Paul Remfry is an expert on the Norman castles of Wales and the families who owned them, especially along the Welsh Marches of the old Angevin empire, with at least 39 books and other publications to his credit. In this volume Remfry provides the first complete translation into English of the Cambro-Latin "Annals of Wales," which survive in the five manuscripts listed in his title. These he labels Texts A-E for convenience. The first three, Texts A-C, were edited by J. Williams Ab Ithel for the Rolls Series in 1860 and derive from a lost set of annals begun almost certainly at St. David's in southwest Wales around the year 795. The first entry is for the year 453 with very sporadic additions thereafter based upon Irish, Northern British and North Welsh sources. Entries become more regular after the year 863 and Text A, British Library MS Harleian 3859, concludes with a notation for the year 954 (= 953). This version most closely represents the original text of the Annales Cambriae as it was completed in the mid-tenth century. It was copied in its current form, Remfry concludes, sometime between the years 1130 and 1200. Texts B and C were copied in the late thirteenth century, but stem ultimately from the same mid-tenth-century archetype as A, continuing their entries through the year 1286 (= 1284) and 1288 (= 1289), respectively. Text D, the Cronica de Wallia in Exeter Cathedral Library, also derives from the lost tenth-century archetype. Somewhat unusually, Remfry has chosen to include a fifth text (E), the Exchequer Domesday of the Public Records Office, even though this manuscript has no direct or demonstrable connection to the original Annales Cambriae. It was compiled from different sources at the end of the thirteenth century with an interest in the lordship of Glamorgan, beginning sub anno 600 (= 597) with sharply increasing content for the years 1066-1298. It is translated here for the first time and included in the collection in order to bring wider accessibility to these various annalistic records of early and later medieval Wales.
As noted above, the best surviving text of the Annales Cambriae and closest to its original form is that in Harleian 3859, where it is inserted after chapter 66 of the Historia Brittonum (History of the Britons), attributed to Nennius and compiled in northern Wales from historical and legendary sources ca. 829-30. Both texts are edited and translated by John Morris in Nennius: British History and The Welsh Annals (1980), with some additions to the Annales Cambriae from Texts B and C as well. However, since there is still no critical edition of the Annales Cambriae in which its four surviving versions, Texts A-D, have been collated, Remfry provides in its place what he calls a "critical translation" (1), that is, one which follows Text A as edited by Ab Ithil for the first five hundred years from 453-953--the coverage of the original tenth-century text as preserved in Harleian 3859--then switches to Ab Ithil's edition of Text B through 1286, with further additions from Texts C, D and E "where necessary" up to the year 1298 (40). The different manuscript sources of these entries are clearly marked. In addition, a separate translation of each one of the five texts follows this composite rendering.
The information Remfry supplies in his introduction and notes is often useful for clarification and reference, but also sometimes incomplete, cryptic, incorrect, or uninformed by recent scholarship. For instance, under the year 516, Text A reads in Latin: Bellum Badonis, in quo Arthur portavit crucem Domini nostri Jhesu Christi tribus diebus et tribus noctibus in humeros suos et Brittones victores fuerunt, which Remfry translates: "The Battle of Badon, in which Arthur carried the Cross of our Lord Jesus Christ for three days and three nights on his shoulders (upper arms) and the Britons were victorious" (42). In his introduction, Remfry suggests that this entry "would appear to owe more to Geoffrey of Monmouth (d. 1154) and folk law, than to any original source" (4). It is true that Text A in Harleian 3859 was quite possibly copied after Geoffrey's Historia Regum Britanniae (History of the Kings of Britain) appeared around 1136. But essentially the same entry appears in Texts B and C as well, which derive from the same tenth-century archetype as A, with the only difference being that Text B designates Arthur as a "king." This similarity suggests that the entry was indeed part of the original Annales Cambriae and not interpolated from Geoffrey's account of Arthur's lifting of the Saxon siege of Bath, which event that author leaves undated. The site of the Battle of Badon, by the way, has never been confidently located, though it is also mentioned by Gildas in his sixth-century De Excidio Britanniae (On the Ruin of Britain) as the obsessio Badonici montis (siege of Mount Badon). In his foot-of-page note to this entry for 516 Remfry confidently suggests that "Bellum Badonis most likely occurred at the hill fort above Bath in Somerset" (42, note 67), thus making Remfry himself rather than the scribe of Text A the one who is imposing a Galfridian interpretation upon the location of this battle in the Annales Cambriae. In addition, Remfry notes that "Nennius' poem suggests that the siege lasted three days." Not all readers will know that the translator is referring to chapter 56 of Nennius's Historia Brittonum, where that author lists in prose twelve battles fought by Arthur culminating in the bellum in monte Badonis (battle on Badon Hill). (Nowhere does Nennius call this battle a siege--that was Gildas--nor does he refer in any way to its duration, except that in one day 960 Saxons fell in a single charge by Arthur, "and no one laid them low save he alone.") By Nennius's poem Remfry is presumably referring to the fact that the prose battle list, some of whose place-names rhyme, has suggested to Welsh scholars that it was taken from a vernacular praise poem, like those attributed to the sixth-century Northern British poet Taliesin, in which similar series of victories are celebrated.
