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

  Vortigern Studies > Wansdyke > Articles > Grigg

Guest Author:
Erik GriggErik Grigg

Erik Grigg is a history graduate and freelance writer with a passion for Early Medieval history. He has visited countless Dark Age sites across the country dragging his poor family behind him! His biggest claim to fame was his discovery of the skull of an Iron Age bog body near Lincoln in 2003. Read more of his Dark Ages' articles on his website.

Vortigern Studies Index


.Wansdyke Project 21
is part of
Vortigern Studies



Articles about Wansdyke
click here


Dark Age Dykes
Erik Grigg

All across Britain there are earthworks that are thought to have been built in the Dark Ages which have rarely been analysed, which is rather odd as early Medieval Britons are archaeologically often invisible. With many historians shying away from the term ‘Dark Ages’ because of its negative connotations I wonder if the period may be more accurately be named the ‘Dyke Ages’ as so large and impressive are these structures. Some, however, may not all date to this period as the lack of documents from this time period makes it very easy to ascribe any mysterious boundary to this time especially as linear boundaries are difficult to date through archaeology as domestic rubbish, burials and building foundations which usually provide good buried dating evidence are rarely found associated with them. They are a widespread phenomena in Britain and as such they are worthy of closer scrutiny as they may be able to tell us more about the politics, economics, methods of war, culture and population levels of Dark Age societies. No historian has ever given an extensive list of these moments (though scraps of pottery or the odd coin can stimulate reams of speculation) so I give a brief gazetteer below. I do not believe the list is exhaustive or the accompanying descriptions completely error free so please do not hesitate to contact with any fresh insights about these mysterious barriers. With some of the dykes the only information I have has been taken from old or extremely basic local guide books so the information I have is very patchy (in an ideal world I wouldn’t have even started this article without another six months full time study!)

Norfolk has a series of Dark Age dykes: Fossditch, Bichanditch, Launditch, Panworth Ditch and the Devil’s at Garboldisham. Launditch and Panworth both face westward and cut Roman roads (therefore they must post-date 407AD). Peter Wade-Martins thought they were built by the Britons against the Anglo-Saxons in the Wash area and the dykes were patrolled by cavalry
[1]. He claims Fossditch and Bichanditch were built by the Anglo-Saxons as they face east and adjoin fenland where the Anglo-Saxons could have first landed (early –ham places names seem to be concentrated behind the ditches). Bichanditch, however, does face Swaffham which was an early Anglo-Saxon site. Dymond postulates that Bichanditch was built to defend Britons against Germanic, and specifically Swabian, settlers in Swaffham[2].

The Roman town of Silchester is surrounded by a series of earthworks, but some of these are Iron Age in date built to protect the pre-Roman settlement. Some of the dykes do cut Roman roads leading to the town from the Thames Valley (where there is evidence of early Saxon settlement). The city should be bisected by the northern border of Hampshire, but the border bulges out in a semi-circular salient following the dykes. This salient is made up of the two parishes of Silchester and Mortimer West End which form a circular piece of territory 3 miles in diameter centred on the city and may be a Roman urban district boundary that has influenced the layout of Saxon parishes.

Just to the east of Housesteads fort on Hadrian’s Wall is a north-south earthwork with a ditch on the western side called Black Dyke which is traditionally thought to be the border between the English kingdom of Bernicia and the British kingdom of Rheged. On older maps it runs from northern Northumbria to Allenheads in Durham, but now it just runs from the North Tyne at Tarset to Moraleeon the South Tyne

Between Galashields in the Borders and Peel Fell is a dyke called the Cattrail, which is 45 miles long, 26 feet wide and with a bank on each side at least ten feet high. As it divides the coastal area from inlands parts it may have been built by the Britons against the English kingdom of Bernicia.

Grey Ditch in the Peak District is a mile long bank with a ditch on the south side. The area to the north of the ditch seems to show signs of British survival and the area to the south signs of early English settlement so it has been seen as a British built. Barnatt and Smith say of Grey Ditch: “This large linear bank and ditch cuts right across the main valley which provides a route between the Hope valley and the fertile areas of the plateau to the south. There are two other short linear ditches in the region which could also been seen as separating Anglian and British communities”
[4]. Map reference SK 173/818 to SK 184/813.

