Heroic Age Logo The Heroic Age, Issue 1, Spring/Summer 1999  

The Age of Arthur:

Some Historical and Archaeological Background


Christopher Snyder,
Marymount College

The question "Did Arthur really live?" is often followed with the query "When did Arthur live?" According to the Annales Cambriae, the Historia Brittonum, and Geoffrey of Monmouth's History of the Kings of Britain, Arthur's activities belong to the fifth and/or sixth centuries AD. Most modern advocates of an historical Arthur place him within this period as well. While we may never be presented with historical proof of Arthur's existence, we can learn more about this period in which he may have lived. In recent years, both historians and archaeologists have shed new light on this murky and often overlooked era.

Naming the Period

First of all, a word or two needs to be said about the name for this period in British history. It is not Roman Britain: Rome was "officially" gone (i.e. the civil and military administration had withdrawn) by 410, despite lingering traces of Roman culture (Dark 1998). Nor is it rightly Anglo-Saxon England: the Britons, not the Germanic newcomers, were in control of most of the island (and were certainly the majority population) until the late sixth century. Calling it an "interlude" or a "transitional period" is also a mistake (though one many writers have committed), for it does a disservice to the fascinating evidence we possess and makes it even harder for us to understand how the legends of Arthur and Merlin could have emerged from it. Indeed, for many this is The Age of Arthur (Morris 1973).

In recent years, scholars have warned us against using the problematic Arthur to define an historical period (Dumville 1977; Thomas 1981). Most archaeologists have opted to use the label "Sub-Roman Britain," first used by ceramics experts to describe declining standards in Romano-British pottery. The problem with this label, apart from the obvious negative connotations, is that "sub-Roman" can mean anything from the first few decades of the fifth century to the entire fifth and sixth centuries. Post-Roman, Early Christian, and Early Medieval are similarly vague in chronology. I have proposed "The Brittonic Period" to cover the years from 400 to 600 (Snyder 1998). The main advantage of this label is that it focuses our attention on the Britons, who otherwise get lost in general surveys of Roman and medieval Britain.

That is not to say that the Britons are the only people of historic import in fifth- and sixth-century Britain. In the extreme northern parts of the island lived various tribes whom the Romans called collectively Picts. (The Latin Picti means "painted ones," perhaps referring to the practice-common among Iron Age Celts-of warriors tattooing their bodies.) These Celtic-speaking peoples, and their enigmatic stone symbols, are still little understood, despite a recent flourish of scholarly attention.1 On Britain's western coasts, especially Dyfed and Argyll, were groups of Irish settlers whom the Latin sources call Scots (Scoti or Scotti). These Irish brought to Britain the script known as Ogam (McManus 1991), founded the Déisi dynasty of Demetia (Thomas 1997), and carved out the kingdom of Dalriada, in Argyll, that would eventually come to dominate and give identity to medieval Scotland (Bannerman 1974). Finally, in the eastern lowlands that would come to be called England, were scattered groups of Germanic-speaking peoples from the Continent including Angles, Saxons, Jutes, Frisians, and Franks.2 Called simply Saxons by their Brittish neighbors, these peoples would coalesce into large kingdoms that would dominate much of the island and, eventually, merge and adopt a common "English" identity (Wormald et al. 1983).

The Written Evidence

All of these peoples raided Roman Britain in the fourth century, turning to settlement in the fifth and sixth centuries. Apart from some inscriptions, they left no written records of their own in the sub-Roman period. The very fact that our only native accounts of these centuries come from Britons is reason enough to view this as the Brittonic Period. The most important of these sources are the writings of Patrick and Gildas. Patrick, born in late Roman Britain but kidnapped and taken to Ireland by slavers, composed an autobiography called the Confessio as well as a letter, the Epistola, to a British tyrant named Coroticus.3 Gildas (fl. 500) wrote a substantial work called De Excidio Britanniae (On the Ruin of Britain), and two additional works, the Fragmenta and De Poenitentia (a monastic penitential), have also been attributed to him.4 Neither Patrick nor Gildas attempted to write history, thus their writings must be used carefully in the writing of history for the period.

