King Arthur: the Making of the Legend 
by Nicholas Higham.
Yale, 380 pp., £25, October 2018, 978 0 300 21092 7
Show More
Show More

Modern academic historians​ want nothing to do with King Arthur. ‘There is no historical evidence about Arthur; we must reject him from our histories and, above all, from the titles of our books,’ David Dumville wrote in 1977; and he was backed up by, for instance, J.N.L. Myres in 1986: ‘No figure on the borderline of history and mythology has wasted more of the historian’s time.’ In his new book, Nicholas Higham cites neither opinion but certainly knows of them, and indeed, in the end, agrees with them (except about book titles). Still, whatever historians may say, legends of King Arthur have remained deep-rooted in popular imagination, giving rise to whole libraries of fiction, and one Hollywood movie after another. Higham’s aim is, first, to demolish the wild theories that have gained currency in recent decades, and then to scrutinise the very limited evidence from early times and show its inadequacy. A question which remains unresolved is quite what powers the continuing appeal and often passionate adherence to the legend. So much smoke, there must be a fire burning somewhere. But one might well think that, nowadays, it’s recent politics that supplies the match and blows the flame, and not anything to do with the history of the Dark Ages.

A few conclusions remain deducible. It is about as certain as can be that the name Arthur derives, not from the Welsh word for ‘bear’, arth, but from the name of a Roman gens or family, Artorius. It is one of a whole string of Roman names repronounced by Welsh speakers in the post-Roman era, such as Emrys/Ambrosius, Custennin/Constantinus, Aergol/Agricola. That at least gives Arthur a historical setting, Britain in the fifth and sixth centuries, when Latin still had prestige but few people spoke it much any more. It seems to have been quite a popular name for a while, borne by several young royals from Ireland and Celtic Britain in the sixth and seventh centuries, which suggests that they may all have been called after someone who had made a name for himself not too long before. But who was he? And was he Roman, or British, or something in between?

The major candidate is Lucius Artorius Castus. His career is recorded on two inscriptions found in Croatia, which Higham carefully edits, translates and comments on. He seems to have been an interesting man, a career soldier who served as centurion in four Roman legions, rose to the rank of primus pilus in Legion V (Macedonica), was camp prefect for Legion VI (Victrix), was put in charge of the fleet at Misenum in southern Italy, and became procurator of Liburnia, where his inscriptions are: all of which seems a long way from Britain. But Legion VI was based at York and Roman soldiers moved about. The longer inscription also declares that Artorius Castus commanded the British legions serving against – but at that point the inscription is damaged, and only the letters ‘Arm’ are visible. Could it be the Armoricans, or Bretons, in which case a British connection would become much stronger? More likely, it was the Armenians. As Higham concludes, none of this looks like the career of someone who would have Celtic princes called after him some four hundred years later, for the inscriptions are unlikely to postdate the year 200.

So why bother with him at all? Higham is too scrupulous to speculate, but it seems horribly obvious that the whole source of the trouble is the deep attachment to the image of King Arthur, Sir Lancelot, Sir Gawain and their comrades as mounted knights in armour, preferably shining. That is how they were presented from the 12th century onwards, when armoured cavalry ruled the battlefield and acted as patrons of poets and chroniclers. But were there any in Britain back in, say, the sixth century? Are they not just a medieval anachronism?

The answer to this seems to have been first given by R.G. Collingwood, in a few ill-advised sentences at the very end of his section of the Oxford volume on Roman Britain and the English Settlements. There he suggested that Artorius/Arthur could have been the commander of a late Roman field army, not of tramping legionaries, but of mail-clad cavalry, equites cataphractarii. ‘Through the mist of legend,’ he wrote,

one might still descry … a country sinking into barbarism … and the emergence of a single man intelligent enough to understand [Roman ideas], and vigorous enough to put them into practice by gathering around him a group of friends and followers, armed according to the tradition of civilised warfare and proving their invincibility in a dozen campaigns.

The knights of the Round Table, in other words, were technologically superior, fighting for a doomed civilisation. Collingwood was writing in 1936, and I think that he had fighter planes on his mind, or more likely, tanks. But in any case, he presented an argument for Arthur as a real-life commander of heavy cavalry in an infantry age.

