(This originally appeared in Wargames Illustrated 139 (April 1999), pp.12-13)
The purpose of this and the next part of the series is to evaluate the written evidence available. Obviously, you can only claim a wargame to be historically plausible if it’s based upon reliable evidence. Many ‘Dark Age’ wargamers regard this as pointless ‘nit-picking’, but I can’t see why. ACW or Napoleonic wargamers expect only the best, reliable, historically accurate information and detail these days, and ridicule works based on out-of-date, uncritical sources, let alone supposedly ‘historical’ wargames based on Cornwell’s Sharpe novels. And so they should. So why should the interested wargamer of the post-Roman period expect to put up with being fobbed off with anything less? Why should Cornwell’s ‘Warlord’ novels (great read though they are) be any more reliable as a basis for a historical wargame?
Gildas: On the Ruin and Conquest of Britain.[i]
We kick off with the earliest insular record of the period. Gildas, although acquiring saintly status in the Celtic church, having various works ascribed to him, and making walk-on appearances in late Breton Saints’ Lives, is otherwise unknown. All we can find out about him has to be extracted from his De Excidio et Conquestu Britanniae (On the Ruin and Conquest of Britain). This work is an extended sermon, not a history, but it contains a summary of events up to the time Gildas was writing, to support his argument.[ii] This smallish section of the whole has formed the basis of most attempts to write fifth-century British history since the days of the Venerable Bede. So, what does Gildas say?
He begins with the usurpation of the ‘tyrant’ Magnus Maximus in Britain (383) – typical British perfidy says Gildas. After the departure of the Romans the Britons are left to fend for themselves against Pictish attacks, but prove unable to do so. Twice the Romans return to build walls against the Picts, and instruct the Britons in matters of defence, but to no avail. The Picts sweep down through the province, with much fire and slaughter. The Britons are driven into mountain fastnesses and other remote places. They appeal to Aëtius ‘thrice consul’, then active in Gaul, in a letter quoted in part by Gildas, but no help is forthcoming. Eventually they rally and the Picts are driven off. Peace ensues but the Britons lapse into sinful ways.
Meanwhile, says Gildas, rumours of another attack reached the ears of the ‘British council’, and ‘the Proud Tyrant’ called in some Saxon mercenaries to help defend the island. This they did until the supplies which they had been promised dried up, when they rebelled, and were joined by reinforcements from Germania. Again, warfare, fire and slaughter ravaged Britain from coast to coast, and the Britons were driven into natural fortresses to hide. They rallied again, this time under Ambrosius Aurelianus (one of the few named personages in Gildas’ work) and after a long war, with mixed success, they eventually triumphed at the obsessio Badonici montis: ‘Siege of Mount Badon’. Gildas appears to say that this siege took place forty-four years and one month before he was writing, at the time of his birth. Sadly, as always seems to happen in Gildas, whenever he has anything really important to say his Latin becomes quite impenetrable. Since Mount Badon, the Britons have lapsed in luxury, says Gildas, and he then moves on to his complaint against the kings of Britain, five of whom he names: Cuneglassus, Vortiporius, Aurelius Caninus, Constantine and Maglocunus. Note that Gildas makes no mention of Arthur.
Of the named kings, the last, Maglocunus, ‘Dragon of the Island’, may well be the Maelgwn of Gwynedd recorded by the Annales Cambriae as dying of plague in 547. This would give us a date of some time before then, probably in the 540s, for Gildas’ writing, and we can work back from that. Following the usual reading of his statement about Mount Badon, that would date that battle to c.500. Ambrosius Aurelianus would flourish before that battle (tallying with Gildas’ statement that his grandchildren were alive in his own day), and presumably after the appeal to Aetius, thrice consul. That, in turn, must date to between 445 (the date of Aëtius’ third consulship) and 454 (his death). The ‘Proud Tyrant’ (superbus tyrannus) is held to be a punning reference to the Vortigern (which name means – roughly – ‘Big King’), first named by Bede as inviting in the Saxons, and associated in Nennius’ Historia Brittonum with events of the 420s. So far so good.
Well, no, not really. Unfortunately, as we shall see, the Annales Cambriae are a very late source, based upon Irish annals, but with no accurate information for the sixth century. Their ‘obit’ for Maelgwn, assuming he is Gildas’ Maglocunus (the name is the same) is quite unreliable. This means it is hardly a fixed point at all. David Dumville[iii] showed that all other attempts to date Gildas by external means (such as references to the named kings in Welsh royal pedigrees) were similarly dubious. That only leaves us with Gildas’ own narrative to help us. There are four more or less fixed points there: Magnus Maximus (whom we know, even if Gildas didn’t, reigned 383-8); the appeal to Aëtius (445×454, as stated); Ambrosius, whose grandchildren are Gildas’ contemporaries; and Mount Badon, fought forty-four years before Gildas wrote. That apart, the whole narrative is very imprecise, rhetorical, and hardly the basis of sound chronology. Nevertheless, on the basis of these fixed points, and some fairly arbitrary assignment of time-spans to vague phrases like ‘after a long time’, Dumville did come round to thinking that Gildas must have written in the early-to-mid sixth century.[iv]
But there are still problems. Several writers have recently proposed that the information that Aëtius was ‘thrice consul’ may have been introduced into the letter by Gildas himself, after the event.[v] This could mean that the appeal to Aëtius took place up to twenty years earlier than traditionally thought. I’m not especially convinced by this, but it is possible. The phrase about the date of Badon is, as mentioned, quite opaque, and has been read (by Ian Wood[vi]) as meaning that the battle took place ‘a month ago in the forty-forth year of my age’. This would force Gildas back (or Badon forward) by forty years. This has not convinced everyone (for reasons of Latin syntax) but is, again, possible.