There are other primary sources for the Brittonic period, but all written by outsiders whose knowledge of Britain varies. For the important events of 400-410, we have the observations of Claudian, Olympiodorus, Frigeridus (in Gregory of Tours), Sozomen, Orosius, Zosimus, and Procopius.5 Two anonymous Gallic Chronicles, written in 452 and 511, provide brief descriptions of Saxon raiding and conquest in Britain.6 The Life of St. Germanus, written by Constantius of Lyon in the late fifth century, describes two visits that the saint paid to Britain in 429 and c.445.7 The rest are later sources, like the Llandaff Charters, which describe land ownership in early medieval South Wales (Davies 1978); Bede's Ecclesiastical History of the English People (c.731); the Historia Brittonum (attributed to the Welsh monk Nennius) and the Annales Cambriae, both ninth-century compositions found together in a manuscript of c.1100; and a vast corpus of Welsh genealogies and saints' lives (Bartrum 1966; Doble 1984).


Another category of evidence, indeed a primary source for the period, is epigraphy. For the Roman period, inscriptions exist in many forms: stone, lead, silver, tile (Collingwood and Wright 1965). While not as numerous, inscribed stones for the Brittonic period have been found from Cornwall to Scotland (Okasha 1993; Allen and Anderson 1903), with Wales producing the most examples (Nash-Williams 1950). These inscriptions, mostly in Latin but occasionally (in Wales and Cornwall) also in Ogam, include personal names, names of districts, occupations, and of course religious formulae. Only recently have scholars begun exploring the potential of writing history with such sources.8


The fastest growing body of evidence for the period, however, is archaeological. I have, elsewhere, discussed in detail all of the British sites that have yielded substantial material evidence for activity in the fifth and sixth centuries (Snyder 1996 and 1997). It will suffice here to list the most important sites, by category. Most late Roman towns in Britain have revealed some evidence for continuity of occupation into at least the mid-fifth century (e.g., Carlisle, Catterick, York, Lincoln, Ancaster, Wall, Droitwich, Worcester, Gloucester, Caerwent, Bourton, Dropshort, Colchester, Cirencester, Chelmsford, London, Bath, Canterbury, Winchester, Shepton Mallet, Chichester, Ilchester, Dorchester, and Exeter), while at others, most notably Silchester and Wroxeter, occupation continues well into the sixth century and beyond. The eastern towns, especially those in the Humber and Thames river valleys, were the first to fall (though perhaps more to evacuation than destruction), while those of Wales and the Southwest were held the longest by Britons.

Occupation also continued within many Roman forts and in the small towns (vici) which often developed near them. Binchester, Ravenglass, Catterick, Chester, Caernarvon, Aberffraw, and Caerleon all show signs of continuity. Of the Saxon Shore forts, Richborough and Portchester have yielded the strongest evidence of occupation in the fifth century, while the former appears to have sheltered a Christian community. Hadrian's Wall has provided even stronger evidence of continuity, though building in timber was apparently replacing stone masonry. Vindolanda and South Shields may have had sub-Roman churches, while Castlesteads, Housesteads, Chesters, Corbridge, and Benwell have revealed slighter traces of sub-Roman activity. At Birdoswald, two Roman granaries were converted to other uses in the fifth century, including a new timber feasting hall.

Other Roman sites occupied in the fifth and sixth centuries include a few rural villas (most notably Glan-y-Mor, Frocester, and Bancroft) and pagan temples (West Hill Uley, Chelmsford, Nettleton, Bath, Lamyatt, and Maiden Castle), though many of the latter appear to have been converted into Christian shrines and churches. We don't get a good picture of the rural landscape in this period, and most of the small simple farmsteads go unnoticed. What we do notice is the rural defended places, the hillforts, promontory forts, Cornish rounds, and crannogs. The most prominent of these are Dumbarton, Doon Hill, and Yeavering (in Scotland), Degannwy, Dinas Emrys, and Dinas Powys (in Wales), and Cadbury-Congresbury, Glastonbury Tor, South Cadbury, and Tintagel (in the Southwest). Most of these sites have yielded good evidence of long-distance trade and military engineering along with more domestic activities.

