Skene's Four Ancient Books of Wales

CHAPTER III.
SOURCES OF THE EARLY HISTORY OF WALES.

   In order to discriminate between what is true and what is fabulous in the early history of Wales as presented to us in the historic literature subsequent to the twelfth century, and to disentangle the fragments of real history contained in them, so as to enable us to form something like a true conception of its leading features, it is necessary to test it by comparing it with the statements in contemporary authorities of other countries, and by referring to such earlier native documents as have come down to us. Of the latter class there are only three, and it is requisite that we should form a right conception of their authority. The first are the works of Gildas, who wrote in Latin. They are usually considered as consisting of two pieces, the Historia and the Epistola, but they may be viewed as forming one treatise. Questions have been raised upon the lives of Gildas, as to whether there was one or two persons of the name--an earlier and a later; but, viewing the question in its literary aspect, it is of little consequence, for the treatise is evidently the work of one man, and there is evidence in the work itself of his date. The writer states that he was born in the year in which the battle of Badon was, fought, and that he wrote forty-four years after.1 According to the oldest Welsh annals, the battle of Badon was fought in the year 516, which would place the composition of the treatise in the year 560; and the Irish annals record the death of Gildas in 570, ten years after.
   Only three MSS. of Gildas are known to have existed, and the oldest of these has since perished. It was in the Cottonian. Library (Vit. A. vi.), but fortunately the text of Josseline's edition of Gildas in 1568 was printed from it, and, according to Mr. Petrie, so correctly that it may be taken as representing it.2 The other two MSS. are in the public library at Cambridge (Dd, i. 17 and Ff, i. 27)--the one of the end of the fourteenth or beginning of fifteenth centuries, and the other of the thirteenth century. The first is said to have belonged to the monks of Glastonbury, and the second to the monks of Durham. This latter MS. inserts various passages which are not to be found in the other MSS. Thus the other MSS. mention that the Saxons were invited "superbo tyranno," and the Durham MS. inserts the words, "Gurthrigerno Britannorum duce." Again, where the other MSS. mention the "Obsessio Badonici montis," the Durham MS. inserts "qui prope Sabrinum ostium habetur." The work of Gildas had early found its way to the Northumbrian monks, as Bede evidently uses it in his history, and they are probably answerable for the additions contained in this MS. It has been remarked that the account given by Gildas of the departure of the Romans from Britain, and the events which followed, are inconsistent with the statements of contemporary Greek and Roman authors; but this appears to me to arise solely from Gildas having misplaced the only document directly quoted by him, which has forced upon his narrative a chronology inconsistent with the true sequence of events, and which, unfortunately, has likewise influenced Bede's history. Gildas narrates two devastations by the Picts and Scots, after each of which they were driven back by the Roman troops; then he states the final departure of the Roman army, followed by the occupation of the territory between the walls by the enemy. When he quotes this document, which purports to be a letter by the Britons, addressed "Actio ter consuli," imploring assistance against the "Barbari, who drive them to the sea, while the sea throws them back on the Barbari." He understands by these "Barbari" the Picts and Scots, and places after this latter the invitation to the Saxons, who first drive back the Picts and then unite with them to subjugate the Britons. Now the exact date when this letter must have been written can be at once ascertained, for Aetius was consul for the third time in 446, and the dates of the other events have been fixed in accordance with this; but while this postdates these events when compared with the other authorities, the sequence is the same, with the single exception of the place occupied by this letter. We know from Zosimus that the Roman army really left finally in 409. We see, from Constantius' Life of St. Germanius that the Saxons had already, in alliance with the Picts, attacked the Britons in 429; and Prosper, a contemporary authority, tells us that in 441 "Britanniæ usque ad hoc tempus variis cladibus evenitibusque latæ, in ditionem Saxonum rediguntur." It is impossible to mistake this language. The Saxons must have completed their conquest six years before the letter was written, and it follows that the "Barbari" to which it refers must have meant the Saxons, and that it was an appeal to the Romans to assist them against the Saxon invaders. The language of the letter, too, which seems exaggerated and inapplicable to the incursions of the Picts and Scots from the north, is much more natural if directed against the steady and permanent encroachment of the Saxons from the east. Take the letter from its present place, and place it after the narrative of the Saxons turning against the Britons and attacking them, and the order of events at once harmonises with the other authorities, while the necessity for postdating them in Gildas no longer exists. It was no doubt his misapprehending the meaning of this document, and misplacing it, which led to the arrival of the Saxons being supposed to have taken place after it, and to the date of 447, the succeeding year, being affixed to it by Bede.
   The second document is the work usually termed Nennius' History of the Britons, and it is very necessary that we should form a right conception of this work, and a correct estimate of its authority. The Origines, of Isidorus of Seville, who died in 636, and which must have been compiled some considerable time earlier, soon became widely known, and led to works being written in many countries upon their early history, in which the traditions of the people were engrafted upon it. Either in the same century, or the beginning of the next, a work was compiled in Britain, termed Historia Britonum. The author of it is unknown, but the original work appears to have been written in Welsh and translated into Latin. It seems to have acquired popularity at once, and become the basis upon which numerous additions were made from time to time. The original work appears to have belonged more to the North than to Wales, or at least the latter part of it, as the events of that part are mainly connected with the North, and it terminates with the foundation of the Anglic kingdom of Northumbria by Ida. Soon after was added what is termed the Genealogia, being the descent of the Saxon kings of the different small kingdoms; but here too Northumbria predominates, and most of the events mentioned in it are connected with its history. It must have been compiled shortly after 738, as that is the latest date to which the history of any of the Saxon kingdoms is brought down; and it too bears the marks of being a translation into Latin from Welsh. An edition of the Historia seems to have been made in 823, the fourth year of Mervyn Frych, king of Wales, by Marc the Anchorite, when that part at least of the text which contains portions of the life of Germanus, and probably the legend of St. Patrick, must have been inserted. Another edition in 858 bears the name of Nennius. The original work was very early attributed to Gildas, but latterly the whole work bore the name of Nennius.