Cornwall has two dykes that are considered to date from this period: Bolster Bank at St Agnes (which is visible between Trevaunance Cove and Chapel Coombe) and the six mile long Giant’s Hedge between Lerryn and West Looe (best seen north of Lanreath). The Map reference for the Giant’s Hedge is SX 245/535 to SX 150/569. Both have been interpreted as Cornish defences against the incursions of Wessex.

There are two distinct dyke groups that have traditionally been seen as defences put up to save this British kingdom from English attack. Grim’s Ditch runs for five and a half miles north-south from near Willington to Whinmoor. It faces east and protects the Leeds area from attack from the Humber area. The bank is still in places eight feet high and the ditch 30-40 feet wide, though housing developments have destroyed much of the northern part. The other set of dykes are called the four and a half mile long Aberford Dykes and they are a varied group of three distinct dykes: The Ridge (also called Becca Banks), South Dyke and The Rein. The Ridge is three miles long, faces south (towards where Elmet traditionally would be) and runs from Barwick to east of Aberford. The Ridge is still 25 feet from the bottom of the ditch to the top of the bank. The Rein also faces south and is a mile long, but is probably an Iron Age boundary. The South Dyke is 1000 yards long and faces north facing the direction of York.

I have read that this part of Scotland was a Pictish enclave surrounded by the British kingdoms of Rheged and Strathclyde that built a long linear earthwork to protect itself. The idea of a southern Pictish kingdom in Galloway sounds a little dubious to me, but may help explain how the Picts raided the west coast and how St Ninian converted Picts from Withorn.

Cambridgeshire has four very impressive dykes: Brant Ditch, Brent Ditch (which faces to the north-east whereas the others all face south-west), Fleam Ditch (which means fugitive’s ditch) and the huge Devil’s Ditch (the inner ditch if this is a system of ditches built at the same time). Excavations in the 1970s failed to provide any dating material, but did show the ditch was built by a series of gangs (the construction was not uniform along the length of the ditches) and while the ditch at Devil’s Ditch was never cleaned out after construction, Fleam Ditch was regularly cleared. Devil’s Ditch is seven miles long and cuts off East Anglia from Cambridgeshire. There have been many explanations as to who built them such as the kings of East Anglia built them to prevent attacks from the Britons in the 6th century or against attacks from Mercia in the 6th/7th Centuries. They are mentioned in the Anglo-Saxon Chronicle in 905 when it is said King Edward harried the Vikings between the Dykes and the River Wissey (Ouse?) so they must pre-date the 10th Century. Bran is three miles long, Brent under two, Fleam in two sections about six miles in total and Devil’s is about 6 miles long.

Between Reeth and Grinton in the Yorkshire Dales are a series of dykes that have been traditionally been seen as defences set up in 70 AD against the advancing Romans, but could easily been erected by the rulers of Rheged against the Northumbrians.

[7] argued that the nearby linear boundaries are Dark Age and demarking a sub-Roman enclave based on the town deep in the territory of the East Saxons.

Dark mentions a Clawdd Mawr in Dyfed. There are four other possible Dark Age Welsh dykes: Clawdd Mawr (near Penybontfawr/Llanwddyn in Powys), Crugyn Bank (a one and a half mile long dyke to the south of Dolfor, Powys), Short Ditch (near Knucklas, Powys) and Giant’s Grave, a dyke about 250 metres long south east of Llandinam which has had deposits underneath it carbon dated to 340-530 AD (there is a short report in Archeaology of Wales Vol. 43, p77). The Clywd-Powys Archeaological Trust is investigating the last four short dykes and their report is due in March 2006.

