¹ Chronica Minora, iii. 9; cf. Dumville, D. N., ‘Sub-Roman Britain: History and Legend’, History, lxii, No. 205 (1977), 173–92.CrossRefGoogle Scholar I cite this article and the following by author's name only: Barley, M. W. and Hanson, R. P. C., Christianity in Britain, 300–700 (Leicester, 1968)Google Scholar; Collingwood, R. G. and Myres, J. N. L., Roman Britain and the English Settlements (Oxford, 1937)Google Scholar; Freeman, E. A., Western Europe in the Fifth Century (London, 1904)Google Scholar; Jackson, K., Language and History in Early Britain (Edinburgh, 1953)Google Scholar; Lot, F., ‘De la valeur historique du De Excidio … de Gildas’, Medieval Studies in Memory of Gertrude S. Loomis (Paris-New York, 1927), 229–64Google Scholar; Stevens, C. E., ‘Gildas Sapiens’, English Historical Review, lvi (1941), 333–73CrossRefGoogle Scholar; Williams, H., Gildae De Excidio Britanniae, etc., Cymmrodorion Record Series, No. 3 (London, 1899)Google Scholar.

² Stevens, 354, as against, e.g., W. H. Davies apud Barley and Hanson, 138 f., and Gransden, Antonia, Historical Writing in Britain, i (London, 1974), 2 f.Google Scholar

³ 37 (48.15). References of this kind are to Gildas, the first figure giving the chapter, the figures in brackets the number of the page and line in Mommsen's edition. It is hypercritical, I think, of Grosjean, P., Analecta Bollandiana, lxxv (1957), 224Google Scholar n. 2, to assert that the phrase relates only to capp. 27–36.

⁴ 4 (29. 19).

⁵ 5 (29. 24 ff.), 8 (30. 21) interea. Augustus's renowned peace with Parthia in 20 B.C. was in fact the ‘first’ significant Parthian peace, and it was certainly a more celebrated event than Trajan's peace of A.D. 117, proposed by Williams, 19 n. 1, who is criticized by Lot, 233 f. Gildas's other mistakes in this part of his work lie in the motives which he ascribes to the Romans for their actions, 6, 7 (30. 5, 20).

⁶ 23 (38. 19), 28 (41. 28).

⁷ Capp. 6 f.

⁸ ibid.

⁹ 10 (31. 16). On the date of Alban's martyrdom see the conjectures discussed by W. H. C. Frend apud Barley and Hanson, 38. The incolis of 9 (31. 4) are the inhabitants of the Empire, not those of Britain.

¹⁰ 4 (29. 4), 7 (30. 14), 13 (33. 2, 7).

¹¹ 13 (32. 26 ff.); Chron. Gall. a. cccclii (i. 646), though some doubt is thrown on this by Seeck, Untergang, v. 497. The word strenuus is also applied to Maximus by Orosius, vii. 34. 9. Scholars have noted a similarity between Gildas, 13 (33. 7), and Sulpicius Severus, Vita Martini, xx. 2, but it is not close enough to prove a borrowing.

¹² 13 (32. 26).

¹³ Amm. Marc, xxviii. 3. 4 f., xxx. 7. 10, Zosimus, iv. 12. 2.

¹⁴ 4 (29. 6).

¹⁵ Zosimus, iv. 35. 4, John of Antioch, frag. 186. 2.

¹⁶ 6 (30. 5), 7 (30. 16).

¹⁷ 14 (33. 10), though perhaps the armato milite as well as the ingenti iuventute are both alike composed of Britons.

¹⁸ 14 (33. 13) multos … annos. Mr Todd objects that, if this were so, we might have expected the recovery of Britain – say, in the 390's – to have been mentioned in the eulogies of Stilicho.

¹⁹ vii. 17. 7, cf. Aurelius Victor, Caes. xx. 18, Epit. xx. 4, Eutropius, viii. 19, and even SHA Severus, xviii. 2.

²⁰ 14 (33. 12) primum, a word which Bede, HE i. 12, with his usual wisdom omits.

²¹ cf. Miller, M., ‘Stilicho's Pictish War’, Britannia vi (1975), 141–5.CrossRefGoogle Scholar

²² Britannia, viii (1977), 313 f.Google Scholar

²³ 21(36. 13 f.).

²⁴ 6 (30. 13), 14 (33. 18), 15 (33. 22) absque rectore; 18 (34. 21 ff. and 35. 4). One arm in which the Britons must have been immeasurably superior to the invaders was cavalry. If the Britons had shepherds, they had cavalrymen; whereas the Angles of ‘Brittia’, as Procopius, BG viii. 20. 29, says with some exaggeration did not even know what a horse was!

²⁵ 19 (35. 13).

²⁶ Lot, 262.

²⁷ Antonia Gransden, op. cit. (in n. 2 above), I.

