Folk-Lore/Volume 4/Celtic Myth and Saga. Report of Research during the years 1892 and 1893
CELTIC MYTH AND SAGA.
Report of Research during the Years
1892 and 1893.
IT is but fitting that Folk-Lore, the one review published in England which concerns itself with the history and literature of the Celtic races, should pay its tribute of sorrowful respect to the memory of two veterans of Celtic study departed within the last year.
Hector Maclean was the right-hand man of Campbell of Islay in his admirably achieved task of collecting and preserving the oral literature of the Gaelic Highlanders. He had all the qualifications of a great collector, intimate knowledge of the people, mastery of and sympathy with their modes of thought and expression, keen enthusiasm, and untiring patience. No higher praise can be given him than that he was worthy to be Campbell's lieutenant.[1]
Hector Maclean was a collector. Geheimrath Albert Schulz, better known by his pseudonym of San Marte, was a book-scholar. He shared with Maclean a keen and lasting interest in all that related to the legendary past of the Celt. It was but the other day (Folk-Lore, 1890, p. 255, note) that I noticed the last work of the veteran, a contribution to that elucidation of Wolfram's great Grail poem which he had begun sixty years previously, and which engaged his best energies throughout his life. In addition to his work on Wolfram, he first made the Mabinogion known on the Continent; he edited Gildas, Nennius, and Geoffrey of Monmouth; he collected and edited the texts relating to or connected with Merlin; he was one of the first to systematically investigate the origin and development of the Arthur romantic cycle. His works, outgrown in many respects as they are by the progress of study, will always remain landmarks in the history of Celtic scholarship, and even if they cease to be consulted, will be kept alive by the generous and lofty enthusiasm which inspires them.
The important text edited by Father Hogan raises afresh the question of the origin, date, and development of the Irish epic romances. It should be premised that the tale in question, the Battle of Ruis na Rig, is obviously a sequel to the Tain bo Cuailgne, intended to satisfy the curiosity, felt at all times and in all countries, concerning the after history of the heroes of a famous story. The existence of a considerable mass of heroic saga, as well as that of a school of epic narrators, are thus presupposed by our text, and any results which legitimately arise from a consideration of the way in which it has come down to us apply with far greater force to the older stratum of storytelling. Two versions are known, that of the Book of Leinster (the redaction of which cannot be later than 1150) and that of a number of modern MSS. belonging to the 17th-18th centuries. These latter represent a form of the saga differing from that in the Book of Leinster, a form which, as shown by the details of life and customs, must have been redacted at a considerably later date. But the MSS. of this later version, although of comparatively recent date, "exhibit many archaic inflexions, old vocables, and MiddleIrish survivals" which, in the editor's opinion, "seem to show that this version represents one coeval with that found in the Book of Leinster."
We thus have two texts substantially dating back to the 12th century, and neither of which, in its present form, can have been redacted before the 11th century, as is proved both by the texture of the language and the occurrence of personal and geographical names unknown in Ireland much before that time. But one of these texts, that preserved by the later MSS., must, substantially, be considerably younger than the other, as facts to be adduced presently amply prove. What follows? That the Book of Leinster version, although in language, and occasionally in geographical and historical nomenclature, a product of the 11th–12th centuries, belongs, so far as the matter is concerned, to a far earlier period.
What then are the differences between the two versions which warrant their assignment to different periods of national development? In the younger version the heroes wear coats of mail, "stout wonderful foreign armour"; "foreign cavalry" form a part of the forces; the war chariots, though mentioned, play no part. In the Book of Leinster version, on the contrary, the chariot is still the material unit of the army; the hero is practically armourless, and covers himself solely with shield and sword. In fact, the one version pictures the fighting of pre-Viking (i.e., pre-800 A.D.), the other that of post-Viking Ireland.
Thus we see how, when the stress of the Viking incursions had died away, the storytellers and scribes who gathered up the tales of olden time went to work. In some cases—e.g., the Book of Leinster version of our tale—they contented themselves with putting the old saga into language of the day and embellishing it with foreign names, in others they translated the material conditions as well as the language of their models. In this instance the second mode approved itself the more acceptable. The Book of Leinster version was apparently neglected by later copyists, whereas the rival one must have been transcribed frequently before reaching the 17th–18th century texts which alone have come down to us.