Dorset has three north facing linear earthworks that seem to be designed to protect the Britons in the area from the expanding kingdom of Wessex. The most northerly is called Bokerley Dyke which was partially built on top of a series of earlier earthworks called Grim’s Ditch (which were probably Iron Age field boundaries built to divide large cattle grazing areas). Bokerley still defines the Dorset-Hampshire border and seems to protect the Romano-British settlement at Woodyates against attack from Hampshire. Coins from the reign of Valens (364-78) and Honorius (393) found during excavations of the dyke suggest it may date to the early 5th century. It is over three miles long and when I last visited it was in many places taller than myself (that is about 6 feet or more) so it was more than a mere hedgerow, but Bowen
[9] and Dark thought that it was a minor boundary. Dark describes it as a: “heightened, earlier boundary bank of potentially, only local significance”[10]. Map reference SU 025/2000 to SU 063/168.

The middle ditch in Dorset is called Combs Ditch and is situated just to the south-west of Blandford Forum. In parts it is 9 feet from the top of the bank to the bottom of the ditch. At present it is about 2 miles long and like Bokerley the ditch is the north of the bank. Map reference ST 853/022 to SY 887/995 (the southern end is indistinct and that last reference may be wrong).
The final linear Barrier in Dorset is Battery Banks which runs west of Wareham along a ridge between the rivers Frome and Piddle to cut off the Isle of Purbeck from the north. There is also a cross ditch to prevent invaders attacking along the ridge. It is over four miles in length from the western most part to the most easterly part, though it cannot be traced at all points in between.

We can imagine these defensive lines being patrolled by a small group of cavalry based in settlements nearby (Woodyates for Bokerley, possibly Winterborne Whitchurch for Combs and Wareham for Battery Banks). These garrisons need only be a dozen strong as all three lines are quite short and have great views for miles to the north. The patrols could sound the alarm using horns, fires or an errand rider and hold the defences until the local militia turned up. Due to the superb views to the north at Bokerley the settlement of Woodyates is closer to the defences than the raiders would be when they came in sight of a horseman standing on the dyke so the local farmers would have time to grab their spears and get to the dyke before the enemy could. Perhaps it was such garrison that provided the stimulus for the growth of Wareham (popular opinion supposes Wareham to be Roman, but no substantial Roman finds have ever been unearthed there). Employing full time mounted troops would have been a strain on the local rulers, but if it meant that cattle were not stolen and crops burnt could have made it very economic (I bet the raiders would have soon turned their attentions elsewhere), though it would have taken a major organisation of resources to build them in the first place. Map reference SY857/886 to SY 911/874.

Wansdyke is one of the most famous Dark Age dykes, but it is in fact two earthworks (the piece that used to be believed connected the two pieces is now known to be a Roman road). The western half extends nine miles from Dundry Hill south east of Bristol to just south of Bath. The eastern half runs ten miles from south of Calne to Great Bedwyn on the edge of Savernake Forest. Both halves could have been built by the Britons, perhaps originally by Ambrosius, against the Saxons of the Thames Valley. The enigmatic Battle of Fethanleag or Battle Wood between the West Saxons and the Britons recorded in the Anglo-Saxon Chronicle could report a British victory in a campaign along this border as the Saxon king turned back to his territory ‘in anger’. Alternatively either could have been built by the kings of Wessex against the might of Mercia, perhaps around 628 when a battle is recorded against the Mercians at Cirencester. The former is more likely as the name of the dyke builder was forgotten and it was named after the Germanic god Woden when the Saxons of the area still named things after pagan gods and Wessex was converted to Christianity in 628. Its existence is implied by a charter dated to 941 when a hillfort adjoining the dyke which is now called Stantonbury Hillfort is named Merces Burh or border fort. Ian Burrow looked at two hillforts along the western part of the dyke that have been incorporated into its structure (Stantonbury and Maes Knoll) and concluded that the builders used the hillforts to save on labour, but did not refortify the forts, garisson them or patrol the dyke as a walkway through the ramparts was not built. He concluded: “that the dykes were simply an emphatic expression of territoriality.”

