The Reign of Arthur

by Christopher Gidlow

  The Historia, therefore, gives us a picture of Saxon settlement and revolt, followed by British resistance, which accords with the other sources. The Historia’s unique contribution is naming the warleader of the main wars as Arthur.

  The War of Resistance

  Historians fall into two main camps over the Arthurian battle-list. Most see it as almost mythological in its hyperbole, and unworthy of consideration. The others, particularly before Dumville, saw it as preserving almost intact an ancient Welsh praise-poem, translated into Latin. Let us consider the charge of mythologising first.

  At the Battle of Mount Badon, Arthur is said to have overthrown 960 men in a single charge ‘and no-one laid them low save he alone’. This is supposed to set up Arthur as a mythical superman who could destroy whole armies in single combat. Even given the tendency for exaggeration by Dark Age writers, I cannot believe this is how Nennius intended his comment to be read. It was a commonplace, then as now, to ascribe the feats of an army to its commander. The Anglo-Saxon Chronicle, on the battle of Chester, says: ‘Aethelfrith led his levies to Chester and there slew a countless number of Welsh.’ In the report of the death of Natanleod, a literal reading would suggest that Cerdic and Cynric had killed all 5,000 Welsh. Only Aelle and his son Cissa are said to have captured Pevensey and slain all the inhabitants. The absurdity of suggesting that the battle-list means that Arthur in person had killed all the 960 men is obvious. The presence of armies with the leader is implicit. If Arthur was leading a charge in which 960 men were overthrown, the claim is modest compared with the totals recorded by the Saxon leaders.

  If we look at the report of Mount Badon in context, it is plain that the statement ‘and no one overthrew them save him alone’ is not meant to exclude Arthur’s troops (they have never been mentioned), but his partners, the kings of the Britons. In previous battles, they fought together, but in this final charge at the last battle, the glory went to Arthur alone.

  The alternative explanation of the list, repeated since the 1930s (Chadwick and Chadwick 1932), is that it is a translation of a Welsh praise-poem. The implication is that we are reading a genuinely ancient source, composed close to Arthur’s time, in which his exploits were celebrated.

  Fortunately, we can compare the list with an actual surviving praise-poem, that of Cadwallon of Gwynedd. Bede recorded the death of King Edwin of Northumbria at the battle called in English Haethfelth in 633. The victor was ‘Cadwalla’, king of the Britons. The Historia knew this too, giving the battle its Welsh name, Meicen, and identifying its victor, ‘Catguollaun’, as the King of Gwynedd. Here is the praise-poem celebrating the campaign:

  Cadwallawn, before his coming waged them for our good fortune – fourteen chief battles of fair Britain and sixty encounters . . .

  The camp of Cadwallawn on the Don; fierce affliction to his foe, a lion of hosts over the Saxons.

  The camp of Cadwallawn the famous, on the upland of Mount Digoll, seven months and seven battles each day.

  The camp of Cadwallawn on the Severn, and on the other side of the River Dygen, almost burning Meigen.

  (translation by Bromwich in Barber 1972)

  The other eleven battles follow. Most give the King’s camp on a riverbank. Two were on hills, one was at a Caer (usually a Roman city), where there was ‘A besieging army and a hundred zealous men’. At one waterside camp, Cadwallon ‘satiated eagles after battle’.

  Arthur’s battle-list has twelve battles, most on rivers, two on hills, one at a Roman city, quite convincing correspondences. Unfortunately, Cadwallon’s praise-poem is itself a late document. It appears in the late fourteenth-century Red Book of Hergest, and is written in language suggesting a date of composition in the ninth or tenth century (Bromwich 1961). It could, therefore, post-date the Historia battle-list, and even have borrowed features from it.

  In all other respects, Arthur’s battle-list does not read like the poem. Arthur’s battles are in a numbered sequence, a feature not found in the praise-poems, and only two have incidents linked to their names. The main thrust of the ‘praise-poem’ argument does not relate to style, however. It is argued that certain linguistic features are only explicable if they derive from a Welsh original. The names Guinnion, Celidon and Badon, for instance, rhyme, as do Dubglas and Bassas, which might indicate a poetic source.