The literary problems which the story raises are perhaps more interesting than the tale itself, yet it contains some picturesque and admirable touches; we assist at the bivouac of the invading Ulstermen: "their fires were kindled, cooking of food and drink was made; baths of clean-bathing were made by them, and their hair was smooth-combed; their persons were cleansed, and tunes and merry songs and eulogies were sung by them." Nor can we easily find a finer example of old Irish chivalry of feeling (by the modern editor rightly and characteristically condemned as foolishness) than the statement: "for Conchobar concealed not even from his enemy the place in which he would take station or camp, that they might not say that it was fear or dread that caused him ot to say it." Most characteristic, too, is the way in which the heroes revile their adversaries and belaud themselves, as well the habit of rapid sententious dialogue, so pithy that each phrase is almost a proverb.
Like many of the oldest examples of Irish storytelling, the Battle of Ruis na Rig is in alternate prose and verse, the great variety and complexity of metre in the latter being remarkable. But it is noticeable that the apparently oldest verse portions are in the so-called rosc, a measure distinguished by no stanzaic form and no rhyme, but by alliteration and a "certain laconic and oracular diction". In this measure have likewise come down to us pieces that profess and approve themselves among the oldest remains of Irish speech, such as the so-called lorica of Patrick, the formulae of the Brehon laws, etc. It has generally been held that metrical complexity and rhyme are both early characteristics of Irish verse which in these respects, it has been maintained, has influenced both Latin and Scandinavian versification. But rosc would seem to be the protoplasm out of which the very complex Irish metres developed, and its persistence in texts so comparatively modern as the 11th–12th century would show either that the complex metres are younger than is generally supposed, or throws back the date of the rose poems to a very early period, proving, moreover, that there must have been a written or a very strong oral tradition to allow of their preservation.
Mr. Coffey's admirable monograph upon the great group of funereal tumuli and inscribed stones at New Grange forms, though only incidentally, one of the most important contributions ever made from the archæological side to the study of Irish legend and romance. It would be important merely for the fact that it prints and translates a number of 10th-11th century texts relating to these monuments. But it does far more than this. Mr. Coffey's archæological inquiry defines with as much precision as is likely to be obtained the nature and date of these monuments, and thus furnishes a series of fixed points by which we can estimate the nature of the traditions he prints from mediaeval Irish sources. Mr. Coffey, on purely archæological grounds, is inclined to date the New Grange tumulus "approximately about the first century (A.D.)", the Dowth tumulus being possibly somewhat earlier. Now the passages quoted by Mr. Coffey from texts which cannot be later than the early loth century show that the antiquaries of the time had a tradition that the burial-place at Brugh was used by the kings of Ireland from the days of Crimthann Niadh-nar to that of Loeghaire, son of Niall, with the exception of three kings, Art mac Conn, Cormac mac Art, and Niall of the Nine Hostages. Elaborate stories are told to account for the absence of the first two of these monarchs from the customary burialplace of their race, the purport of which is to connect them with Christianity, and thus, implicitly, to insist upon the pagan nature and associations of the New Grange monuments. The date of Crimthann is given by the Four Masters as A.D. 9, that of Loeghaire (the contemporary of St. Patrick) as A.D. 429. As Mr. Coffey remarks, "the evidence discussed in regard to New Grange would bring some of the tumuli in question within that period."
Here we have apparently a very remarkable convergence of testimony archaeological and historical, and there would seem good warrant for asserting both that the New Grange graveyard was started in the early years of the Christian era by the high-kings of Ireland, and also that the dates ascribed to these kings by the 10th–12th century annalists are substantially correct. But the question is a great deal more complicated than appears at first sight. For the very same texts which mention the fact that Crimthann was the first high-king of Ireland buried at New Grange, also insist most strongly upon the importance of the district as the burial-place of the Tuatha de Danann, that euhemerised race of ancient deities who, in the 10th-12th century annals, figure as genuine kings and heroes A.M. 3300-500. Indeed, Crimthann is definitely stated to have fixed his burial-place at Brugh, instead of at Cruachan, where his ancestors were interred, because his wife Nar was of the Tuatha De.
All later romantic tradition in Ireland connected with the Brugh district is concerned, not with what we may provisionally assume to be historic, the first-fifth century burial-place of the high-kings of Ireland, but solely with the legendary burial-place of the Tuatha De.
Mr. Coffey would account for these facts as follows. "The association of particular monuments with the Dagda and other divinities and heroes of Irish mythology implies that the actual persons for whom they were erected had been forgotten, the pagan traditions being probably broken by the introduction of Christianity. The mythical ancestors of the heroes and kings interred at Brugh, who, probably, were even contemporarily associated with the cemetery, no doubt subsequently overshadowed in tradition the actual persons interred there" (p. 82).