Furthermore, in his translation itself, Remfry offers a parenthetical gloss upon "shoulders" as "upper arms," apparently finding this a more plausible location for Arthur's bearing an image or replica of the Cross. In an earlier battle in Nennius' list, however, Arthur is said to have worn an image of the Holy Mary Ever-Virgin super humeros suos (upon his shoulders), again suggesting to Welsh scholars a vernacular source for this information in that humeros may be a mistaken translation of ysgwyd (shield), easily confused with ysgwydd (shoulder), which words were sometimes spelled ambiguously in medieval Welsh orthography. John Morris accepts this view in his 1980 edition and translation of the Historia Brittonum and the Annales Cambriae in Harleian 3859, correcting in square brackets his translation in both texts from "shoulders" to "shield," which indeed makes better sense for the location of these images on Arthur's person. Remfry seems unaware of this widely accepted interpretation, unless he wishes us to understand "upper arms" as a metonymy for "shield." Even more puzzling is his suggestion that the entry quoted above is based upon "folk law," the substance of which he neglects to specify. One is tempted to suppose that Remfry means "folk lore" here, possibly referring to the martial hyperbole of medieval Welsh oral tradition or its characteristic "triplism" in reference to a battle that is said to have lasted three days and three nights. Such triple patterns are preserved in texts like the Trioedd Ynys Prydain (Triads of the Island of Britain). Others might see Arthur's three-day deliverance of his Christian compatriots from the heathen Saxons as modeled on Christ's Harrowing of Hell between the Crucifixion and Resurrection as described in the popular Gospel of Nicodemus. In sum, the inadequate note and commentary on the entry sub anno 516 somewhat damages this reviewer's confidence in the information supplied in other notes upon which he is less well informed.
To be fair, Remfry is far more interested in and knowledgeable about Wales in the Norman era. He usefully provides a map of 66 political subdivisions of the country referred to in these texts, as well as an index of the surnames that appear therein, plus 25 neatly annotated genealogical charts adapted from P. C. Bartrum's authoritative Welsh Genealogies AD 300-1400, 8 volumes (1974). These include the kings, princes and lords of the various Celtic-speaking kingdoms and other polities of western and northern Britain in the early and later Middle Ages, followed by a general index of names, places and topics. There is no bibliography or list of works cited.
Unfortunately, the production values of the volume are weak. Abbreviations and punctuation marks are inconsistent, many pages come loose upon first opening and turning, and the attractive cover illustration of a solitary tonsured scribe at his easel is mislabeled "Llywelyn ab Iorwerth on his deathbed with his sons Gruffydd and Dafydd from Mathew Paris." This volume thus does not replace John Morris's edition and translation of the Annales Cambriae in Harleian 3859, but does provide a convenient further reference for students of later medieval Welsh history and society, who now have to hand a ready if imperfect resource.