- c652) who was all-powerful among the English kings.

As a frontier demarcation line, then, Wat’s Dyke may have had little or nothing to with divisions along English Welsh lines that later became associated with it, but, rather, pre-dated these divisions and reflected the political realities of rival kingdoms of an earlier age; a period of confused loyalties, military alignments and dynastic splits. It might, then, have been an isolated frontier marking measure that subsequently gained renewed importance with the late sixth/early seventh century emergence of English Mercia, in the area of what is now commonly referred to as the Midlands.

The line of Wat's Dyke, marked by a hedgerow, near Wrexham.
The line of Wat's Dyke, marked by a hedgerow, near Wrexham.
Click the image to enlarge.

This is just where the narrative becomes confused. It does so because of the overlapping of British and English communities and elite control, and the emergence of little known hierarchies and the forging of unrecorded (mostly fleeting) military alliances. Significantly in this context, we have to consider the powerful, intrusive influence exerted by Northumbria. In addition, acculturation (acceptance, out of self-interest, by one group of the powerful cultural features of another) was not unknown in Mercian territory. It was highlighted by the archaeological findings from a British/Romano-Saxon cemetery at Wasperton, Warwickshire. Here emerged evidence of two-way contact between Saxon and Briton, making the background distinctly blurred and politically ambiguous. It was an unrecorded phenomenon that was doubtless much more widespread than has been previously understood.


As the sole authority offering a continuous narrative of what might have been going on going in 5th century Britain, Gildas’s De Excidio is without doubt the key account to this past. There are compelling reasons – or at least as compelling as they might be in the murky circumstances - to suggest that he was much more intimately acquainted with the north-Wales region than has been previously acknowledged. And on Vortigern, as with certain other aspects of the post-Roman period, his splenetic tract, written circa 540, bears the unmistakable ring of an authentic, if lone voice; but of its time and engaging in an ecclesiastical theme of prophetic dimensions and conveyed with an Old Testament vehemence.

All a bit unreal but, more specifically, there are grounds for arguing that De Excidio offers an essentially north- western view of events.   And it is worth noting that, even before the disclosure of the recent archaeological findings pointing to the new dating evidence for Wat’s Dyke, there have been suggestions that the earthwork might have been in existence sometime before Gildas’s day.[2]

The difficulties that arise here are thus many, varied and difficult to unravel. Unlike the documentary references to Offa’s Dyke, those for Wat’s Dyke are relatively late. However, what is not less well known is that cryptic references to dykes occur in early Welsh poetry, in particular those preserved in Powys in the collection bearing the name of Llywarch Hen (the Old). They have been described as the finest creations of the Cyfarwydd, as the storyteller was known in medieval Wales. A sixth century figure, Llywarch – maybe a poet-chieftain or even an heroic amalgam of the two - is said to have been a cousin of the celebrated British king, Urien of Northern Rheged, whose base lay close the borders of present-day Westmorland and Cumberland. The narrative of the poem confusingly switches scene, from the north to Powys, where the territory of the British kingdom of the Welsh borderland had been divided, probably by partition among heirs.  

But the importance here is that the poet makes what are arguably the earliest of all references (though not by name) to the presence in the landscape of the two earthworks, Offa’s and Wat’s Dykes, in this area – and where the distance between the two is at its shortest.

The context is very much in keeping with the heroic nature of the times - the death song of the warrior prince, Gwen, one of the sons of Llywarch Hen.

Gwen by the Lawen kept watch last night
In the fight he fled not
Sad is the tale, on Gorlas dyke..

Gorlas is believed to be a scribe’s error for Morlas, referred to in a later line ‘At the Ford of Morlas Gwen fell.’  The Morlas, a tributary than runs into the river Ceriog, here runs close to Wat’s Dyke, and excavation has shown an earthwork to be present along the southern length of the Morlas Brook.  Elsewhere, there is a reference to ‘dykes.’  Again, it can only refer to earthworks in this same locality- just where battle lines might be drawn.

The relevant sequence – characterised as an epitaph of a nation - is rendered by a man reconciled to permanent exile in the hills. As a lament, it seems to confirm the worst fears expressed by Gildas, many generations later. It is also a convincing one, because the poet is also talking about a genuine material loss - the loss of good land, adjoining a major river source, territory now denied to the men of the hills who now find themselves glowering behind the dykes.

The key sequence runs thus:

Gone are my brethren from the lands of the Severn
Around the banks of the Dwyryw

Glad am I, my God that I am yet alive
Brothers I had who never lost heart
Brothers who grew like hazel saplings
All are gone, one by one
The dykes endure. 
He who dug them is no more.

And endure they did, along with the richly dramatic handed down tales of the early Welsh storytellers. But here it appears that the poet was setting down tales in the 850s. They were, in essence, bard-related stories incorporating the memory of battles, engaged at long remembered sites: conflicts that had badly hit Powys more than 200 years before, if not earlier. The Welsh scholar, Sir Ifor Williams, considered these tales to be Powys-based sagas, produced in the mid-ninth century, but reflecting on momentous events of the 7th century.

We may argue with the hyperbole of these richly expressed themes - but surely not with their underlying accuracy and pinpointing of known landscape features. These stories are not the result of later scribal interpolations; they represent interpretations of events that lay deep within a local folk memory, a word-of-mouth inheritance that was not committed to written form until much later and which had little or nothing to do with later subsequent church rivalries, or the accretions of the later Arthurian romances. The words of the poet stood alone, pristine and deeply evocative of their time.