Now, where would Romans recruit heavy cavalry, and find horses big enough to carry them? One popular theory says from the Sarmatian tribes, out on the steppes beyond the Danube, not hopelessly far from where Artorius Castus was stationed in Croatia. There was indeed a unit of Sarmatian cavalry in Britain, stationed at Ribchester in Lancashire as confirmed by several inscriptions, and they may have descended from a large group drafted by Marcus Aurelius and sent (good Roman practice) to serve as far as possible from their homelands. But all this happened long before the Arthurian era, when the Roman Empire was still in firm control. Could the Sarmatians have preserved their ethnicity and culture for many generations in an alien land? And also their memories of glorious victories won in Britain under the leadership of Artorius Castus, their enemies later being misunderstood as not Caledonians from the north, but Saxons from the east? The first supposition is highly unlikely, the second even more so. There is ‘not a shred of evidence for it’, Higham reports. (It becomes one of his favourite phrases.)

The Sarmatian theory was taken up in Antoine Fuqua’s movie King Arthur (2004), starring Clive Owen and Keira Knightley, which must rank as one of the silliest films ever to have come out of Hollywood. Its tenuous grip on history is best exemplified by the claim in its tie-in novel that ‘the victory of Arthur at Badon Hill was so complete and so devastating that the Saxon army retreated for ever from Britain’ – which, if true, would mean that we must all be speaking Welsh to this day. Its grip on geography is if anything even worse. But the real point is that the film is clearly about Vietnam, mapped onto Dark Age Britain. The baddies are Roman bureaucrats (think Washington) and Saxons invading from the north (Ho Chi Minh’s army). The good guys are the British ‘Woads’ (South Vietnamese), protected by the Sarmatians, draftees, slave-soldiers, hoping vainly for a return to freedom and their homes – in other words, noble but unfortunate American ‘grunts’. The big difference, of course, is that in the movie the good guys win. For Collingwood, Arthur defended civilisation. For Fuqua (and, with less excuse, his on-set advisers), Arthur fought for freedom and democracy.

Higham goes on from the Sarmatian theory to look at the associated notion that vital parts of the Arthurian legend – such as the Holy Grail, and Bedivere’s return of Excalibur to the Lady of the Lake – are derived from the sagas of the Narts, another people dwelling far away in the Caucasus mountains. Could such stories have been imported by the Alans, a tribe which seems to turn up everywhere on the wilder shores of speculation? There is, Higham concludes, ‘no connection’. Magic cauldrons and mysterious swords are common items in folklore, and in ritual practice. The same goes for a further suggested origin in Greek mythology, centred on the name Arktouros, or Arcturus, the familiar star-name which medieval scribes sometimes substituted for Artorius. But there was no ongoing awareness of the Greek language or culture in early Britain: again, confusion is not connection.

Having​ dealt with modern academic fantasies, Higham turns his attention to what we actually have to go on. It does not amount to much, and it has been cherry-picked into a mistaken consensus. Our only extensive near contemporary account for the post-Roman years is Gildas’s De Excidio Britanniae, ‘On the Destruction of Britain’, probably written in 536 ad. This mentions the siege of Mount Badon as a major victory over the incoming Saxons, but doesn’t ascribe it to Arthur. It has become a cornerstone of the belief that Arthur stopped the Saxons (for a while) in their tracks.

Early Welsh poems mention Arthur, but the most interestingly historical of them, Y Gododdín, does so only tangentially. Whatever John Morris and Leslie Alcock may have written in the 1970s and 1980s, the evidence provides ‘no space … for an “Age of Arthur” during which a victorious British emperor-like figure held back the barbarian hordes’. That image really dates back to the Historia Brittonum of ‘Nennius’, the work of one or more Welsh compilers, dating in the form we have it from 829-30. Its chapter on Arthur, which amounts to less than a page in a modern book, names him not as a king but as dux bellorum or ‘war-leader’, and lists his 12 battles, culminating in Badon. Most of the modern ideas of ‘historical Arthur’ images, like Collingwood’s ‘dozen campaigns’, go back to Nennius. However, Higham concludes that this was ‘in large part a work of historical fiction’, with an evident propagandist purpose in the Welsh politics of the time.

Nennius nevertheless picked up the name of Arthur from somewhere, and his list of battles does not look like pure invention. There are also many independent references to Arthur in early medieval Welsh tradition, which Higham discusses. Still, Nennius’s Arthur might have remained in scholarly oblivion had he not been picked up by Geoffrey of Monmouth 300 years later, and expanded into a whole book of his Historia Regum Britanniae. This immensely successful work – more than 200 surviving manuscripts, and translations into Old French, Middle English, Welsh and even Icelandic – brought Merlin into the story, gave Arthur an ancestry, a biography, as well as inflating him into a European conqueror who marched on Rome itself, until betrayed by Mordred and Guinevere. The modern attempt (by Leslie Alcock) to find an Arthur really campaigning in Gaul, under the name or title of Riothamus, is another act of desperation, Higham points out: Riothamus was probably Breton rather than British. It was the complete lack of corroboration from any European chronicle that eventually – and much to the annoyance of Henry VII – put an end to Arthur-based claims of ancient seniority for English kings.