This famous linear earthwork marks the Welsh border and as the bank is on the east and the ditch on the west it was built against the Welsh, probably by the Mercian King Offa (who ruled 757 to 796). It was Asser, King Alfred’s friend and biographer, in the ninth century who first ascribed the dyke to Offa. Offa’s Dyke is newer (so we think we know who built it and when), it has been more heavily studied than most other dykes and is larger so more of it survives so we know more about it than other linear boundaries. Sir Cyril Fox
[12] carried out many surveys on the dyke in the 1920s and said it ran 149 miles from sea to sea, though only 80 miles were built (the gaps being covered with thick forest on the heavy clay soil). In parts the bank is still 8 feet or 2.5 metres high. A recent book[13] claims it was built by the Emperor Septimus Severus about 200AD, but this is probably not correct as Roman deposits have been found sealed under the dyke (at Ffrith) and the sources of the story of Severus building a turf wall in Britain are probably confused references to him rebuilding the Antonine Wall. Bede, who realised that Gildas’ date for the Antonine Wall was too late, found a reference in the life of Severus by Aelius Spartianus. This described Severus building a wall across the island from sea to sea and, presuming the turf wall was the older, Bede assumed this referred to Hadrian's Wall. Aelius Spartianus also wrote a biography of Hadrian which Bede obviously did not read or if he did he didn’t believe (these two biographies are part of the Augustan History that historians now believe is all the work of the same person so ‘Aelius Spartianus’ may be a pseudonym). Severus fought against the Picts and would have refurbished the Antonine Wall and would have no reason for building a barrier in the middle of the Roman province of Britain.

A guide written by two historians who have spent decades surveying and digging the monument[14] claims that the dyke is only 64 miles long from Rushock Hill (SO 300/595) north to Treuddyn (SJ 268/577) and lots of the parts marked by Fox do not exist and dismiss the idea that there was heavy forest on the clay soils that mark the ‘gaps’ (the Domesday Book shows Herefordshire as comparatively unwooded). Fox was confused with other smaller dykes (Whitford Dyke, Rowe Ditch, Lyonshall Bank and Scutchditch among others) that could date from prehistory to Tudor times and Wat’s Dyke in the north. Hill and Worthington argue quite convincingly that it was not a border between the English and the Welsh, but a border between Mercia and Powys (which would explain the northern and southern gaps as they cover the borders of Gwynedd and Gwent). We know from the Pillar of Eliseg that a Powys king who was contemporary to Offa was taking land from the English. It was used to be thought that the dyke was purely symbolic with large gaps in it to allow trade with the Welsh, but many suspected gate sites have been dug and no gaps in the ditch have ever been found. Hill and Worthington say the dyke was built by different gangs (as in Cambridgeshire) and probably patrolled (which would take 300 foot soldiers or fewer on horseback), but not garrisoned. They postulate a system of fortified villages (there are lots of place names of villages near the rear of the dyke that suggest they were enclosed and may be militia muster points) linked by beacons though a reviewer on the Clywd-Powys Archaeological Trust website say they have rather stretched the evidence.

The Burghal Hideage, an Anglo-Saxon document that dates to about 900AD, implies that one man is needed to maintain (or possibly build) just over 4 foot of defences (1.26 metres) so it would have taken about 75,000 men to build it or fewer if built over a number of years. A man could no doubt construct more than 4 ft of defences in a year, but this calculation is probably based on what a man can do in his spare time who is engaged in agriculture full time.

The Tribal Hideage, an eighth century document, says the north Mercians had 7,000 households and the south Mercians 5,000, so Offa must have conscripted labour from the other tribes that made up Mercia as well as kingdoms that accepted his overlordship.

Wat’s Dyke was thought to have been just over 20 miles long, but is now thought to run 38.6 miles from Lower Morton (SJ 305/233) to Basingwerk (SJ 195/775) and incorporates the hillfort at Oswestry. Fox dated it to the reign of Aethelbald (716-57), but a recent radiocarbon date suggests the late fifth or early sixth century, so perhaps it was built against the kingdom of Gwynedd under the powerful king Maelgwn.


Dark is one of the few historians to discuss these vast monuments and says they show the power of Dark Age British kings to conscript labour, especially in the Somerset area[15]. He suggested the labour may have been organised using service renders in lieu of money taxation which had ended when the widespread use of coinage ended in the early 5th century in much the same way that the Romans organised the Britons to work on the maintenance of Hadrian’s Wall in the late Roman period. He also obliquely suggests the West Country dykes may have been used by the local royal families to control trade.