So from this, we can reasonably deduce that the references they contained reflected the existence of the two dykes, as they appeared to the poet in the mid-9th century. But this, it might be argued, only establishes that both earthworks were visible at that point. It does not in itself indicate that Wat’s Dyke was of earlier origin. But then the crucial reference to the Morlas brook (almost adjoining as it does Wat’s Dyke) does suggest that. As a long-remembered battle location – one, moreover, signified by the death of a major figure, a Powysian chieftain – we might reasonably assume that the presence of the dyke made its mark locally and was already in existence as a frontier barrier and thus the place to do battle, sometime in the mid-7th century. Yet to the Welsh at least, it then bore no personal name. This indicates there was little or no diplomatic contact or exchange with the Saeson (English) to the east, then occupying one-time Powysian lands surrounding the Severn, territory that had been irrevocably lost.

But what of the later references, for they are the only ones upon which we can rely for identification by name. The earliest of these documentary references to Wat’s Dyke are far removed from the time of its construction. The 19th century Wrexham historian A.N. Palmer recorded in 1897 [4] that he encountered three documents referring to the old name of the earthwork.

One occurred in a deed of year 1431, where it appeared as Clauwdd Wad; another of 1433 referred to Claude Wode – and a mine lease reference of 1472 refering to Claghwad – that is Clawddwad  – allowing lessees the liberty of taking iron from upon as well as under the land from the ditch called Claughwad – that is Clawddwad or Wat’s Dyke to the mountain of Glassffrey (Glasfre) in the parish of Ruabon.’  Until the 17th century, Ruabon Mountain or the Mountain or Mabon was known as Glasfre.  It was this latter name that is believed to have led to the confusion linking St Collen with Glastonbury in Somerset. Glast or Glas is also considered to be a personal name for a son of Cunedda’s son, Dogfael, identified with the southern part of what is now called Dyfryn Clwyd, in the Vale of Clwyd, the latter river name linked to the name of the great Scottish waterway, the Clyde and the region of Strathclyde.

Clawdd Wad and Glasfre, according to Palmer, were ‘practically and severally’ the eastern and western boundaries of the parish of Ruabon There are further pointers to the presence of the dyke in this region.  Land held by a prominent landowner in 1620 in the Acton area, to the north of Wrexham, included a Cae Wad field name which continued in use into modern times, appearing – along with the distinct line of the dyke – on a map of 1844. The field name clearly referred to the then prominent and largely undamaged earthwork, which ran nearby to the west, forming a township boundary. Adjoining fields to the earthwork were known in English as Acton Moor, and in Welsh Gwaun y terfyn (the boundary moor). The highest point of Acton hill, overlooking Wat’s Dyke, was known in the 18th century, and earlier as Allt y Groes, after the cross that once stood on its crest. Among the many crosses in the area mentioned in 1699 by the Welsh antiquary Edward Lluyd (the scholar who transcribed the ancient text on the Pillar of Eliseg, near Llangollen,) was ‘Croes in ye Township of Acton.’  It survives today, incorporating a Welsh personal name as the suburb of Croes Eyneurys. From all this it is clear that for centuries, the dyke, whether as boundary or an ancient feature with deeper associations, had been attracting significant field names and Christian symbols.

But just what lay behind the importance of Wat’s Dyke? Its role marking the fiscal limits, surely, because its line remained fossilised in the tribute raising arrangements along the Northern Marches long after it had lost its military purpose. Even in Offa’s time, it seems to have been an ancient landmark, protecting and marking off the productive lowland areas and, crucially, transmitting the tribute levying authority of an earlier, maybe dynasty linked, order.

Scholars have suggested that this northern borderland territory was probably hidated shortly after it fell under the direct control of the mid-7th century Mercian authorities. (Worth noting here, too that the early royal notables of Mercia no doubt encompassed a British/ Welsh element in their hierarchy, and most certainly among their subject peoples and that much the same sort of mix seems to have applied to the English kingdom of Northumbria). And when the 7th century change over of borderland authority or control occurred, Wat’s Dyke had already been constructed and was functioning as a frontier.

Wat’s Dyke and Offa’s Dyke

Unlike the earthwork bearing the seemingly obscure name of Wat, its famous parallel neighbour was mentioned in an early source, and of course attributed to the Mercian king Offa (r 757-796). Offa was a contemporary of Charlemagne and internationally recognised as ‘king of the English.’ He is more familiar to us today from the great dyke that has long been recognised as a byword for the separation and/or separateness of the Welsh and the English.

It was Asser, the Welsh-born bishop of Sherborne and biographer of the great Wessex King Alfred – writing within 100 years of the Mercian king’s death – who tells us that the great dyke was built by the all-powerful Offa. Using familiar if confusing literary conventions, he described it as stretching ‘from sea to sea.’ That has been proved to be misleading: ‘river to river’ might have been a more accurate definition. But Asser’s reference does offer one of the very few instances where a linear earthwork has been dated through documentary evidence. Moreover, it is the only one of its class in Britain to which the name of the builder has been consistently attached. [5]

But the relationship between the two dykes is still puzzling. If, as has long been argued, Wat’s Dyke was constructed by Offa’s 8th century predecessor, Aethelbald  (716 –757), why then was this similarly prominent feature in the landscape, one performing largely the same frontier role, directed against the hills men to the west, not also given the name of its builder?