As for the many modern attempts to make sense likewise of Nennius’s list of battles, by identifying place-names and trying to construct a plausible campaign history, Higham prints a map showing scholars’ preferred fields of Arthurian activity. Seven plump for southern Scotland, the old Debatable Land and the cockpit of Britain for a thousand years, including Longtown kirkyard where Guinevere is allegedly buried. Six vote for the east coast of England, eight for Wales and three for the south-west. Everywhere except the Midlands, in fact. There is no fire, Higham concludes, just smoke, and even that looks more like ‘highland mist’. He agrees in the end with Dumville and Myres that ‘the only safe conclusion must be that [Arthur] was a fictional or fictionalised character introduced for rhetorical purposes in 829-30. We should cast [him] out of our histories,’ and accept that we know nothing of his real-life origins.

But Arthur has regularly proved hard to finish off, even by the most determined of historians. There is one evident reason for the ‘historical’ Arthur’s unexpected revival in the 20th century, and that is contemporary experience. If Collingwood was seeing Arthur in terms of fighter planes, and Fuqua in terms of Vietnam, a number of authors saw him very clearly in imperial or post-imperial terms. Kipling may have started the vogue with his ‘Parnesius’ stories from Puck of Pook’s Hill (1906), in which he draws a very detailed parallel between Roman and British Empires, notably the thanklessness that greeted the devoted service of British-born Romans, at Hadrian’s Wall, and Indian-born Britons on India’s north-west frontier. Arthur does not come into the stories, but that is only because Parnesius’ Britain has not (yet) been abandoned, though the signs of decay are there. Conan Doyle pushed that idea further in his grumpy story ‘The Last of the Legions’ (1911): ungrateful people demanding independence might get more than they bargained for, he warned. Alfred Duggan’s two novels about the end of empire, The Little Emperors and Conscience of the King (both 1951), once again drew an explicit parallel, with Duggan pointing out that neither the Roman withdrawal from Britain nor the British withdrawal from India was prompted by military defeat. So what caused the collapses? Incompetent bureaucrats sent out from Rome, or ‘Home’, not speaking a word of the language of the people they affected to rule, was his answer, and Duggan’s stepfather was Lord Curzon, Viceroy of India, so perhaps he knew what he was talking about. Arthur and Badon creep into the end of Duggan’s novel sequence, though the ‘siege’ is presented as less than decisive. For these and other British writers, the image of Arthur as imperial holdout had turned personal.

A further powerful modern propellant for Arthur’s legend has been archaeology, with its mix of hard facts and openness to interpretation. Even ‘highland mist’ has its appeal. Who can forget Legio IX Hispana marching off into Caledonian mystery, in Rosemary Sutcliff’s Eagle of the Ninth (1954)? Part of the inspiration for that was the discovery at Silchester of a wingless Roman eagle. How did it get there? Her Arthurian novel, Sword at Sunset (1963), centres on another discovery, at Newstead beyond Hadrian’s Wall, of the skeleton of a diminutive woman, lying in a pit beneath the bones of nine warhorses, all buried in the middle of the fort’s old parade ground. The mix of mystery, ritual, old civilisation and new barbarism has proved intoxicating. The Morris-and-Alcock revaluation of the 1970s, to which Higham is still responding, also got a lot of its force from dig discoveries, notably at Cadbury and Tintagel. There is still an urge for the knights in shining armour, but if they can be combined with a real-life guerrilla warfare scenario, with a sense in it of the sun setting on an empire – note Sutcliff’s title – so much the better.