Cemeteries are a very promising avenue for further investigation. Large late and sub-Roman cemeteries have recently been excavated at Ancaster, Brean Down, Shepton Mallet, Cannington, and Poundbury. Here we see changing burial practices among the Britons that may, in some cases, signal the conversion of these communities to Christianity. Apart from many portable objects decorated with Christian symbols, the material evidence for Christianity in late and sub-Roman Britain is meager. The best candidates for churches are the small basilicas excavated at Colchester, Richborough, Silchester, Canterbury (St. Pancras), and Icklingham. Other urban Christian communities are suggested at Carlisle, Worcester, Cirencester, St. Albans, Wells, and Shepton Mallet. Rural Christian sites, many of which were hermitages and monasteries, include Ardwall Isle, Caldey Island, Lundy Island, and St. Helens in the Scillies. The extensive and recently concluded excavations at Whithorn have revealed evidence for a thriving early Christian community, from the late fifth century to the Viking Age, where Roman technology was used by Britons bearing Latin names-north of Hadrian's Wall (Hill 1998)!

The final category of material evidence is hoards of precious-metal objects. In the late Roman period, currency entered Britain in the form of coins, bullion, and ingots. Coin hoards are found throughout the Roman period and are usually associated with a political crisis. The ultimate political crisis was that of 410, so many hoards have been found containing coins issued by the emperors Honorius, Arcadius, and Constantine III in the first decade of the fifth century. While coins later than 410 are extremely rare, the recently discovered Patching Hoard in West Sussex contained many Roman coins (and Visigothic copies) datable to the mid-fifth century. The next most frequently hoarded object was silver wares. A hoard of silver spoons bearing Christian symbols was found at Canterbury, while beyond the frontier, at Traprain Law, a hoard of cut and folded silver plate indicated a possible payment to a Roman ally.


Small finds of portable objects are often the best indicators of date for the Brittonic period. For while radiocarbon and other scientific methods can yield accurate dates, often there is no trace of organic material to extract samples. Coins, of course, give us very precise dates, but only the terminus post quem, the date after which an associated structure may have been built or used. Moreover, imperial mints ceased operating in Britain in the late fourth century, and regular shipments of coin (to pay the troops and administration) stopped arriving in the first decade of the fifth century. After 410, a few coins circulated but mostly a barter economy prevailed. This should not imply primitive, however, for there is increasing evidence that the Britons carried on an active trade with Gaul, Spain, and the Eastern Mediterranean in the late fifth and sixth centuries, bringing in such exotic imports as olive oil, wine, fine table wares, and glass drinking vessels (Thomas 1990; Campbell 1996). The fine wares, glass, and amphoras (which contained the wine and oil) can be dated with some precision, and thus their presence at sites like Tintagel and Whithorn have helped us to talk with confidence about these sites as thriving British settlements in the fifth and sixth centuries.

Historiographic Trends

Until fairly recently, the student of fifth- and sixth-century Britain had to rely on large surveys of the Roman and Anglo-Saxon periods, which occasionally offered a few words on subjects like Arthur and the adventus Saxonum (coming of the Saxons) (e.g. Collingwood and Myres 1936). Indeed, many scholars still treat the period as an appendage to either Late Roman Britain (e.g. Esmonde Cleary 1989; Jones 1996) or early Anglo-Saxon England (e.g. Arnold 1984; Higham 1994). In the early 1970s, however, two well-respected scholars, one an historian and the other an archaeologist, made this period the sole subject of two ambitious works (Morris 1973; Alcock 1971). Both men saw Arthur as the key figure in the period, and provided the king with a respectable historical backdrop.