But is it likely that the "contemporary association", which Mr. Coffey assumes, existed unless there was some basis of fact for it, unless, that is, Crimthann really did choose an ancient hallowed spot for his burial-place? And is it not strange that the introduction of Christianity should, ex hypothesi, have "broken the pagan traditions" connected with the high-kings of Ireland and left whole the far more pagan traditions connected with the Tuatha de Danann?
Future archæological investigation may perhaps tell us if there are in the Brugh district traces of older burial than that of the first century Irish kings, or of an overlapping or mixture of races such as would seem to be implied by the historical tradition.
One point should be noted in view of recent controversies as to the origin of the belief in fairies. This belief, as still held by the Gaelic-speaking peasants of Ireland and Scotland, is, essentially, the same as that found in the Irish premediæval and mediæval romances concerning the Tuatha de Danann. As early as the 10th century at least, and probably very much earlier, the Tuatha De were prominently associated with the monuments in the Brugh district, and these monuments are not the dwelling-places of any former dwarf races, but, without doubt, served as a burial-place to the ancestors of present Irishmen.
To praise Mr. Standish Hayes O'Grady's Silva Gadelica is an easy matter. The first requisite in the study of Celtic antiquity and literature is still the publication and translation of texts, so that the bringer of such a stately pile of sheaves gathered from eight centuries of Irish storytelling and comprising many of the remains of Irish romance most interesting to the artist, most valuable to the historian, cannot but be sure of a hearty welcome. And when the gathering is made by a scholar who joins to a native knowledge of the Irish language and literature greater than that of any other living man, wide familiarity with literature at large, and acquaintance with the methods and results of historical and philological criticism, the welcome is intensified. A work such as Mr. O'Grady's at once takes rank as a classic in its line of study, and the critic best pays his due of admiration and respect when he treats it as a classic to be studied with minute and searching attention. In the remarks that follow I shall take it for granted that the book is in the hands of every serious student of Celtic lore, to whom I shall endeavour to be of use by supplementing the information to be found therein or by challenging statements and conclusions for which there seems to me to be insufficient warrant.
As it is probable that Mr. O'Grady's version will for some time to come be the standard of quotation for nonspecialist students of Celtic matters, it is necessary to say a few words as to the way in which he has dealt with his texts. I do not refer to the Irish original; I must needs, it is true, point out that in the opinion of other Irish scholars Mr. O'Grady has deprived his collection of value to the philological student of Irish by his practice of largely modernising the texts he draws from MSS. ranging in date from the nth to the 18th centuries. He has, in fact, edited his Irish on the system used by Mr. Henry Craik in his recently published English Prose from the 14th to the 16th Century. The system is a defensible one, and as folklorists the matter does not affect us save remotely. But if an editor deliberately discards philological merit for his texts, is it too much to ask that he should also discard the shackles which strict philological accuracy imposes! Of what use is it to print an imperfect iitheentury text when a perfect 14th-century one exists, save as a specimen of 11th-century form of speech? Yet Mr. O'Grady, while refusing to supplement the 11theentury scribe even where the latter can be proved to have skipped a couple of lines in his transcript from an earlier MS., as steadily refuses to give the exact grammatical forms of the version he in other respects slavishly follows! An example will make this plain.
The Boroma is one of the most important tales edited and translated by Mr. O'Grady. It so happens that Mr. Whitley Stokes, whilst Silva Gadelica was passing through the press, published text and version of the same tale in vol. xii of the Revue Celtique. Means is thus afforded to the non-Celtic student of testing the method of editing and translating of both scholars. One singularly reassuring result of the comparison between the two versions is that for practical purposes Middle Irish has been mastered; substantially, the two renderings, made independently of each other, agree.[2] The Boroma, which tells of the tribute levied upon Leinster by an over-king of Ireland in the second century, and continued by his successors until the seventh century, is preserved mainly in two MSS., the 12th-century Book of Leinster and the 15th-century Book of Lecan. Mr. O'Grady prints the former version, which is incomplete, at the end, and leaves out a number of passages found in Lecan. Mr. Stokes supplies all deficiences in the Leinster text from that in Lecan, bracketing the passages thus dealt with. I select a few of the passages to show what is lost in Mr. O'Grady's version.