The construction of the dykes does show there was a vast amount of labour available (which goes against the theory of a mass depopulation in the immediate aftermath of the fall of Roman Britain as expounded by historians like Dark). There are at least 240 miles (or 385 km) of dykes in Britain that probably date to the Dark Ages and if the Burghal Hideage is correct to think that a man could be expected to construct 4 foot and one and a quarter inches of fortifications in a year that is nearly a third of a million people needed to maintain and construct them (if they are each built in a single year). This shows borders were stable long enough both to be worth delimiting and to allow the labour to be conscripted, organised and the dyke built. There seems to have been conflict (anything from mass invasion to cattle rustling) or the fear of conflict among many societies so the belief that the Dark Ages were peaceful seems to be false. Both Anglo-Saxon and British monarchs seem to have built dykes so both cultures were copying tactics off each other and off the Romans (who famously built linear fortifications). They are named after the Devil, giants and pagan gods so obviously the proponents of the continuity school of Dark Age history need to explain how the undisturbed local peasantry who obviously had to build the things so quickly forgot who had ordered the building these vast structures.

Many of these dykes have been categorised in the past as the border between Celtic/British kingdom A on one side and Anglo-Saxon kingdom B on the other, but this may be overly simplistic. Many of these dykes could have been built between kingdoms where the monarchs on both sides thought of themselves as English or they could be older than we think and they could be borders between British kingdoms. Even if the king on one side spoke English and worshiped Woden that does not necessarily mean that all his subjects had ancestors from northern Germany. These dykes do influence cultural patterns as the earliest mass produced Anglian pottery (Illington-Lackford pottery) does seem to have a distribution pattern: “bounded by three early medieval linear earthworks”[16]. These dykes may, however, have hardened cultural divides in Britain and may be one of the reasons that the English language adopted so few words from the older British language. Two genetic studies[17] both show that even today genetic signatures differ between central England and Wales and we know the nations either side of Offa’s and Wat’s dykes show linguistic and cultural differences even today despite the fact that the border has not been a political one for about 700 years.

Apart from Offa’s Dyke and possibly some of the Cambridgeshire dykes there is a good case for assigning most of the as being of British construction. Perhaps this tendency for the native Britons to adopt defensive tactics against an aggressive invader is why the Britons could never defeat the initially far fewer invading Anglo-Saxons and why the Anglo-Saxon invasion stopped at the Welsh border when the English adopted these defensive techniques.

How these dykes worked as military structures (or even if they were military structures) is open to debate. Some may have been garrisoned and patrolled, but others seem not to have been so to generalise may be dangerous. Some could have been purely symbolic borders or places where the movement of goods was controlled (like the Salt Tax Hedge in India, a purely economic border between two areas both under British control showing a barrier does not have to be a political border). They seem to have been based on military fortifications and so at the very least were used to deter invaders, perhaps being designed to slow down small, mobile raiding parties to stop them carrying off goods or cattle across open country (most of the dykes stop at heavy woodland or marshy valleys).

This rash of dyke building may have been initially inspired by Hadrian’s Wall (Gildas dates its construction far too late, perhaps being confused by stories of a final refurbishment of the wall in late Roman times or possibly being confused by stories of dyke building by the Britons after the end of Roman Britain) or by the rash of hill-fort refurbishment that went on in this period. They need further study: excavations to look for dating evidence, surveys to find out how many man hours were needed to build them and serious research into the origins of their names. Perhaps these enigmatic structures could tell us so much more about this crucial period of British history.