  The most widely cited ‘proof’ of a Welsh original is the description of the battle at Castellum Guinnion. Here, Arthur carried the image of the Virgin Mary on his shoulders. This is supposed to be explicable only if Nennius mistook the Welsh word for shield as the similar word for a shoulder and then put it into the plural.

  This argument is so weak it is surprising to find it as a given in every book on the historical Arthur. Even if the design was on his shield, a writer could justifiably have said this was being carried on his shoulders, whatever his language. Medieval guesses put the image on Arthur’s shield, on a banner, or made it a three-dimensional image carried for piety. The same phrase occurs in Annales Cambriae, where, at the battle of Badon, Arthur carried the cross of our Lord Jesus Christ on his shoulders for three days and nights. It seems highly unlikely that this terse entry was a snippet from a Welsh praise-poem, and unlikelier still that its author should have made exactly the same mistake in translation. This is especially true if, as the proponents would have it, the phrase ‘on his shoulders’ was meaningless to a reader of Latin.

  An ivory carving thought to be of the Roman Magister Militum Stilicho, from the early fifth century, shows a variety of ways in which the reference can be understood without resorting to a lost source in another language. He has on his shoulder an embroidered patch decorated with crosses. These shoulder-patches were standard in the late Roman army. His shoulder-slung cloak is embroidered with images of saints or perhaps his imperial commander and his shield has an inset cameo of a mother and child, who could be seen as the Virgin and Jesus.

  Concentration on the shield/shoulder confusion has obscured a much more fruitful line of enquiry. When Dark Age and medieval writers needed to explain or translate material they were copying, they would usually add a ‘gloss’, a short phrase usually beginning ‘that is . . .’. The battle in the Caledonian wood is given first in Latin, then glossed in Welsh. If the lists were composed in Welsh, then translated into Latin, we would expect to find the opposite e.g. ‘the battle of Cat Coit Celidon, that is [in Latin] the Celidonian Wood’. Elsewhere, in instances such as Vortimer’s battles or the English-derived genealogies of the northern dynasties, Nennius’s glosses follow this same pattern, giving the information in Latin, then adding glosses giving Welsh translations. For example: (HB 57) ‘the battle called Gueith Lin Garan’; (HB 61) ‘He is Eata Glinmaur’. The conclusion is inescapable. Nennius had the entry relating to the Caledonian wood, if not the whole list, from a source which was not Welsh. It could only have been in Latin or even English material of some kind pre-dating the early ninth century.

  Although the details are different, the basic framework of the Saxon wars in Historia Brittonum accords with that presented by Bede and derived from Gildas. There is one striking area in which they differ: the role played by one of the major figures of the period.

  As Gildas tells it, after the Saxon revolt, the Britons banded together under Ambrosius Aurelianus and challenged the Saxons to battle. ‘From then on the victory went now to our countrymen, now to their enemies . . . This lasted right up till the year of the siege of Badon Hill, pretty well the last defeat of the villains and certainly not the least.’ Bede took Gildas as saying that Ambrosius Aurelianus was the victor of Mount Badon.

  The Historia says much more about Ambrosius than Arthur. He is known both by his Latin name and as Embreis Guletic, showing he figured in Welsh tradition. He appears as a fatherless child about to be sacrificed by Vortigern and as the son of one of the consuls of the Roman people. He is a native of Mais Elleti in Gleguissing, the region of South Wales between the Gower and Gwent. His prophecies to Vortigern, revealing two
fighting worms under his fortress, are the key point to a story with more in common with the legends of the Mabinogion than sober history. After Vortigern’s death, Ambrosius rules in western Britain and is described as ‘King among all the Kings of the British nation’. It is by his permission that Fernmail of Builth’s ancestor is given his lands to rule. Just about the only thing we are not told about Ambrosius is that he fought the Saxons. On the contrary, he is said to be an adversary of Vortigern, who lived in dread of him. It is implied that this is one of the factors which impelled Vortigern to employ the Saxons in the first place. The warfare, sometimes in favour of the Britons, sometimes against them, is instigated by Vortimer and carried on by Arthur and Outigirn.