In the course of the tale it is told how Aed, son of Ainmire, is defeated and slain in his attempt to levy the tribute. Lecan adds: "but though Aed fell on account of the Boroma he had levied it twice without a battle." Now whether this be addition to the original text by a non-Leinster scribe, or its absence in the Book of Leinster be due to deliberate omission from patriotic motives, there can be no doubt as to the importance of the passage for estimating the historic value of the narrative. When Aed dies his wife laments as follows:
"Dear to me were the three sides
Whereon I shall never look again:
Telltown's little side, Tara's side.
And the side of Aed, son of Ainmire."
Telltown being, so to say, the religious, and Tara the political headquarters of the Irish kings. This exquisite quatrain is only found in Lecan, and is thus absent from Mr. O'Grady's pages, the chief object of which is to bring the beauty of Irish romance home to the English reader!
The next passage is of greater importance and of special interest to folk-lorists as presenting the oldest example of a familiar incident of Gaelic story-telling—the counterspell. It is told how Cummascach, son of the high-king of Ireland,, starts forth on a "free circuit of youth" throughout Ireland. It was the custom of the free circuiter to "sleep one night with the wife of every king of Erin", whence it may be gathered that the "free circuit" was not an institution favourably beheld of the under-kings. Cummascach comes to the court of Leinster's king, Brandub, and, to quote from Mr. Whitley Stokes' version (Rev. Celt., xii, p. 59): "Then said the king of Erin's son, 'Where is Brandub's wife?' A message was sent by him to the queen. The queen came to converse with him, and bade welcome to the king of Erin's son.
["Then the king of Erin's son said to Brandub's wife, 'Let a boon be granted by thee to me.' 'What boon dost thou ask?' says the lady. 'Not hard to say,' quoth he. 'Thou to stay with me that I may sleep with thee.'"]
"'Grant thou a boon to me,' she saith. 'What boon doth thou ask?' says the king of Erin's son.
"'Not hard to say,' she replied. 'A respite, not to detain me until I have finished distributing food to the host, so that I may purchase my honour from them.'"
Of course the queen escapes, and Cummascach is slain by Brandub's men.
The bracketed passage in above extract is omitted by the Leinster scribe, obviously owing to his having skipped the first boon through inattention in copying. As Lecan gives the full passage we have here ample proof that the 15th century MS. is not copied from the 12th century one, but goes back to the common original, a fact in itself of the utmost interest and import. Again, without the omitted phrases the whole passage loses all point and meaning. Yet Mr. O'Grady prints the Leinster text, nonsense though it be, and takes no account of the omitted passage, precious as it is to the folk-lorist and the textual critic.
These examples will suffice, I think, to justify regret that Mr. O'Grady should have given forth an incomplete and mutilated version when better ones lay ready to his hands. Unfortunately, I have to add that Mr. O'Grady does not even translate the whole of the text he prints. A single example will show this. The cause of the levying of the Boroma tribute was this: the king of Leinster's son weds one of the two daughters of the over-king of Ireland. After a while, pretending she was dead, he sought for and obtained the other in marriage. The two sisters meet, and to quote from Mr. Stokes' translation: "But when Fithir beheld Darfine she dies at once of shame. When Darfine beheld her sister's death she dies of grief [Thereafter the washing of the two maidens was performed in Ath Toncha, so that everyone said 'Rough is this washing'. Hence the neighbouring fortress 'Rough Washing' is so called.]"
Mr. O'Grady prints the Irish of the bracketed portion, but does not translate it, nor does he in any way indicate that he has omitted a very curious and important passage. In the first place we have plainly here an interpolation from the Dinnshenchas, that remarkable early mediæval list of Irish topographical legends, a portion of which recently appeared in these pages, which is thus proved to have existed before the composition of the Boroma; in the second place we have an allusion to an incident no trace of which survives otherwise in the story.
It is not necessary to multiply examples of this most regrettable practice, nor would I have mentioned this one were it not that important questions of Irish literary history are concerned. Some of Mr. O'Grady's omissions seem due to a mistaken standard of delicacy. The few naturalistic touches of the original might well have been left entire, considering the cost and bulk of Mr. O'Grady's work.