²⁸ 25 (40. 12), cf. Mommsen, Chron. Min.. iii. 10.

²⁹ 4 (29. 20).

³⁰ 3 (28. 8).

³¹ Williams, 14 n. 1.

³² 4 (29. 17).

³³ 4 (29. 16), cf. Migne, Patrol. Lat. xxiii. 634.

³⁴ Chron. Min. iii. 7.

³⁵ Levison, W., ‘St Alban and St Alban's’, Antiquity, xv (1941), 337–59CrossRefGoogle Scholar, at 340, 348 with n. 18. On Gildas's reading see also F. Kerlouégan apud Barley and Hanson, 158 f., ‘… sinon des textes in extenso, du moins des extraits importants de ces textes’.

³⁶ Mommsen, Chron. Min. iii. 6.

³⁷ Lot, 233.

³⁸ 1 (25. 5.).

³⁹ 12 (32. 21). He also mentions heresy at 1 (26. 26); 67 (63. 24); 75 (69. 6).

⁴⁰ Few will follow Myres, J. N. L., ‘The Adventus Saxonum’ apud Grimes, W. F. (ed.), Aspects of Archaeology (London, 1951), 221–41Google Scholar, at 227 f., in supposing that ‘the acceptance of Satan for an angel of light’, 21 (37. 4), and other phrases, are indirect references to Pelagianism. If so, they are indeed indirect. The opposite conclusion from mine is also drawn by Williams, 18 fin., but Lot, 235, agrees that Gildas knew nothing of Pelagius or of the Vita of Germanus.

⁴¹ Mossir, J., J. Theolog. Stud, xvi (1965), 36Google Scholar, cf. Migne, Patrol. Lat. xxx. 167 B. The sentence is also found, but with non enim actum est in place of non agitur, in the Pelagian De Operibus, xiii. 1 (Caspari, 90). I am not persuaded by the remarks of Davies apud Barley and Hanson, 140, that ‘perhaps some lingering Pelagian influence may be detected in Gildas himself’. The passages cited in his n. 105 do not bear out this opinion.

⁴² On the author of the De Virginitate see, besides the standard works on Pelagianism, Morris, J., ‘Pelagian Literature’, J. Theolog. Stud., xvi (1965), 26–80, at 36CrossRefGoogle Scholar; Hanson, R. P. C., St Patrick: His Life and Times (Oxford, 1968), 44 ff.Google Scholar

⁴³ Vita Germani, xxvii, though admittedly he says in illis locis, not in Britannia. Professor Markus points out that Constantius is perhaps trying to credit his hero with success in a distant land about which he and his readers knew little. On the end of Pelagianism see Plinval, G. de, Pélage: ses écrits, sa vie, et sa réforme (Lausanne, 1943), 382Google Scholar ff.

⁴⁴ 26 (41. 1 ff.)

⁴⁵ 65 (61. 28 ff.)

⁴⁶ 65 (61. 23), with Chadwick, O., ‘Gildas and the Monastic Order’, J. Theolog. Stud., v (1954), 78–80CrossRefGoogle Scholar, with Grosjean, P.apud Gordon, D. J. (ed.), Fritz Saxl, 1890–1948: A Volume of Memorial Essays (Edinburgh, 1957), 70Google Scholar f.

⁴⁷ 69 (64. 17), 92 (77. 3), 110 (85. 23 f.).

⁴⁸ 50 (55. 26).

⁴⁹ Williams, 64 n. 1, 161 n. 1; Hughes, Kathleen, The Church in Early Irish Society (London, 1966), 43Google Scholar, thinks it possible that he is referring to the monks.

⁵⁰ Collingwood and Myres, 316.

⁵¹ 17 (34.6).

⁵² 7 (30. 19).

⁵³ These are some good remarks in Freeman, 154 f.

⁵⁴ 19 (35. 10 ff.). Stevens, 360, and Morris, John, The Age of Arthur (London, 1973), 44Google Scholar ff., date these events to 410–11 or even earlier, thus giving the Britons some 35 years of defeat before they thought of appealing to the Western government for help. Even British hesitancy could hardly have reached these dimensions!

⁵⁵ So Stevens, 360. True, at 21 (36. 17) in extrema parte insulae does indeed have this sense – unless Gildas is telling us there (rightly or wrongly) of Pictish settlements south of the Wall. The story of the migration of Cunedda and his sons from Mannau Gododdin to North Wales is to be rejected: wick, Nora Chad, Studies in the Early British Church (Cambridge, 1958), 32–4Google Scholar; Dumville, 182. But in view of Hanson, op. cit. (in n. 42 above), 22 n. 4, the question deserves re-examination.

⁵⁶ Chron. Min. i. 660, with Mommsen's discussion, ibid. 618.