Patrick Sims-Williams[vii] has argued that the chronology of Gildas’ account is so vague that Badon could have taken place any time after about 450. This would be supported by (to my mind) the most serious obstacle to the traditional dating: between the account of the war with the Picts and the war with the Saxons, each of which forms a discrete section of Gildas’ narrative, occurs the word interea – meanwhile.[viii] This in itself means that there must be some unspecified chronological overlap between the separate accounts of the two theatres of conflict, and it destroys any neat sequential link between the appeal to Aëtius (in the section on the Pictish War) and the appearance of Ambrosius (in the account of the Saxon War). Ambrosius could even, conceivably, have been the person who sent the appeal. A final argument in favour of ‘the early Gildas’ was put forward by Michael Lapidge,[ix] who showed that Gildas must have received a classical rhetorical education, and posed the question of whether such an education was likely in the sixth century.
Finally, Gildas’ account, as you may have noticed from the potted synopsis above, is incredibly vague, rhetorical, and stylized too. Note how the Eastern Saxon War and the Northern Pictish War follow more or less exactly the same pattern.
Thus we have no idea when Gildas was writing. He may have been writing in the late fifth century, which is no bad thing as it brings him closer to events. It also, if you are so inclined, removes the problem of why he never mentions ‘Arthur’. If ‘Arthur’ was active around 500, then that might have been after Gildas. I wouldn’t want to push that, but previous attempts to explain Gildas’ silence about the ‘Once and Future King’ – that he was an enemy of the Church, or that he was too well-known – are pretty unconvincing, too, if not more so. It was famously argued by Kenneth Jackson that a preacher in 1850 would not need to say ‘the battle of Waterloo, which was won by the Duke of Wellington’, but, as Padel has riposted, he could very easily have said ‘Wellington’s victory at Waterloo’, and this is even more likely in Latin. One has to ask why Gildas did mention Ambrosius and the current kings but not Arthur. That Arthur was an enemy of the church is based upon very late and doubtful Celtic saints’ lives, where the prestige and power of the saint is simply emphasized by having him triumph over the great king of legend.
With all this in mind we must abandon attempts to use Gildas to construct a traditional narrative history of fifth-century Britain. This is a shame, especially for the wargamer, but all is not lost. As is so often the case with post-Roman history, it is a case of asking the right questions. Gildas may not answer the question of ‘what happened and when?’, but his work does contain other snippets of information which there is no reason to discount – about kingship in his own day (which, as I’ve just said, might be the late fifth century) and about the mechanics of the initial Saxon settlement.
[i] Gildas. The Ruin of Britain and other Documents. ed. & trans. M. Winterbottom (Chichester 1978).
[ii] On the Ruin and Conquest… c.4-26
[iii] D.N. Dumville, ‘Sub-Roman Britain – History and Legend.’ History vol.62, no.205 (1977), pp.173-92; D.N. Dumville, ‘Gildas and Maelgwn: Problems of dating’, in M. Lapidge & D.N. Dumville (ed.), Gildas. New Approaches (Woodbridge, 1984), pp.51-60.
[iv] D.N. Dumville, ‘The chronology of De Excidio Britanniae Book I’, in M. Lapidge & D.N. Dumville (ed.), Gildas. New Approaches (Woodbridge, 1984), pp.61-84
[v] P.J. Casey; N. Higham, The English Conquest. Gildas and Britain in the Fifth Century (Manchester, 1994).
[vi] I.N. Wood, ‘The end of Roman Britain: Continental evidence and parallels’, in M. Lapidge & D.N. Dumville (ed.), Gildas. New Approaches (Woodbridge, 1984), pp.1-26
[vii] P. Sims-Williams, ‘Gildas and the Anglo-Saxons.’ Cambridge Medieval Celtic Studies 6 (1983), pp.1-30 (n.b. this is the same journal as the later Cambrian Medieval Celtic Studies; it switched title when its editor and creator changed universities!).
[viii] This point was drawn attention to, but never really exploited, by M. Miller, ‘Bede’s use of Gildas.’ English Historical Review 305 (1975), pp.241-61.
[ix] M. Lapidge, ‘Gildas’ education and the Latin culture of sub-Roman Britain’, in M. Lapidge & D.N. Dumville (ed.), Gildas. New Approaches (Woodbridge, 1984), pp.37-50.