This implies that those who took over its authority as a fiscal demarcation line were unaware of the identity of the man or people behind it, were far removed from the time of its construction and the significance of the name from which it was derived. Wade/Wad/Wado was in fact a saga hero, originally a ‘sea-giant’ – essentially a helper in need’. [6]

But just as puzzling: why are these similarly aligned earthworks never more than three miles away, sometimes much less than that?   And why are they especially close to each other in this area? A military reason, no doubt, because both are focussed here on two major river valley routes (Ceriog and Dee) from the mountainous areas. The stragetic background for their siting, however, would have been just as compelling in the 5th century as it was, 300 year later, in Offa’s time. And it is to the post-Roman fifth century that the new dating evidence from Wat’s Dyke at the Shropshire border town of Oswestry (a name bearing associations with the 7th century Northumbrian king Oswald in both its English and Welsh forms) points. Combined with this, it is claimed by the excavators that the earthwork was a contemporary of the great Wiltshire boundary of Wandsyke.[7] If that is so, such a context in this region at least, suggests a scenario that conceivably links the earthwork with the ever-controversial figure of Vortigern and/or his successors.

Nennius’s Historia Brittonum (parts originating from the 7th century?) puts Vortigern’s accession at 425. According to Continental sources, Britain was subjected to the Saxons in the 440s, perhaps 445/6. Interpretations of Gildas’s account suggest that the British appealed for help to Aetius ‘thrice consul’ in the west in 446. Then there is the story of the transfer of the Votadani chief, Cunedda, and his follows from the Edinburgh/southern Scotland region to north Wales. This movement, one consistent with typical military Roman strategy of the time, has been placed in the mid-5th century.Other, almost uncanny chronological coincidences occur in this context. The tombstone of the Irish chieftain/warlord/city guest, Cunorix, found at the former Roman city of Wroxeter in Shropshire, has been dated on linguistic grounds to 450-475. Local tradition – admittedly one now much disputed – claims that the so-called ‘Alleluia’ victory over the Picts and Scots of St Germanus, Bishop of Auxerre, occurred at a site near Mold, Flintshire, in 447.   Lastly, we have the dating for the finds from a hearth beneath the line of Wat’s Dyke at the borderland town of Oswestry - independent laboratory carbon-dating analysis focussing on 446, and as controversial as any of the other dates offered in this broad sequence.

More questions arise here than answers - and several difficult queries immediately occur: are these, sometimes almost converging dates, merely a coincidence? And do we summarily dismiss as of no consequence persisting local (Welsh) traditions, because they are not verifiable, either archaeologically in the soil or in the annals? And here is yet another one: was the construction of Wat’s Dyke one of the military measures adopted to counter the Pictish and Irish threat in north Wales? The presence of the Irish seems beyond question, with inscribed stones pointing to Christian and Irish elements in the population in the 5th and 6th centuries. Equally, there should be no surprise about the emergence of Pictish intruders in the area, for they seem to have left their dynastic mark. There are suggestions that remnants of a defeated Pictish army were given permission to settle in parts of Powys in the Roman period, and that they later sought permission to take as brides the daughters of local British nobles. But having failed to do so, they instead turned to the women of Gwydel – or Gwyddelig (Irish). Documentary evidence survives in a 14th century manuscript at Jesus College (Oxford’s Welsh college) pointing to the presence of a fifth century ‘king of Gwyddel Ffichti of Powys’. [8]

But what do the topographical/features, and with them local traditions, suggest? In truth, a mesmerising amount, but much of it attached to legend, traditions and folklore, though together forming an underlying narrative that bears the ring of what might be called a coherent folk-truth – the sort of the ‘synthetic’ history now referred to by scholars. But the central core of it all, in this context - the one that might be assembled from these various sources - sounds highly plausible, and nowhere more so than when it focuses on the area around Llangollen (llan – church or enclosure of Collen/Gollen) in the period in Wales when obscure monks were elevated to the status of local saints. Llangollen, above the prehistoric, revered rock-tumbling waters of the river Dee, lies at the centre of a region of almost haunting natural beauty, one teeming with stories of Arthur and his knights, along with references to giants, demons, devils and wizards. Here lies the interface between the natural and the spiritual. Not surprisingly, the religious and secular leaders were powerfully drawn to the area – and left their mark upon it.

The Political Background

In the context of sub-Roman Britain, the name of Vortigern inevitably looms large. And despite the myths, and the demonising legends – the so-called incest slur – that have accumulated around this unique figure, the underlying narrative about the ‘superbo tyranno’ and those followed him, is compelling, and especially so as the tales converge on the Welsh kingdom of Powys.

But then there is the argument pointing to an Irish genealogical lineage for that other key figure in early north Welsh history, Cunedda.Again, here is a background central to so much that followed in this part of Wales in the post-Roman period and beyond. The movement of Cunedda could not have taken place before the mid-fifth century. Further, his grandson was still fighting the Irish intruders in Anglesey circa 500 [9] Chester may have been the base for his campaign, or possibly it functioned jointly in this role with Wroxeter. None of this would not be surprising if, as we are told, Cunedda’s ancestors bore Roman names and his men were trained in the Roman military fashion. And at Wroxeter, it has been established that the local craftsmen were still formally trained in the Roman manner, building according to classical measurements and standards - but in wood rather than stone.