[1] Wade-Martins, P. (1974): ‘The Linear Earthworks of West Norfolk’, in: Norfolk Archaeology Journal XXXVI pp. 23-38.
[2] Dymond, D. (1985): ‘The Norfolk Landscape’.
[3] Embleton, R. & F. Graham (1984): ‘Hadrian’s Wall in the days of the Romans’, esp. pp. 126-7.
[4] Barnatt, J. & K. Smith (1997) : ‘The Peak District’, esp. p. 53.
[5] Further details can be found at: http://www.hjsmith.clara.net/3943.htm.
[6] Any information on an earthwork in this area would be gratefully received.
[7] Basset, S. (1989): ‘In search of the Origins of Anglo-Saxon kingdoms’, esp. pp. 25-6.
[8] Further details can be found at: http://www.cpat.org.uk/projects/thisyear/progress.htm#929.
[9] Bowen, H.C. (1990) 'The archaeology of Bokerley Dyke'
[10] Dark, K. (1994): Civitas to Kingdom, British Political Continuity 300-800, Studies in the Early History of Britain, esp. p. 117.
[11] Burrow, I. (1981): ‘Hill-forts after the Iron Age: the relevance of surface fieldwork’ in G. Guilbert (ed) ‘Hill-fort Studies’, esp. p. 145.
[12] Fox, C. (1955): Offa's Dyke, a field survey of the western frontier-works of Mercia in the seventh and eighth centuries AD.
[13] Blake, S. and S. Lloyd (2000): The Keys to Avalon: the true location of Arthur’s Kingdom revealed. Shaftesbury.
[14] Hill, D. & M. Worthington (2003): ‘Offa’s Dyke: History and Guide’.
[15] Dark, K. (1994): Civitas to Kingdom, British Political Continuity 300-800, Studies in the Early History of Britain, p. 125 and p. 150.
[16] Hamerow, H. (2005): ‘The earliest Anglo-Saxon kingdoms’ in: Fouracre (ed): ‘The New Cambridge Medieval History: c500-c700 AD’, p. 279.
[17] Weale, Weiss, Jager, Bradman and Thomas (2001): ‘Y chromosome evidence for Anglo-Saxon Mass Migration’, in: Molecular Biology and evolution 19 (7) pp. 1008-1021 and Goldstein Goldstein, prof. D. et al (2003): ‘A Y chromosome survey of the British Isles’. Current Biology Vol. 13 (11) pp 979-984.


  • Barnatt, J. & K. Smith (1997) : ‘The Peak District’.
  • Basset, S. (1989): ‘In search of the origins of Anglo-Saxon kingdoms’, in: Basset, S. (ed) (1989): ‘The Origins of Anglo-Saxon kingdoms’.
  • Blake, S. and S. Lloyd (2000): The Keys to Avalon: the true location of Arthur’s Kingdom revealed. Shaftesbury.
  • Bowen, H.C. (1990) 'The archaeology of Bokerley Dyke'.
  • Burrow, I. (1981): ‘Hill-forts after the Iron Age: the relevance of surface fieldwork’ in G. Guilbert (ed) ‘Hill-fort Studies’.
  • Dark, K. (1994): Civitas to Kingdom, British Political Continuity 300-800, Studies in the Early History of Britain.
  • Dymond, D. (1985): ‘The Norfolk Landscape’.
  • Embleton, R. & F. Graham (1984): ‘Hadrian’s Wall in the days of the Romans’.
  • Fox, C. (1955): Offa's Dyke, a field survey of the western frontier-works of Mercia in the seventh and eighth centuries AD.
  • Goldstein, prof. D. et al (2003): ‘A Y chromosome survey of the British Isles’. Current Biology Vol. 13 (11) pp 979-984.
  • Hamerow, H. (2005): ‘The earliest Anglo-Saxon kingdoms’ in: Fouracre (ed): ‘The New Cambridge Medieval History: c500-c700 AD’.
  • Hill, D. & M. Worthington (2003): ‘Offa’s Dyke: History and Guide’.
  • Moxham, R. (2001): ‘The Great Hedge of India’.
  • Wade-Martins, P. (1974): ‘The Linear Earthworks of West Norfolk’, in: Norfolk Archaeology Journal XXXVI pp. 23-38.
  • Weale, Weiss, Jager, Bradman and Thomas (2001): ‘Y chromosome evidence for Anglo-Saxon Mass Migration’, in: Molecular Biology and evolution 19 (7) pp. 1008-1021. A presentation of the above source can be found at: http://www.ucl.ac.uk/tcga/presentations/ASdemo/AS-26-11-03b.html

Dark Age Dykes is Copyright 2006 Erik Grigg, used with permission.

Comments to: Erik Grigg

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