For the student not the least important section of Silva Gadelica consists of the illustrative extracts, occupying, in English, forty-eight closely printed pages. An immense amount of valuable matter is here brought together and for the first time rendered accessible to the non-Irishspeaking student. But here, even more than in the body of the work, there are grave defects of editing, the effect of which is to seriously diminish the value and utility of this section to the mass of readers. How is the non-specialist to know that MD at the end of an extract means that it is from the Martyrology of Donegal? A number of passages are quoted from the Kilbride MSS. 3 and 16, in the Advocates' Library at Edinburgh, but no information is given as to the date of these MSS.; nor, more important still, is one told from what tracts the passages are taken. Now both of these, like nearly every other early Irish MS., are libraries in themselves, made up of pieces of various date and provenance. To refer simply to the MS. is much as if an English editor should refer to Parl. Deb. or Stat. at Large., without vouchsafing a hint as to the date and nature of the passages referred to. The same remark applies to the citation from the Books of Leinster, Lecan, and Ballymote; but of these MSS. facsimile editions exist, and it is possible by an expenditure of £15 and several hours' work to trace the passages quoted by Mr. O'Grady and to form some idea as to their nature and value. One class of references to the Books of Leinster and Ballymote requires special mention. Mr. O'Grady has—and one cannot be too thankful to him for it—translated a considerable portion of Dinnshenchas, but this is a fact the ordinary reader would never find out, as the references are simply to LL or BB, the quotations being impartially drawn sometimes from the 12th century and sometimes from the 14th century text.[3]
In other respects the student is left in the lurch just where he requires the expert editor's guidance. Thus, p. 522, Mr. O'Grady quotes as follows respecting Ossian: "Blae Dherg from the rushing Banba, the formidable Ossian's mother. In a doe's shape she used to come and join the outlawed band; and thus it was that Ossian was begotten upon Blae Dherg disguised as a doe, LL. 164, marg. sup." It is evident that the value of this passage for the criticism of the Ossianic romance generally depends largely upon its date. The ordinary reader, knowing that LL stands for the 12th century Book of Leinster, naturally concludes that we have here a genuine 12th century testimony to the animal parentage of Ossian. It may well be so; on the other hand it may possibly not be so. For the quotation comes from a marginal note, and what one expects of the editor is that he should give us the benefit of his knowledge as to the date of this gloss. Is it in the same handwriting as the body of the MS.? does it present the same linguistic features as the text to which it is appended? These are questions Mr. O'Grady could answer but does not, and in the meantime the reference is useless, or misleading, to anyone ignorant of Irish palaeography and linguistics.
The criticisms I have felt bound to make could, it will be seen, have been obviated by more definite ideas of the editorial function, and by a very slight extra expenditure of time, work, and space. It is earnestly to be hoped that Mr. O'Grady and his publishers will receive sufficient encouragement to continue the issue of Silva Gadelica, and that the editor will, in future, bear the requirements of the ordinary student more fully in mind than he has done in the present volume.
The contents of the volume (already given in full, Folk-Lore, ii, p. 125) are of too miscellaneous a character to allow of detailed criticism. Suffice to say that whilst the earliest stratum of Irish story-telling is practically unrepresented, the middle and later stages are fully illustrated. These stages are of especial interest to the student of oral literature still surviving among the Celtic-speaking populations of these islands. The wonderful continuity of mode of thought and expression, upon which I have so often insisted, is once more brought into relief. The Gaelic story-tellers of to-day work in a convention which has subsisted for over a thousand years.
Undoubtedly the most important text translated by Mr. O'Grady is the Agallamh na Senorach, or Colloquy with the Ancients, the chief representative of the second stage of the Ossianic romance, and one of the most characteristic specimens extant of Irish story-telling, with its fondness for annalistic and topographical minutiæ, its mingling of dreamy romance and would-be historic accuracy. Renewed acquaintance with this text has not led me to modify the opinions I expressed concerning its nature and date three years ago (MacInnes, p. 411), nor to view with added favour Prof. Zimmer's hypothesis concerning the origin of the Fenian cycle.
Mr. O'Grady has been as chary of exegetical as of critical comment, and this is greatly to be regretted. A romantic literature such as is the Irish, singularly self-contained and cast in a traditional mould equally familiar to reciter and to hearer, offers many pitfalls to the outsider. It is so easy to attach undue importance to an expression or an epithet in a particular passage before one learns that it is merely a conventional cliché. Mr. O'Grady's unrivalled knowledge of Irish romance would enable him, if he but would, to give precious assistance to the student. The few obiter dicta scattered through the volume are pregnant and illuminating. But I must confess my disbelief in the soundness of one, and as the question is of interest to the student of Celtic belief and custom, I will briefly set forth Mr. O'Grady's view and my grounds for taking exception to it.