The reaction was polemical. The general public read these works with great enthusiasm, and gave a boost to the tourist industry in Arthurian locales like Glastonbury and Tintagel. Academic skepticism, however, was given voice in a provocative essay by the Cambridge textual historian David Dumville (Dumville 1977). Dumville called both Morris and Alcock to task for placing too much emphasis on Arthur, a figure whose historicity was at best debatable. Morris in particular was criticized for basing his history on ambiguous Celtic "traditions," while both men were using much later sources to construct the history of the fifth and sixth centuries. Dumville, followed later by Charles Thomas (Thomas 1981), called for scholars to remove Arthur from their histories and above all from the titles of their books.

Dumville's hyper-criticism was felt far and wide. Very few academic historians ventured into this territory in the years following his essay, while many archaeologists used Dumville's skepticism to justify their own innate distrust of the written sources. Meanwhile, the quest for the historical Arthur hardly slowed down in the popular market. The Glastonbury historian Geoffrey Ashe produced the most noteworthy of these books, offering the intriguing possibility that the historically attested British ruler Riothamus may have been the real Arthur (Ashe 1981 and 1985). Since then, many writers have offered alternative candidates, ranging from the Roman commander Lucius Artorius Castus to the Welsh prince Owain Ddantgwyn (Littleton and Malcor 1994:62-63; Phillips and Keatman 1992). Merlin has not fared as well, being the subject of only one substantial work (Tolstoy 1985), though the prolific Norma Goodrich has produced lamentable "studies" of Guinevere and Lancelot as well as books on the magician and the king.

The scholarly community has taken little notice of these popular developments. Arthurian skepticism has prevailed throughout the academy in the last two decades. Still, a few important works have been produced drawing our attentions back to non-Arthurian questions of the fifth and sixth centuries. The late Edward Thompson was one of the rare historians to take on the textual materials of the fifth century, producing provocative studies of both Patrick and the events surrounding Germanus' visits to Britain (Thompson 1984 and 1985). The archaeologist Charles Thomas gave us an exhaustive look at British Christianity in the fifth century (Thomas 1981), while another prolific archaeologist, Nick Higham, weighed in with a controversial history of the same century claiming that Badon was actually a British defeat and that Gildas was writing under Saxon dominance (Higham 1994)!

Higham, Simon Esmonde Cleary, and Michael Jones have all concentrated on the fifth century in attempts to explain, essentially, the collapse of Roman Britain and the creation of Anglo-Saxon England. While these works have much to offer, including new theories for the adventus Saxonum,9 none devote much attention to the culture of the Britons per se. Two very recent books have done this, however, treating the period as a separate entity and the Britons as the dominant culture. Ken Dark's lengthy treatment is essentially founded on the premise that Roman Britain never really ended, and that the post-Roman Britons carried on many Roman-period traditions for several centuries after 410, when Roman civitates were transformed into British kingdoms (Dark 1993). My own study portrays the Britons as a unique hybrid of Celtic, Roman, and Christian cultures, looking both at the terminology used by Britons like Patrick and Gildas and at the extensive archaeological evidence to date (Snyder 1998).

With so many substantial works produced in the last ten years, there is good reason to be optimistic about this field. The period of great skepticism seems to be receding, and now writers are feeling more comfortable expressing opinions about the fifth and sixth centuries using both written and material evidence to produce their histories. I doubt that we will ever be able to produce a reliable narrative history of the period, but we can look at the culture of the Britons in many new ways, as, for example, more studies of their inscriptions come to light, and as we begin to understand their architecture better. I predict that early in the next century we will express more confidently the reasons for the decline of Roman Britain and the adventus Saxonum. And, hopefully, the Britons will make their way into general histories of the Middle Ages alongside their English and Frankish neighbors.

Whether or not Arthur is ever proven to have been an historical personage, no one can doubt that the fifth and sixth centuries, the Brittonic period, played an important role in shaping his legend. Admirers of the Arthurian legends can do no harm to the story by attempting a better understanding of this age, while those who follow the historical Arthur trail can learn some important lessons about sources and methodology by reading the sober texts of historians and archaeologists working in this field. Perhaps by learning from one another enthusiast and academic can once again investigate an Age of Arthur.

Notes References Next
Return to Table of Contents   Return to Homepage