And what​ , finally, of the non-historical or romantic Arthur? Even more powerful than Geoffrey of Monmouth’s rather tedious chronicle was the string of romances that also began to appear in the 12th century, and which transferred Celtic tales to an up-to-date French setting: Chrétien de Troyes’s Erec, Lancelot, Perceval, Yvain. It was these which introduced the Waste Land, the Holy Grail, and above all the doomed romance of Lancelot and Guinevere to the European imagination. In the English imagination they powered Thomas Malory’s 15th-century Morte D’Arthur, created the Victorian vogue for William Morris’s ‘Defence of Guenevere’ (1858) and Tennyson’s Idylls of the King (1859-85), and returned in T.H. White’s Once and Future King (1958) and the Disney movie based on it in 1963, with a dozen successors including John Boorman’s 1981 film Excalibur. It’s not unreasonable for Higham to give what we might call ‘the Guinevere tradition’ little space, for he is concerned above all with the ‘truth’ of Arthur, i.e. what if anything lies behind the quasi-historical setting of the stories. But Guinevere, Mordred, Uther Pendragon and Morgan le Fay have even so become an inextricable part of Arthur-as-now-received.

Maybe the trouble just is that there are so many clues loose, and everyone can pick their favourite. Mine is the battle of Camlann, ‘in which Arthur and Medraut fell’ (Annales Cambriae entry for 537 ad, but written four hundred years later, though not taken from Nennius). Higham locates this at the ‘small vill of Camlan in the Dovey (Dyfi) valley’, but Camlann could well derive from earlier Camboglanna, a fort on Hadrian’s Wall. Some identify this with Castlesteads, but the name means ‘crooked glen’, and that is a good fit for Birdoswald, which looks down on a dog-leg of the River Irthing, and was occupied in the post-Roman era. Arthur fell there, say the Annales, but Geoffrey of Monmouth says he was taken to Avalon to have his wounds healed. The big fort on the western end of Hadrian’s Wall, a day’s walk from Birdoswald, was at Burgh-by-Sands – Aballava in Latin, a name soon forgotten but easily garbled. So they took Arthur to the base hospital?

As for Burgh-by-Sands, its castle was held in the late 12th century by Hugh de Morville, and Hugh is thanked by the Swiss author Ulrich von Zatzikhoven for giving him the book from which he translated his romance Lanzelet. Clearly Hugh – who was lord also of Pendragon Castle close by – knew something, maybe a lot, about the Arthurian story. If, after Camlann, Excalibur had to be thrown away, as the stories insist, what better place than into the notorious Solway quicksands at the foot of Hugh’s castle? De Morville was also one of the murderers of St Thomas à Becket, Archbishop of Canterbury. What kind of a story could one not make out of all those facts? A hidden tomb for Arthur, pagan Pendragons guarding the secret, Excalibur maybe not beyond recovery. Indiana Jones would love it.

Higham would surely say, ‘cherry-picking again!’, and he would be right: all these ‘facts’ are open to doubt (except for the murder of Becket), and they are pulled from sources of quite different date and type. Yes, we might reply, but the cherries are so enticing. In Chrétien’s Erec, Arthur is warned by Gawain that his plan to hunt the White Hart is dangerous, for, according to custom, he who catches the White Hart wins the right to kiss the fairest lady of the court, and that means deciding who she is, and behind every lady stands a knight committed to asserting her pre-eminent beauty. It will mean duels for ever!

Maybe so, Arthur replies, but still,

a word a king has said
must not then be taken back.
Tomorrow morning, with great delight,
we will all go to hunt the White Hart
an la forest avantureuse.
Ceste chace iert molt mervelleuse.

And, he might well have added, ‘get this, Gawain, what I said about “great delight”, that’s an order!’ Like the White Hart, and whatever historians may say, the truth is that Arthur still lives in the Forest of Adventure, and the chase after him remains ‘full of marvels’. But not, one must hope, with Higham, downright silly ones. Not, as the scholar Caroline Jewers once quipped, King Arthur, or 101 Sarmatians.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 41 No. 1 · 3 January 2019

‘Legends of King Arthur have remained deep-rooted in popular imagination,’ Tom Shippey writes (LRB, 20 December 2018). He takes most of his examples from the 20th century, and doesn’t mention the fascinating ideological use of the King Arthur legend for the purpose of legitimising British colonial conquest and dominion in the late 16th century. A key figure setting this process in political motion was Dr John Dee, Queen Elizabeth’s court mathematician and geographer since her accession in 1558. Dee had studied at Louvain under the Dutchman Jemme Reinerszoon, born in the Friesian seaport of Dokkum in 1508. Better known as Dr Gemma Frisius, this polymath was a major figure in the development of global geography. One of Frisius’s students (and collaborators) was the Flemish cartographer Geert de Kremer, who became famous as Mercator. The young Mercator had previously studied in ’s-Hertogenbosch, at the time a cosmopolitan cultural centre.