⁵⁷ Constantius, Vila Germani, 25 ff. There is no reason whatever for thinking that Germanus' second visit to Britain is a mere doublet of the first: contra, Chadwick, Nora K., Poetry and Letters in Early Christian Gaul (London, 1955), 255 ff.Google Scholar

⁵⁸ 19 (35. 20).

⁵⁹ 19 (35. 26). It is beside the point that at 35. 26 the MS A reads insula for regio.

⁶⁰ e.g. Collingwood and Myres, 313; Jackson, 113. Some scholars believe that Aetius had a fourth consulate in 454. But the Aetius who was consul then was an Eastern general and must be distinguished from the Patrician: Seeck, P.-W. i. 703.

⁶¹ 20 (36. 5), cf. the plague mentioned on p. 35 below. On this and other plagues of the period see Todd, M., ‘Famosa Pestis and Britain in the Fifth Century’, Britannia, viii (1977), 319–25.CrossRefGoogle Scholar

⁶² 20 (36. 7).

⁶³ 21 (36. 16 ff.). Miller, M., ‘Bede's Use of Gildas’, English Historical Review, xc (1975), 241–61CrossRefGoogle Scholar, at 247, 260, makes the fascinating suggestion that the earlier, unquoted part of the letter to Aetius may have been Gildas's source of information about the wars of the years preceding 446. If so, Gildas is not likely to have been unaware of the identity of the Britons' enemies in those years, as Stevens and others would have us believe: see p. 214.

⁶⁴ The first was in 409: Britannia, viii (1977), 313Google Scholar ff.

⁶⁵ 20 (36. 9) where note the imperfect strages dabant: there were several strages.

⁶⁶ Miller, art. cit., 260, is wrong, in my opinion, to say that at interea at the beginning of cap. xxii Gildas turns from the north to the south of the island. In fact, he does not give the slightest hint that the scene of events has changed or that he is now talking of a different area of Britain. It is not the case that in cap. xxi Gildas is bringing the history of the north down to his own time: he does not do that until cap. xxvi. In xxi he is speaking of the period after the British victory and is moralizing about it: and interea at the beginning of xxii simply resumes the narrative from 20 (36. 12).

⁶⁷ Bede, HE i. 15 init.; Gildas, 22 (37. 23). The ‘old enemies’ ought to include the Irish, but Gildas's narrative goes on as if the Picts alone were in question.

⁶⁸ 2 (28. 3). See Todd, art. cit.

⁶⁹ 23 (38. 12). Whether Vortigern's name ought to be inserted in the text at line 12 may be decided when we have a new edition of the De Excidio: Dumville, 184. Such a translation as ‘outstanding leader’ is misleading.

⁷⁰ 23 (38. 17). So Williams, 52 n. 1.

⁷¹ Hydatius, 171 (ii. 28). Invading forces carried in three ships are recorded in the Anglo-Saxon Chronicle, s.aa. 477, 514, 787 (789), or in five ships, s.aa. 494 (preface), 495.

⁷² 22 (37. 22) inhabitare, but for some reason Bede, HE i. 14, omits this point.

⁷³ Stevens, 365, thinks that the Elaphius, primus regionis, of Constantius, Vita xxvi, was the ‘first man of Britain’, since regio here, he thinks, means Britain itself. This is hardly credible.

⁷⁴ 23 (39. 1 f.).

⁷⁵ JRS xlvi(1956), 66.Google Scholar

⁷⁶ 23 (39. 7), cf. Stevens, 369. Epimenia is also found in Juvenal. In view of the frequency of the word in Greek it is doubtless an accident that it does not recur in Latin documents. I find it hard to follow Mommsen, Chron. Min. iii. 6, in thinking that Gildas had read Juvenal (or Persius, Martial, and Claudian, for that matter).

⁷⁷ 23 (39. 6 f.), where multo tempore need not mean much more than three or four years – after all, Gildas is a rhetorician.

⁷⁸ Kenney, J. F., The Sources for the Early History of Ireland (re-printed Dublin, 1966), 148Google Scholar, remarks that ‘it is noteworthy that from this date we hear but little of plundering expeditions sent out from Ireland’. He suggests that ‘possibly the progress of Christianity, especially over the central and northern parts of the country, may explain the change’. But in fact the conversion of a barbarian people to Christianity hardly ever diminished its enthusiasm for attacking the Roman provinces. Plunder is more attractive than a principle.

⁷⁹ See e.g. Myres, J. N. L., Anglo-Saxon Pottery and the Settlement of England (Oxford, 1969), 62 ff.Google Scholar

⁸⁰ 23 (39. 8).

⁸¹ 24 (39. 10 ff.).

⁸² 25 (40. 1 ff.). The remark of Stevens, 369, that ‘an agricultural labourer remained as useful to a barbarian as to a British master’, is untrue in view of the low stage of economic development that the Saxons had reached at this date.