   The oldest MSS. are of the tenth century, and are three in number. They represent two different editions of the work. The Vatican MS. bears the name of Marc the Anchorite, and contains the date of 946, and the fifth year of King Edmund. It is remarkable enough that this was the year in which that king conquered Cambria, and made it over to Malcolm, king of Scots. It would seem as if this conquest had brought it first under the notice of the Saxons, and this conjecture is further strengthened by the fact that the Paris MS. exactly corresponds with this, and that this MS. alone, of all the numerous MSS. which have come down to us, has the names, of the Saxon kings in the Saxon and not in the Welsh form.
   The MS. which represents the other edition is one in the British Museum (Harl. 3856). It contains in it the date of 796, but there are additions to it not found in any other MS., which must have been compiled in the year 977. These, are, first, a later chronicle of Welsh events, from the year 444, and though the last event recorded is in 954, the "anni" have been written down to 977; the second is a collection of Welsh genealogies, commencing with that of Owen, son of Howel dda, king of South Wales, who reigned from 946 to 985,--both in the paternal and maternal line,--from which we may infer that the writer was connected with South Wales. The Chronicle was made the basis of two much later chronicles, in which the events are brought down to 1286 and 1288, and the whole have been edited under the name of Annales Cambriæ, but the two later chronicles have in reality no claim to be incorporated with it, as the differences are not various readings of one text, but later additions. The great value of this Chronicle arises from the fact that it was compiled a century and a half before the Bruts were written, and it detracts from that value to add to it later additions taken from chronicles compiled as many years after the Bruts, and which are obviously derived from them. It is also the source from which many of the entries in the Welsh Brut y Saeson and Brut y Tywysogion have been translated. It is obvious that both the Chronicle and the Welsh genealogies were additions intended to illustrate the Genealogia attached to the Historia Britonum, and to bring the Welsh history down to the date of the compiler. The Chronicle inserts the events in the Genealogia in the very words of the latter; and when the Genealogia enumerates four Welsh kings as fighting against one of the kings of Bernicia, the Welsh genealogies give the pedigree of these four kings, in the same order.
   The Historia Britonum was translated into Irish by Giollacaomhan, an Irish Sennachy, who died in 1072, and various Irish and Pictish additions were incorporated in the translation.
   The work, therefore, as it existed prior to the twelfth century, may be said to consist of six parts: First, The original nucleus of the work termed Historia Britonum; second, The Genealogia, added soon after 738; third, The Memorabilia; fourth, The Legends of St. Germanus and of St. Patrick, added by Mare in 823, the latter being merely attached to his edition, and incorporated in that of Nennius; fifth, The Chronicle and the Welsh genealogies, added in 977; and, sixth, The Irish and Pictish additions, by Giollacaomhan.3 The MSS. of Nennius amount to twenty-eight in number; and of the later MSS. several seem to have been connected with Durham. To the monks of Durham many interpolations may be traced very similar to those in Gildas: in some MSS. they are written on the margin, and in others incorporated into the text. Thus, when the Mare Fresicum is mentioned, the Durham commentator adds, "quod inter nos Scotosque est." The result of my study of this work is to place its authority higher than is usually done; and, used with care and with due regard to the alterations made from time to time, I believe it to contain a valuable summary of early tradition, as well as fragments of real history, which are not to be found elsewhere.
   The third native authority prior to the twelfth century is The Ancient Laws and Institutes of Wales. They were published by the Record Commission of England in 1841, and the oldest of them, the Laws of Howel dda, are of the tenth century.
   Such are the native materials upon which, along with the old Roman and Saxon authorities, any attempt to grasp the leading features of the early history of Wales must be based.

Footnotes
1. Bede understood this well-known passage as implying that the battle of Badon was fought forty-four years after the arrival of the Saxons; but it is now generally admitted that this is a mistaken construction of the passage, and that the true import is as above, to which I also give my adhesion.
2. Josseline says it had belonged to Christ Church, Canterbury, and was 600 years old.
3. The original work will be quoted under the title of the Historia Britonum, the second portion under that of the Genealogia, or both generally as Nennius, and the fifth as the Chronicle and Genealogies of 977. The Irish Annals will be quoted from the Chronicles of the Picts and Scots, recently published, being the first of the series of Scottish Record publications.

1. Bede understood this well-known passage as implying that the battle of Badon was fought forty-four years after the arrival of the Saxons; but it is now generally admitted that this is a mistaken construction of the passage, and that the true import is as above, to which I also give my adhesion.
2. Josseline says it had belonged to Christ Church, Canterbury, and was 600 years old.
3. The original work will be quoted under the title of the Historia Britonum, the second portion under that of the Genealogia, or both generally as Nennius, and the fifth as the Chronicle and Genealogies of 977. The Irish Annals will be quoted from the Chronicles of the Picts and Scots, recently published, being the first of the series of Scottish Record publications.