Cunedda it was who gave rise to the important Gwynedd dynasty.  Maelgwn, king of Gwynedd in the first half of the sixth century, was his great grandson, and of course the powerful north Wales monarch was one of the five rulers denounced by name in De Excidio. 

The ferocious and lengthy condemnation of Maelgwn (Magloconus,..dragon of the island’) is detailed enough to suggest that Gildas possessed an almost first-hand knowledge of the personality of the target of his venom. However, Gildas does not offer any detectable information on the dynasty of Powys that, in its early period, was clearly focussed on Wroxeter. And it is here that, arguably, we might look for the origins of Wat’s Dyke.

John Morris refers[10] to the Wroxeter dynasty of Constantine. The name is found, indirectly, in a reference by Gildas as Constantine ‘tyrant whelp of the filthy lioness of Dumnonia’ (covering Devon, Cornwall and part of Somerset). According to Morris, leading Cornovii (Wroxeter) families moved to Dumnonia, probably around 430, and that one Ducco – also called Congar (died 473)- lived as a monk on an estate he established there.This latter name was subsequently preserved in the present-day name of Congresbury, in Somerset, south of Bristol). Leslie Alcock[11] tells us this is the only major fortification (Cadbury-Congresbury) in Wales and Dumnonia to have produced acceptable evidence for continuous occupation from the third century through to the sixth.

But what about the new discoveries from Wat’s Dyke at Oswestry, some ten miles or so to the north west of Wroxeter – an area that lay within Wroxeter/ and early Powysian control? The finds from a fire or hearth on a site beneath the dyke at Maes- y- Clawdd (field of the dyke) have produced carbon dating evidence pointing to the mid-fifth century dating. If we accept these and the arguments that followed, the previous reading of the early medieval background surrounding the construction of the earthwork is effectively turned upside down. The excavator’s conclusion from this is that the dyke is likely to have been the work of the ’skilled men of post-Roman Wroxeter and that it should be regarded as a contemporary of that other great 5th century earthwork, Wandsyke in Wiltshire’.[12]

To repeat a key point – the laboratory analysis of the burnt charcoal remains focuses on the date of 446. This dating evidence, first published in 1977, has now been robustly challenged, giving rise to a lively debate, one that is likely to continue for some time.[13] But the proposed Wroxeter link is by no means beyond the bounds of possibility. Long considered to have been a cultural backwater in the prehistoric, Roman and later periods by scholars, the north-west Midlands and the hinterland around Wroxeter has been shown to have been as highly productive and wealthy zone as any in the rest of the country. This was rich agricultural territory worth protecting - as much in the post-Roman period as before.

A long pioneering excavation campaign at Viroconium - symbolised today by the still standing remains of the 25 ft high south wall of the baths complex, the ‘Old Work’ - has revealed the presence there of a large and skilled workforce, deep into the post-Roman period. Major building work was going on there in the 560s and it included the timbered villa-like headquarters of a major figure. The annals don’t tell us that, but archaeology has.

The 'Old Work' at Wroxeter, the south wall of the baths basilica.
The 'Old Work' at Wroxeter, the south wall of the baths basilica.
Click the image to enlarge.

The tribal civitas capital of the Cornovii has not yielded its secrets easily. But the lack of closely datable finds has been relieved by that one single fascinating if enigmatic object – the Cunorix tombstone commemorating an Irishman. He is likely to have been a high-ranking guest or maybe a leader of a band of mercenaries hired by the ruler of the former Roman town.

The 'Old Work' at Wroxeter, site of a major refurbishment and building in the post-Roman period.
The 'Old Work' at Wroxeter, site of a major refurbishment and building in the post-Roman period.
Click the image to enlarge.

His hosts, or their church representatives, possessed sufficient Latin to give him a memorial in keeping with his status. As indicated earlier, this has been dated on linguistic grounds to 460-475. The strengthening of the ditch-and-rampart defences at Wroxeter is known to have taken place in the fourth century, when there is marked evidence of a military presence. This has been demonstrated by the discovery of cross-bow brooches, and lead-weighted darts – the largest known number uncovered from a single civilian site in Britain not located in the militarised frontier zone.If nothing else, it implies that the area was coming under increasing threat, and that challenge could only have been coming from the west, or via the mountains for the west.The brooches uncovered are known elsewhere to be badges of rank and status, associated with equipment worn by both military and civil service officials, in the late Empire.

Wroxeter, in its re-constituted form, outlasted the Roman Empire in the west by some 150 years, its craftsmen and artisans still following classical building forms. In its final stages, the old capital was not stormed or put to flames by outside invaders; it was abandoned, in the early 600s, peacefully and in an orderly manner – either that or under armed guard – and its timber buildings systematically dismantled, no doubt for use elsewhere.

Some scholars have argued that a Vortigern-Wroxeter link is a ‘tempting ‘suggestion, one that places the fifth century developments on the baths basilica site in the political context described so confusingly yet so vividly by Gildas.[14] Given its location and central role within the emerging Welsh kingdom of Powys, Wroxeter could have been one of his citadels of power – or retreats into Wales.