Located in what is now North Brabant in the Netherlands, ’s-Hertogenbosch is also associated with Jacobus Cnoyen, author of Itinerarium, a travel book, now lost but influential in the 16th and 17th centuries. Cnoyen not only described his own voyages but incorporated old narratives, including stories about King Arthur’s legendary seafaring voyages and conquests. The Antwerp book trader and geographer Abraham Ortels (Ortelius) owned a copy. He loaned it to Mercator, who used it for his map of 1569 detailing (largely imaginary) Arctic geography.

As Mercator explained, Cnoyen’s Itinerarium ‘makes some citations from the Gesta of Arthur of Britain’, but, significantly, ‘the greater and most important part he [Cnoyen] learned from a certain priest at the court of the king of Norway in 1364.’ In 1577, Mercator dispatched a letter to his friend Dee in which he had transcribed relevant parts of Cnoyen’s book identifying King Arthur as an ancient conquering ruler. Dee was an active promoter of British colonial exploration and expansion; he is thought to have been the first to deploy the term ‘British Empire’. He translated and abbreviated Mercator’s transcription, and incorporated it in his treatise Brytanici Imperii Limites (1578).

Soon afterwards, Dee’s fellow Brits expanded on his book, among them Richard Hakluyt, author of Principall Navigations. Hakluyt’s first volume portrays King Arthur as the brave and devout British conqueror of multiple ‘wild and savage’ peoples inhabiting Ireland, the Scottish Isles, Iceland, Greenland, Scandinavia, Lapland, ‘and many other Islands beyond Norway, even under the North Pole’.

Harald Prins
Kansas State University

Vol. 41 No. 3 · 7 February 2019

I knew nothing of the use made of King Arthur by 16th-century geographers, till enlightened by Harald Prins, though it just shows that Arthur legends die hard (Letters, 3 January). I feel that both John Dee and Richard Hakluyt should nevertheless have known better. As I mentioned briefly in my review of Nicholas Higham’s book (and at greater length in a piece in the LRB of 31 July 2008), Geoffrey of Monmouth’s legend of Arthur’s European conquests had been disproved by Polydore Vergil’s Anglica Historia, commissioned by Henry VII but only published after a twenty-year delay in 1534. Polydore pointed out, very reasonably, that if Arthur had conquered half of Europe, some European chronicler would surely have mentioned the fact: but no one had. This provoked furious patriotic denials, notably by John Leland, whose Assertio inclytissimi Arturii regis (1544) wrote Polydore off as a bitter, jealous foreigner ‘steeped in Italian vinegar’ (Italo perfusus aceto), protested the truth of the legend, and listed all 149 knights of the Round Table to prove it. So there!

Just the same, as far as I was aware, the learned world reckoned that Polydore had settled Geoffrey’s and Arthur’s hash, and by the 1570s and 1580s Dee and Hakluyt might be expected to have caught on. Milton commented in his History of Britain (1670, but begun decades earlier) that he doubted Arthur’s very existence, adding with a dash of English vinegar (and so much for Geoffrey): ‘He who can accept of Legends for good story, may quickly swell a volume with trash.’ But alas, people believe what suits them, especially when it comes to national pride. I’m surprised at those Dutch geographers, though. It can’t have been ideology in their case. Word just hadn’t got round.

Tom Shippey
Buckland Newton, Dorset

Vol. 41 No. 4 · 21 February 2019

In the discussion of the way King Arthur’s legend has been used to further contemporary political debates through the ages, I have been surprised to see no mention of what must be the most powerful role played by the mythical king – namely, as part of the Plantagenets’ campaign to fashion a unified British culture out of the hotchpotch they inherited (Letters, 3 January and Letters, 7 February). It can be no coincidence that the Arthurian legend received so much royal attention after the loss of Normandy and other French possessions under King John. England, and eventually Wales and parts of Scotland too, were ruled by a French-speaking aristocracy, cut off from their ancestral homeland, but with no clear tie to their new lands and peoples, either in the Germanic heartland or the Celtic fringe. Tapping the already ancient, pre-Saxon Arthur as a symbol of the English monarchy was a brilliant move. Here was a rare unifying figure in a cultural landscape shaped by conquest and displacement. The cult of Arthur reached its height during the reigns of Edward III – who explicitly modelled himself as a new Arthur – and his successor Richard II, at precisely the time when Chaucer and his contemporaries were fashioning a national language, a linguistic compromise between Old English and French that was accessible to everyone.

Hassan Damluji
London SW1

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences