Gildas is remembered as the author of De Excidio Britanniae (On the Ruin of Britain). The Anglo-Saxon historian Bede, who made use of De Excidio in his Historia Ecclesiastica, calls him the Britons’ “own historian” (HE I, 22), but, though he touches on history, it was not Gildas’ purpose to write a ‘history book’.

Whatever my attempt shall be in this epistle, made more in tears than in denunciation, in poor style, I allow, but with good intent, let no man regard me as if about to speak under the influence of contempt for men in general, or with an idea of superiority to all, because I weep the general decay of good, and the heaping up of evils, with tearful complaint. On the contrary, let him think of me as a man that will speak out of a feeling of condolence with my country’s losses and its miseries, and sharing in the joy of remedies. It is not so much my purpose to narrate the dangers of savage warfare incurred by brave soldiers, as to tell of the dangers caused by indolent men. I have kept silence, I confess, with infinite sorrow of heart, as the Lord, the searcher of the reins, is my witness, for the past ten years or even longer; I was prevented by a sense of inexperience, a feeling I have even now, as well as of mean merit from writing a small admonitory work of any kind.
De Excidio Britanniae §1

De Excidio is essentially a long sermon concerned with the evils which the Britons have brought on themselves by their sinful behaviour. Gildas singles out five contemporary British rulers for particular criticism (§§28–36), of whom the last to be addressed, though “first in wickedness”, is named, in Latin, Maglocunus. Gildas’ Maglocunus is conventionally identified with Maelgwn, king of Gwynedd (Mailcun rex Genedotæ), whose demise during a mortalitas magna (‘great death’, i.e. a plague) is recorded in Annal 103, equating to the year 547, of the the Annales Cambriae A-text. Since Gildas must have lambasted the king whilst he was still alive, a date in the region of 545 is often proposed for the composition of De Excidio. However, the validity of the unsubstantiated date attributed by the Annales Cambriae, a relatively late source, to Maelgwn’s death is suspect.[*]

The chronological water is further clouded by a remark Gildas makes – in the surviving text (§26) its meaning is not at all clear – concerning “the siege of Badon Hill” (an unidentified site, where the Britons defeated the Anglo-Saxons). Gildas seems to say (other interpretations are available) that he was born in the same year as the siege, and that he was writing in the forty-fourth year, one month of which had already passed, since the siege (i.e. he was writing 43 years and 1 month after). The Annales Cambriae (A-text, Annal 72) indicate that “the battle of Badon” took place in 516 (another date for which there is no substantiation). By these tokens, simple arithmetic would date De Excidio to 559, which is, of course, after Maelgwn is said to have died.[*] Bede, though, whose own copy of De Excidio must have been earlier than any surviving copy (the earliest is 10th century[*]), in his paraphrase of Gildas, says (HE I, 16) that the interval “about 44 years” was the time that elapsed between the arrival of the Anglo-Saxons in Britain and their defeat at the siege of Badon Hill – making no connection to the time when De Excidio was written.

Some scholars, drawing inferences from Gildas’ text, prefer to date De Excidio as early as the end of the 5th century.[*] The Annals of Ulster, however, place Gildas’ death in 570 (this date is echoed by the Annales Cambriae).

At any rate, the debate about just when Gildas was writing continues, as, indeed, does the debate about just where he wrote – the most popular suggestion seems to be somewhere in south-western Britain.

Gildas De Excidio Britanniae translated by Hugh Williams

The majority of entries for the 5th and 6th centuries in the Annales Cambriae are based on entries found in Irish annals (of which the Annals of Ulster is representative). The Annals of Ulster lists victims of a mortalitas magna s.a. 549. In an paper titled ‘Gildas and Maelgwn: Problems of Dating’, David Dumville writes:
We have no grounds whatever for supposing an antecedent Welsh annal which has been conflated with the Irish source text. Rather we must suppose that the tenth century compiler [of the Annales Cambriae], knowing from legend that Maelgwn had died in a plague,[*] made some approximate calculations (whether from a pedigree or from whatever synchronistic information was available to him) and concluded that that noted at ‘549’ in his Irish annals was most likely the one in question.
However, having dismissed the validity of the Annales Cambriae’s record of Maelgwn’s death, Dr Dumville, in a companion paper*, ‘The Chronology of De Excidio Britanniae, Book I’, concludes that “the usual broad dating of Gildas’s literary activity to the second quarter of the sixth century is correct” (though he adds “an argument for the third quarter might well be possible too”).
* The two papers were first published in Gildas: New Approaches (1984), and later in a compilation of David Dumville’s papers: Britons and Anglo-Saxons in the Early Middle Ages (1993).
The earliest surviving copy, British Library MS Cotton Vitellius A vi, mid-10th century, was badly damaged by fire in 1731. The manuscript had previously been employed to produce printed editions of De Excidio: by Polydore Vergil in 1525, and Josselin in 1568.
(There is a fragment, possibly copied in the late-9th or early-10th century, preserved in Bibliothèque Carnegie de Reims MS 414.)
Guy Halsall, for instance:
There is some evidence for an ‘early Gildas’, writing in the late fifth century. This includes Gildas’ rhetorical education, his Latin style, his theological concerns, and a rereading of his historical section and where he places himself within it. I tend towards this interpretation, although it cannot be proven. It is unlikely that Gildas wrote before 480/490 or much after about 550; beyond that we cannot go.
Worlds of Arthur (2013) Chapter 4
s.a. = sub anno = ‘under the year’.
Molly Miller argues that Gildas was actually referring to calendar years, and concludes that:
De Excidio was published in the February of a calendar year [because one month of the year had passed] from which, counting backwards and inclusively, the forty-fourth calendar year was that in the course of which Badon was fought and Gildas born.
‘Relative and Absolute Publication Dates of Gildas’s De Excidio in Medieval Scholarship’, Bulletin of the Board of Celtic Studies Vol. 26, 1974–76
Supposing this interpretation to be correct (it is an attractive proposition), and supposing the date suggested by the Annales Cambriae for Badon to be correct, then De Excidio could be precisely dated to February 559. Unfortunately, there is no reason to suppose that the Annales Cambriae’s date for Badon is correct. The entry itself is influenced by legend: “The battle of Badon, in which Arthur carried the Cross of our Lord Jesus Christ for three days and three nights on his shoulders and the Britons were the victors.”
A ‘Life’ of St Teilo in the 12th century Liber Landavensis (Book of Llandaff: National Library of Wales MS 17110E) says that Maelgwn was a victim of “the Yellow Pestilence”, which “destroyed his country”.