  For Nennius, Arthur is famous as a leader in battles, but nothing else is revealed about him. In contrast to the obviously fictitious material attached to Ambrosius, Arthur’s role does not strain credulity nor smell of folklore. Yet, despite the legendary accretions, historians accept Ambrosius as a real character, while Arthur is now treated as all but entirely mythical.

  What is Historia Brittonum?

  On the Saxon invasions and British Resistance, Historia Brittonum merely supplements available earlier material. With Ambrosius, it seems to part company entirely. We will have to establish how acceptable the Historia’s version is, by examining the author’s sources and how he incorporated them into his work, before analysing his picture of Arthur and his age.

  The major issue is how far the Historia is the work of a single author, in the sense that Bede or Gildas are the authors of their work, or, as the ‘Nennius’ prologue suggests, a barely edited collection of historical materials. We can begin by asking where ‘Nennius’ might have got his facts from.

  Certain areas of Britain are given more attention in the Historia than others. This could be an indication of the location of the author, or of his sources, or both. As a comparison, Bede’s Ecclesiastical History gives prominence to both Northumbria (Bede’s home kingdom) and Kent, where his earliest and most reliable sources are found.

  The main geographical areas covered by the Historia are:

  1. South Wales, the lower Severn Valley and estuary and marches of England. A striking example is the inclusion of the full genealogy of Fernmail of Builth ‘who now rules’, an unparalleled reference to a contemporary British ruler, and one so minor he can only be included for local interest.

  2. Kent. Much material, especially relating to the first settlement of the Saxons, focuses exclusively on Kent.

  3. North-east England/south-east Scotland. In particular, the Historia includes lists of Kings of the area, as well as the origins and British equivalents of place-names there.

  4. Ireland. The Historia presents foundation legends of Ireland, such as are found in the Irish Lebor Gabala (the Book of Invasions). There is also a major section on the life of St Patrick.

  Of equal importance are the areas about which the Historia knows little or nothing. Beyond Kent, there is nothing about lowland England. Aside from the list of twenty-eight cities at the end of the Historia, there are no place-names in the south between the Isle of Wight and Gloucester and the Severn Valley. The Isle of Wight is mentioned once, as one of the three adjacent islands of Britain. Sussex is the only named region of southern England beyond Kent.

  While we might expect a ‘History of the Britons’ to say little about English areas, it is striking how limited its interest is in the British areas. There is nothing about Devon and Cornwall. North Wales and the north-west (Gwynedd and Rheged) are hardly mentioned. Apart from references to Snowdon and Caernarfon, North Wales features only as the homeland of great British kings such as Mailcunus and Catgablaun. These do not appear in a North Welsh context, but as adversaries of the English kings of Northumbria or descendants of the Gododdin. Urbgen (Urien of Rheged) is not located by Nennius. He, too, is only important as an opponent of the Northumbrians.

  It is difficult to imagine an author with such disparate geographical interests. The only convincing explanation is that Nennius follows sources from discrete geographical regions.

  The South Welsh, Kentish, north-eastern and Irish elements of the Historia are of different styles and content, linked only by Nennius, interweaving them in his book. This interweaving between passages is generally not smooth. One moment we might be reading about how St Germanus prophesies about Catel, whose descendents ‘rule Powys even to this day’, then find ourselves moving immediately on to ‘And it came to pass, after the English were encamped in the aforesaid Island of Thanet’, aforesaid, that is, in that it was mentioned four chapters earlier before the St Germanus material which knows nothing of Thanet, Kent or the Saxons and seems to be taking place a generation earlier.