As in the case of Offa, the name of Wat survives today in north-east Wales, attached to landscape features and modern street and district names. The name Wada, from which it derives, was not uncommon among the Anglo-Saxons. This is no doubt due to the persisting tale of Wade, a sea-giant celebrated by the coastal tribes of the North Sea and the Baltic. In his De Nugus Curialium, the 12th century writer and royal court official of Welsh descent, Walter Map refers to Gado (Wade) a prince of the Vandals, as part of an anachronistic and fanciful tale, in which the prince was associated with Offa in a fight with the Romans. In the 7th century Anglo-Saxon poem Widsith, Wade appears as a saga hero ruler. The place name Wade occurs in North Yorkshire, attached to a Roman road (Wade’s Causeway) on Wheeldale Moor, south of the seaport town of Whitby.Intriguingly, it lies close to a village called Goathland (Godelanda c1100).

According to local legend, the road was built by a giant called Wade to enable him to take his cattle to market. Two standing stones at Barnaby, North Yorkshire, known as Waddes Grave (one still standing) were traditionally considered to mark the resting place of a giant. There was also a site on Hadrian’s Wall, called Wades Gap – Wade, according to tradition having a castle in the area.[15]

In the Welsh borderlands, the personal name of Wat has long been a mystery and its absence from local place names in the region has been deemed to imply that its context is an early one. In this region, only the dyke was so named. In addition, there is no evidence of legendary giants associated with the earthwork here, though it is sea-linked, ending on the Dee Estuary in present-day Flintshire. It seems clear, then, that the name Wat – familiar and of evident early origin in Northumbria – became attached, at some imprecise date on or after its construction, to the boundary in the Welsh borderlands. And this same demarcation line turned out to be administratively important for centuries - a border marker serving as a fiscal (manor-hidation payments) territorial line for the western limits of Mercia. It performed that role up to the Norman Conquest. Hides were the standard method in early England of assessing renders due from an estate. To the west, beyond the dyke, other movements were afoot – at the site of what later became the 13th century-founded abbey at Valle Crucis (a short distance from Eliseg’s Pillar). There are suggestions that this important riverside site could have housed an early Celtic church.

Amidst the ruins of the abbey cloister, the 19th century excavators found vestiges of Saxon and Roman occupation, including what were claimed to be remnants of a hypocaust and tiles. But then below this layer, they uncovered the burnt remains of an oak foundation. These were considered to be the remnants of an early Christian church.[16] Above the Valle Crucis site lies Ruabon Mountain, a topographical name that incorporates the Welsh name of Mabon (Divine Youth). Intriguingly, Romano-British altars have been found dedicated to the Celtic diety Maponus (four equated with Apollo) in the Cumberland and Northumberland districts, on or close to Hadrian’s Wall. The dedications were made at frontier sites, including Corbridge, and were made by high-ranking Roman officers. A shrine to the god is known from the place-name in Dumfreisshire, Lochmaben, known from the early documentary record, Ravennna Cosmography, in a way that suggests that the cult centre was located there.[17] The Greeks and Romans equated this Celtic diety with Apollo, and he became the Mabon fab Madron of early Welsh legend.[18]

On more tangible ground, perhaps - the parish church at Llangollen contains an inscription to Collen, indicating that he was a descendant of Cunedda Wledig:

‘St Collen son of Gwennog, son of Coleddog, son of Cawdraf son of Caradog Freichfras, son of Lly Marini, son of Einion Yrth, son of Cunedda Wledig, Ethni Wyddeles, daughter of Lord of Cwl, in the kingdom of Ireland Which saint was buried here’[19]

The same study indicates that the tomb of Collen at Llangollen was mentioned in the Rural Dean’s report of 1749 and then again by the Welsh antiquary Thomas Pennant, in 1771, when the much worn remains of the tomb were lying in the church belfry. This alabaster figure was originally in the possession of Valle Crucis. It was removed to the parish church at Llangollen, which was owned by the abbey.   The magnificent carvings in the church were rescued from the abbey following the Reformation.

Four miles downstream from Llangollen, the parish church at Ruabon, is now dedicated to St Mary. Oddly, its original dedication was to St Collen. In 1699, the antiquarian Edward Lhuyd recorded another site associated with St Collen. This lay in the township of Dinhinlle Isaf, in the south of the parish of Ruabon, and close to the line of Wat’s Dyke. Lhuyd also recorded a further site associated with St Collen: ‘Clawdd Collen on Cefn Ucha, in the township of Pengwern, a short dyke, similar to Offa’s Dye.’ The church at Ruabon, and that at nearby Chirk (both now named St Mary) - and possibly the parish church at Wrexham - all had their dedications changed when they came into the hands of the monks of Valle Crucis in the 13th century.  It is worth noting, here, that the Cistercian order, to which Valle Crucis belonged, was accused of hostility to the claims of saints that lacked a Catholic reputation.[20]

But there are echoes here, surely, of earlier and deeper hostilities towards the Celtic/British church – enmities that the Venerable Bede had been expressing centuries before? It may be just a coincidence - but the Collen–associated locations lie close to Wat’s Dyke. Moreover, the dyke here is situated just a few miles west of the monastic settlement at Bangor-is-y-coed, where Gildas is believed to have lived at one point. There have been other suggestions that he might have been spent some time at Chester, a centre of early ecclesiastical activity and venue of a major synod of the British church in 603.

At some date before 1202, Madog ap Gruffydd, a prince of northern Powys, gave to the monks of Valle Crucis, both the site of the abbey and also lands elsewhere. These lands included that portion of the Wrexham area known for centuries as Wrexham Abbot, and where a modern town centre street names survives as Abbot Street.

But are there distant echoes here of something else - the so-called demonising of Vortigern? The characterising of him as the ‘devil incarnate’ of British history is a deeply controversial issue. Yet perhaps his true sin was his heretical role - a Pelagian, who struck up a pact with the devil, (the pagan Saxons), though of course in engaging them as federate troops, he was only was following a practised Roman precedent. His other enduring sin, of course, was that his strategy went disastrously wrong.

But is there, indirectly, evidence here that touches upon that vexed point? More than 100 years ago, the Wrexham historian A.N. Palmer[21] encountered an intriguing name-attachment linking its construction of Wat’s Dyke with the devil - though this is common in relation to ancient boundary dykes elsewhere in Britain. At Wynnstay Park estate, previously known as Watstay at Ruabon (the Welsh name incorporating the name of Mabon as detailed earlier) he came across the folk-tale that the dyke here was built by the devil and that a plea for a ‘stay’ accounted for the gap in the earthwork at this point. In the early 19th preparatory drawings for the Ordnance Survey maps in this area, the line of the earthwork was indeed assigned the name of ‘Devil’s Dyke.’

A few miles to the north, in the Acton area of Wrexham, near the boundary formed by Wat’s Dyke, he also noted a field name lying close to the line of the earthwork that just might be significant: Tal y geifr (the goats’ headland). A common enough name attachment in a rural area of hills, maybe. But nearby, again close to the old dyke, there was also a field bearing the name Cae’r swynwr (wizard’s field).

A mere coincidence this, maybe, but according to Bede, the original name for the great 6th century royal ceremonial centre at Yeavering in Northumberland was ‘Ad- Gefryn – ‘hill of the goats’ in Welsh. It was here, significantly, that archaeologists discovered a rare burial of a pagan priest with a stylised goat decorated metal staff and a goat’s skull, considered to have been a badge of office. This, it has been argued recently, is evidence of a taboo activity known to be deeply offensive to the early Christian church: the early Anglo-Saxon veneration of goats that, in turn, gave rise to the portrayal of the anti-Christ as a goat-like figure with hooves and horns.[22]

St. Guthlac attacked by devils, which are portayed as humans with the faces of goats.
St. Guthlac attacked by devils, which are portayed as humans with the faces of goats.
From the 12th-century Guthlac Roll.
Click the image to enlarge.

So who, in this confused past was given this image of the old demonic goat? Vortigern, no less. The traditional depiction of the anti-Christ in this context occurs in the picture from a 15th century English manuscript.[23] It shows clearly ‘the strangely hoofed’ image of Vortigern.[24] The significance of the goat-like figure comes from the role of the Nordic sky god, Thor, or Thunor to the early English, who rode the heavens in a chariot drawn by goats. Depictions of goats are said to remain unknown from material evidence from the 8th and 9th centuries. This suggests that, like runes and the swastika symbol (associated with Thor) - and featured on early Anglo-Saxon brooches and pottery and sword hilts – their potency as taboo symbols of paganism survived.

It would be unwise to stretch the devil and demonising references too far. But it does seem that the 6th/7th century Northumbrian influence in the Welsh borderland region has been little understood. And perhaps this disguises the degree of deeper effects its warring leaders and, more importantly its churchmen - exerted in the region. The overwhelming Northumbrian success over the Welsh at the Battle of Chester, circa 616 is an important political event, just one example of how those influences began.

Militant and expansionist Northumbria was a powerful force in the 7th century and, indirectly, it brought to Mercia the influence of the Celtic church, albeit but from another direction. The present-day church on the banks of the Severn at Atcham - mostly built of re-used stone from the ruins of the nearby post-Roman city of Wroxeter - bears a unique dedication, to the Northumbrian monk, St Eata, Abbot of Hexham and Bishop of Lindisfarne. He died in 685. The church is one of the earliest surviving buildings in Shropshire.

At nearby Frogmore, Atcham, aerial photography has picked out crop marks of an unexcavated group of linked Anglo-Saxon (or British-influenced?) timber built halls. They are unique in Shropshire, and are strikingly similar to the seventh century ‘royal halls’ of the Northumbrian kings, excavated at Yeavering.[25]

The Welsh borderlands region, in its northern sector at least, is littered with references, place names and monuments that touch upon this difficult period. Sometimes they seem to confirm the evidence in the annals, sometimes almost contradicting them. Yet, strangely, together they confirm the essence of a consistent and coherent historical theme.

For example, Wat’s Dyke clearly incorporates the magnificent Iron Age hillfort known as Old Oswestry (or Hen Dinas) – a site with claimed Arthurian links - as part of its line as it travels through the area of the Shropshire border town. This is the same fort noted by the 17th century Welsh antiquarian, Thomas Pennant, as being called Caer Ogyrfan. Ogyrfan was supposedly the father of Gwenhwyfar (Guinevere). To the north of this dyke-edged borderland, is the field (Maes Garmon), near Mold, linked by local – and it has to be said much disputed tradition - to the site of the famous ‘Alleluia Victory.’

This conflict is supposed to have been fought in 447, when the Britons were led by St Germanus. The site, now marked by an obelisk, lies at the western end of the easiest valley route (and classic ambush point?) through the Clwydian hills. Before the battle, the area immediately around Maes Garmon is said to have been ruled by an Irish pagan (or indeed a Pelagian) chief called Benlli - from his hillfort at Moel Fenlli, the second highest peak (1,876 ft) in the Clwydian Range.[26] This was also the name of Benlli Gawr, a giant of Welsh mythology. A large, late 4th century Roman coin hoard was found at this typically aligned Iron Age rampart-and-ditch fortress.It was one of the hill forts in the west that seems to have been reoccupied in the late/post Roman period.

It might be noted, in passing, that Wat’s Dyke - which has been confused with Offa’s Dyke in some places in this Flintshire region - runs a short distance away, to the east across the valley of the River Alyn. Nothing however exists here, in the annals or elsewhere, to link the dyke with that battle tradition. And tradition is no more than it might be: a ninth century attempt to connect the historic visits of Germanus with an existing local cult figure (St Garmon) in Wales. But it has to be said that not everyone agrees with that narrow interpretation.[27]

More recent on-the-ground explorations of neighbouring sites have to be considered. Extensive work has been carried out on Wat’s Dyke in recent years by David Hill, Senior Research Fellow in Anglo-Saxon Studies, at Manchester University. But he says that the earthwork is undated, except for the recent single radiocarbon date uncovered by the Shropshire archaeologists at Oswestry.

But he argues that the dyke must be later than the 6th century, and that its positioning suggests it was built in the 850s, when Gwynedd and Powys briefly became a unified state. The linking of hillforts along the dyke indicates, in his view, that the forts - possibly supported by beacon encampments - might have been used as bases by dyke patrols. But the remarkable lack of Anglo Saxon evidence from either Offa’s or Wat’s Dyke suggests, he says, that people were not settling or spending much time in what were then wild border zones.[28]

The enlarged Mercia (Mierce – boundary people) created by Penda (died 655) is set out in the tax list, probably of the late seventh century, and known as the Tribal Hidage. But it is argued that, when the area came under Mercian rule, Wat’s dyke had already been constructed and was functioning as a frontier, and that the renders in kind paid to the central court authorities (dues then approximating to then standard English units) were still commonplace in Shropshire at Domesday.[29]

For centuries after its construction Wat’s Dyke was a striking feature in the landscape but also an administrative boundary. In some areas, where it came into contact with lands owned by the Cistercians, it seems to have been considered to be the work of the devil – and for that reason it attracted a range of Christian crosses and symbols designed to counter the influence of the ‘unspeakable.’

So which devil - or topographical cythraul in Welsh - was being confronted here? Could it conceivably have been the key sub-Roman figure of Vortigern? Clearly, his name was not ‘unspeakable’ to the Welsh Powysian princes in the 9th century, who famously celebrated his role in their lineage on Eliseg’s Pillar - on a memorial that, for its time, must be without parallel in Britain.


  • Alcock, Leslie (1987): Economy Society & Warfare among the Saxons and Britons, University of Wales Press, Cardiff.
  • Archaeology in Wales (Roman period) Plas Coch, Wrexham, vol. 37, 1997.
  • Blake, Steve and Scott Lloyd (2000): The Keys to Avalon: The True Location of Arthur’s Kingdom Revealed. Element Books.
  • Blinkhorn, Paul (1999): Tolerating Pagans for The Sake of Trade, in British Archaeology, No. 44, May 1999, at: http://www.britarch.ac.uk/ba/ba44/ba44feat.html#blinkhorn.
  • Bu’lock, J.D (1972): Pre-Conquest Cheshire, 383 – 1066, Vol. 3, History of Cheshire, Cheshire Community Council, Chester.
  • Fox, Sir Cyril (1955): Offa’s Dyke: A Field Survey of the Western Frontier-works of Mercia in the Seventh and Eighth centuries. Published for The British Academy by Oxford University Press.
  • Gelling, Margaret (1992). The West Midlands in the Early Middle Ages, Leicester University Press.
  • Haigh, Mike: ‘ The Tale of Wade’ Northern Earth 66 (1996) http://www.danu.co.uk/ne/66/wade.html
  • Hannaford, H.R. (1997): Archaeological on Wat’s Dyke at Maes-y-Clawdd, Archaeology Service, Shropshire County Council, report no. 132., December 1997.
  • Higham, Nicholas J. (1991): Gildas, Roman Walls and British Dykes, in: Cambridge Medieval Celtic Studies 22, pp. 1-14.
  • Higham, Nicholas J. (1993): The Origins of Cheshire, Manchester University Press.
  • Hill, David (2000): ‘Offa Verses The Welsh’, in British Archaeology, Issue 56, December 2000, at: http://www.britarch.ac.uk/ba/ba56/ba56feat.html#hill.
  • Hunt, August (2001): Cunedda as Vortigern, Guest Articles, at: http://www.vortigernstudies.org.uk/artgue/guestdan3.htm.
  • Matthews, Keith J: Wat’s Dyke: a North Welsh linear Boundary: http://www.kmatthews.org.uk/arthuriana/wall_archaeology.html
  • Morris, John (1998): The Age of Arthur, Phoenix paperback edition.
  • Myres, J.N.L (1989): The English Settlements – Oxford University Press paperbacks.
  • Palmer, A.N. (1992): A History of the Parish Church of Ruabon and the Town Fields and Folk of Wrexham in the Time of James the First etc. A.N. Palmer (1847-1919), Bridge Books Wrexham.
  • Palmer, A.N. (1997): A History of the Town of Wrexham, 1893, Bridge Books, 1982/1997.
  • Vermaat, Robert (2000): Who was Vortigern, Vortigern Studies, at: http://www.vortigernstudies.org.uk/artwho/who.htm.
  • Vermaat, Robert (2001): Illumination 3 – Lambeth Palace Library MS 6 folio 43 verso, Vortigern Studies, at: http://www.vortigernstudies.org.uk/artlit/image3.htm.
  • Watson, Michael and Chris Musson (1994): ‘Timber Halls at Frogmore, Atcham’ and St Eata’s Church at, Atcham’ in Shropshire From The Air, Publishing Division of the Leisure Department, Shropshire County Council, 1993, reprinted 1994.
  • Webster, Graham (1986): The British Celts and their Gods under Rome, Batsford.
  • White, Roger & Philip Barker (1998): Wroxeter; Life and Death of a City, Tempus Publishing.
  • Williams, Sir Ifor (1980): The Beginnings of Welsh Poetry, Studies by Sir Ifor Williams, edited by Rachel Bromwich. University of Wales Press, Cardiff.
  • Worthington, Margaret (1997): Wat’s Dyke: An Archaeological and Historical Enigma, Bulletin John Rylands Library, Manchester, Vol 79, no. 3, 1997.


[1] Sir Cyril Fox, 1955.
[2] N.J. Higham, 1991.
[3] John Morris, 1998.
[4] A.N. Palmer, 1992.
[5] Sir Cyril Fox, 1955.
[6] Sir Cyril Fox, 1955.
[7] H.R. Hannaford, 1997. The report concludes: ‘The radio-carbon date obtained for the earthwork at Maes-y-Clawdd centres on 446 A.D. and probably lies within 411-561, at a time when this area may have formed part of a smaller, independent British kingdom whose capital was Wroxeter, the former civitas of the Cornovii. …. the radio-carbon date suggests that a 5th century date for the construction of Wat’s Dyke is more likely.. The dyke should therefore be regarded as being contemporary with that other great 5th century linear earthwork, the Wiltshire Wansdyke, rather than Offa’s Dyke and should be considered as an achievement of the post-Roman kingdom of the northern Cornovii rather than a work of 7th-8th century Mercia’.
[8] Steve Blake and Scott Lloyd, 2000.
[9] Steve Blake and Scott Lloyd, 2000.
[10] John Morris, 1998.
[11] Leslie Alcock, 1987. ‘It would seem reasonable to allow for occupation to be dated by pottery typology…beginning before 300 AD and continuing without a break until after 500 AD.
[12] H.R. Hannaford, 1997.
[13] Keith J. Matthews: Wat’s Dyke: a North Welsh linear Boundary The author is unequivocal in his rejection of the Maes-y-Clawdd, Oswestry, dating. ‘A date between c400 and c600 can be ruled out with some confidence….Wat’s Dyke is almost certainly not sub-Roman despite the statements being made on the back of a single radio-carbon date.’
[14] Margaret Gelling, 1992.
[15] Mike Haigh, 1996.
[16] Steve Blake and Scott Lloyd, 2000.
[17] Graham Webster, 1986.
[18] Sir Ifor Williams, 1980.
[19] Steve Blake and Scott Lloyd, 2000.
[20] A.N. Palmer, 1992.
[21] A.N. Palmer, 1997.
[22] Paul Blinkhorn, 1999.
[23] Lambeth Palace Library, MS6, folio 43v.
[24] Robert Vermaat, 2000 and 2001.
[25] Michael Watson and Chris Musson, 1994.
[26] J.D. Bu’lock, 1972.
[27]J.N.L Myres, 1989.
[28] David Hill, 2000. The author, argues here that, far from being supine victims of the 8th century Mercians, the Welsh were a major force, and that the Offa’s dyke was no symbolic boundary, but a defensive barrier and that the Welsh were ‘on the warpath against the English’, and often won. The dyke was nothing less than Offa’s Western Front.’ Hill also suggests that evidence from Wat’s Dyke (similar but better made, in his view, than the Offan earthwork) must be later than the 6th century.
[29] N.J. Higham, 1993. In ‘An English Empire: Bede and the early Anglo-Saxon kings’ (Manchester University Press, 1995). Higham discusses scholarly views now pointing to a Northumbrian, rather Mercian, origin for the Tribal Hidage – a list of payments or obligations owed to an ‘over king’ by his tributary kings–and that, of the Northumbrian candidates, the most likely monarch responsible for it is Edwin (r 616- 633). Edwin spent part of early life and manhood as a Christian prince in exile in Gwynedd. One of his residences would have been at the famous royal ceremonial centre at Yeavering (Ad-Gefryn – hill of the goats). Wat’s Dyke marked the western limits of the hidation levying territory in Mercia.

WAT’S  IN A NAME? is copyright 2001 Keith Nurse, used with permission.

Comments to: Keith Nurse

VortigernStudies and Wansdyke Project 21 are copyright Robert Vermaat 1999-2007.
All rights reserved