⁸³ Gildas does not say that Ambrosius was descended from parents who had worn the purple. What he says 25 (49. 13), is that they had ‘evidently’ (nimirum) been clad in the purple. It is a conventional form of praise and no inference about their social status should be drawn from it. Was not Gildas himself said to have been one of the twenty-four sons of a king of the Picts? Caradoc, Vita Gildae i (iii. 107).

⁸⁴ Dumville, 183 f.

⁸⁵ 10 (31. 18).

⁸⁶ Stevens, 360.

⁸⁷ 20 (36. 8).

⁸⁸ I do not know why Myres, art. cit. (on p. 211, n. 40 above), 224, says, ‘It is clear that Gildas is thinking of this event (the foedus with the Saxons) as happening at least twenty or thirty years after the appeal to Aetius’. And so he transfers the events which led up to the appeal to the years following upon 410 (ibid. 227). See p. 214, n. 54 above.

^88a 3 (28. 14).

⁸⁹ 24 (39. 17 ff.), cf. 26 (40. 21) desertae dirutaeque.

⁹⁰ 5 (30. 2), 24 (39. 18).

⁹¹ 19 (35. 12). I do not know what to make of the barbed shafts with which they pulled the citizens off the city walls and dashed them to the ground: 19 (35. 17).

⁹² 6 (30. 9 f.).

⁹³ See the passages cited by Thompson, E. A., A Roman Reformer and Inventor (Oxford, 1952), 71Google Scholar f. For earlier times see Oxford Latin Dictionary, s.v. ‘quadratus‘, 3d, esp. Seneca, Ep. 59. 7, and Thesaurus Ling. Lat., i. 1344. 20 ff.

⁹⁴ 15 (33. 14 ff.), cf. Lot, 236 n. 1.

⁹⁵ 25 (40. 3 ff.), cf. 4 (29. 21).

⁹⁶ Mansi, vii. 947.

⁹⁷ Sidonius, Ep. iii. 9.

⁹⁸ Sidonius, Ep. i. 7. 5; Jordanes, Get. xlv. 237. Stevens, C. E., Sidonius Apollinaris and His Age (Oxford, 1933), 138Google Scholar f., dates Sidonius, Ep. iii. 9, after the defeat: ‘the Breton troops, with all discipline gone, were roaming over the country is disorderly bands’. I have followed (for no very obvious reason) Fahy, D., ‘When Did Britons Become Bretons?’, Welsh History Review, ii (1964), 111–24Google Scholar, at 120, who remarks that the end of this letter hardly suits defeated men. Whether they came from Britain or Britanny is not a very meaningful question: if from Britanny, they had only recently arrived there from Britain.

⁹⁹ Sidonius, Ep. ix. 9. 6 ff. On the name Riocatus see Jackson, 457.

¹⁰⁰ See e.g. Maximus of Turin, Sermo lxxxii. 2 (pp. 336 f., ed. Mutzenbecher).

¹⁰¹ Jackson, 13.

¹⁰² D. Greene apud Barley and Hanson, 81, to whom I am heavily indebted here.

¹⁰³ Prosper, Chron. 1307 (i. 473). For the view that a high proportion of the fourth- and fifth-century Roman objects found in Ireland were brought there by refugees from Britain see Bateson, J. D., ‘Roman Material from Ireland’, Proceedings of the Royal Irish Academy, vol. 73, Section C (1973), 21–97Google Scholar, at 30.

¹⁰⁴ For this trade see a brief and lucid summary with bibliography in Adcock, Leslie, Arthur's Britain (Pelican Books, London, 1973), 201 ff.Google Scholar

¹⁰⁵ 3 (28. 12).

¹⁰⁶ Kathleen Hughes, op. cit. (in n. 55 above), 49.

¹⁰⁷ 67 (63. 13, 64. 1.).

¹⁰⁸ Zosimus, vi. 10. 2.

¹⁰⁹ Procopius, Bell. Vand. iii. 2. 38.

¹¹⁰ Vita Germani, 18.

¹¹¹ 30 (43. 6).

¹¹² 33 (44. 24 ff.).

¹¹³ But see 26(36. 13).

¹¹⁴ 30 (43. 8 f.).

¹¹⁵ On his name see some interesting remarks in Jackson, 169 f.

¹¹⁶ 31 (43. 20). For his tombstone, found at Castell Dwyran, see Nash-Williams, V. E., The Early Christian Monuments of Wales (Cardiff, 1950), 138Google Scholar: he calls himself protictor.

¹¹⁷ 28 (42. 3).

¹¹⁸ 33 (45. 7).

¹¹⁹ This is suggested as a possibility by Margaret Deanesly, ‘The Implication of the Term Sapiens as Applied to Gildas’, apud Gordon, op. cit. (in n. 46 above), 53–70, at 68.

¹²⁰ 28 (41. 29 ff.).