  The knowledge that Nennius uses geographically discrete sources is of crucial importance now we come to examine the Arthurian battle-list. It might be that Arthur’s battles derive from areas not elsewhere represented in the Historia, but it is most plausible to assume that it conforms to the general pattern of Nennius’s sources.

  IRELAND

  We can swiftly dismiss the Irish material as the origin of the battle-list. Nennius always presents Irish historical sections as discrete from the British history. In the Harleian, it is only linked to the fifth/sixth-century British material at the beginning by the phrase ‘And St Patrick was, at that time, a captive’ and at the end, six chapters later, by ‘at that time, the Saxons increased in number’. The sources are two late seventh-century Irish lives of St Patrick, with some later additions (VII Dumville 1990). The previous reference to Ireland was forty chapters earlier, a prehistoric origin legend. Ireland is never mentioned again in the Historia.

  KENT

  The Arthurian battle-list is placed firmly by the author in a Kentish context. It begins with an introduction on the kingdom and kings of the men of Kent, continuing: ‘Then Arthur fought against them in those days . . . .’ The style of the list is similar to the Vortimer section, which is indisputably set in Kent, including identifiable Kentish place-names, and is clearly of Saxon and, presumably, Kentish origin. It includes the genealogy of Hengist back to Geta ‘one of their idols, which they worshipped’. Place-names in the Vortimer section are given first in English (Thanet, Canturguolralen, Episford), then glossed as ‘in British Ruoihm’ ‘in our language Chent’ ‘in our language, on the other hand, Rithergabail’. The same feature is found in the Arthur battle-list: a non-British place-name is glossed into British: ‘That is Cat Coit Celidon.’

  The title Dux applied to Arthur is only found elsewhere in the Historia in Roman material linked to the Kent sources. The Roman and Kentish material is not only linked stylistically. They share a common dating system based on consular years. The two Roman consuls were elected annually. Although they played an almost entirely honorary role in the late Roman Empire, their names were used as a system of dating: ‘In the year of the consulship of A and B.’ Nennius himself misunderstands this system, thinking that the consuls are in some way synonymous with Emperors. In the time of the Emperor Maximus, we are told, ‘the consuls began, and they were never again called Caesars’. This was presumably the date from which his Consuls’ List started. Thereafter Nennius uses Consul for ‘Roman ruler’. Nennius’s lack of understanding of the difference between emperors and consuls (‘Gratian ruled for the second time, with Equitius’) is a clear indication that it is his source, not he, which originates the consular system of dating. The Historia links a dating system based on the Passion to the Kentish material, but it is less clear where this derives from. This system, used in the opening quotation from Malory, took as its starting date the death, rather than the birth, of Jesus. It was the work of enthusiasts like Bede which caused it to be superseded.

  Dumville suggests that our author, ‘Nennius’ himself, has the Paschal tables of Victorius, which synchronise post-Passion and consul dates, in front of him (IV Dumville 1990). I think this is unlikely, given that these synchronisms are only applied to particular material,
rather than throughout the Historia. The Paschal tables lead back to Jesus’s time, and do not explain Nennius’s belief that the consuls were a late feature of Roman history, synonymous with the earlier Caesars. These features can only be explained by their incorporation in particular sources used by Nennius.

  The counter-arguments, that the Arthur battle-list is not exclusively Kentish, are all exterior to the text itself. We know that the Caledonian wood cannot be in Kent. It must be in the north. Gildas refers to the City of the Legions and the siege of Mount Badon, but shows no knowledge of Kent.

  SOUTH WALES

  I use this term to mean medieval Wales excluding Gwynedd, but including the adjacent marches in what is now England. In general, the Historia concentrates on south-eastern and central eastern Wales, Gwent, Glywysing and Powys.

  The South Welsh material is recognisably of two types. The first is a Life of St Germanus, said in the Historia to be a ‘Book of the Blessed Germanus’. It seems to have no internal dates. It includes an origin legend of the Kingdom of Powys, probably where it came from, and locates Vortigern in Gwerthrynion and on the Teifi in the land of the Demetians, Dyfed.