A number of stories are extant in which the Irish saints play a part that assorts singularly ill with our idea of the saintly character; they show themselves vehement and unscrupulous partisans, they resort to trick and dodge to achieve their ends. But the interesting point is that whilst they approve themselves to be on the same moral level as the pagan Druid, they likewise approve themselves to be on the same intellectual level. There is the same belief in the irresistible power of the formula, in the irrevocable nature of the oath, in the efficacy of symbol and spell. Mr. O'Grady is much chagrined by these stories, and, says he, "it is idle to suppose that the native Irish writers of remote times, whose general tone indubitably is that of gentlemen writing for gentlemen, knew no better than to seriously credit men like S. Columbkill and Adamnan, for instance, with conduct worthy of Til Eulenspiegel" (p. xviii). So he concludes "these episodes have all the appearance of broad caricatures drawn to raise a laugh." That the mediæval Irishman was quite capable of enjoying a laugh at the expense of an eminent saint I am willing to believe, but is it certain that he would have seen anything laughable in the trick by which Moiling procured the remission of the Boroma owing to the double meaning of the word Luath (Monday and Doomsday), or in how Adamnan outwitted the King of Ireland } The two, namely, were fasting and performing penance against each other, and neither got ahead of the other. So Adamnan dressed up one of his clerics in his semblance, and when the king, who was averse to works of supererogation, sent to ask the saint what he was doing that night, the cleric answered, "I banquet and sleep." The king felt he could do likewise. But meanwhile Adamnan kept fast and vigil, and tarried all night in the river, and so got power over the king. The story is a delightful one—to us—but would it have struck the mediæval Irishman as a joke, and would he have considered the trick as ungentlemanly? I doubt it exceedingly, but what I chiefly doubt is that an Irish story-teller would have woven these jokes into historic and hagiological tales which were obviously meant to be taken au sérieux, if not to edify. Yet such is the case with nearly all the tales that exercise Mr. O'Grady. I submit that it is far simpler to treat these stories as evidences of the fact, in itself most probable, that the early Irish saints were just tribal medicine-men with a Christian instead of a pagan bag of tricks, and to regard them as surviving by force of tradition, than to imagine that several generations of Irish story-tellers, after centuries of Christianity, went out of their way to vilify their national saints by harking back to archaic and pre-Christian modes of thought and act.
What makes it still more unlikely that these stories, in which no trace of humorous intent is perceptible, were meant by way of caricature, is the existence of a mediaeval Irish tale conceived in the truest and broadest vein of caricature. I allude to the Vision of Mac Conglinne. The parodist spares neither heroic saga, nor saint's legend, nor even the gospel narrative, and his work, precious as testifying to the existence in serious literature of the incidents and modes of expression which he caricatures, is still more precious as affording conclusive proof that the mediæval Irishman's appreciation and expression of grotesque humour were essentially the same as our own.
Hitherto we have been considering collections of new material, and have had little to discuss in way of contributions to a constructive criticism of the mythic literature of the Irish. But Professor Zimmer, in the important work on Nennius[4] which he has just published, amongst many valuable hints towards the proper understanding of the Irish literary records in the pre-mediæval and mediæval periods, makes two suggestions the effect of which upon current views of Irish myth and saga cannot easily be overestimated. There is a well-known legend to the effect that the bards of the early seventh century were unable to recall in its entirety the greatest of Irish epic tales, the Tain bo Cuailgne; so they sent to Brittany "to learn the Tain, which that wise man (insui) had taken to the East in exchange for the Cuilmenn." This story has generally been interpreted in the sense which critics attach to the finding of the Law under Josiah, i.e., as implying that the Tain assumed its definite shape in the early seventh century. But Prof. Zimmer seems inclined to take it au pied de la lettre. For him insui, "that wise man", can only apply to Gildas, with his standing epithet of Sapiens, who did come from Brittany (returning thither to die) to Ireland in the middle of the sixth century, and who, he conjectures, carried off a MS. of the Tain in exchange for the Cuilmenn, an historical work dealing with the early history of mankind in supplement of the biblical account, which was held in high esteem in mediæval Ireland.
If this is really so, our MS. tradition for the Tain, and inferentially for other portions of the Ultonian cycle, is thrown back to the early sixth century, and we have the proof that, probably following the firm establishment of Christianity in Ireland, the old heroic literature suffered an eclipse during the sixth century and experienced a revival in the seventh century, thanks to King Guaire of Connaught and to the chief bard Senchan Torpeist. The prominence of both these personages in the romantic history of the period is clear evidence that they did take part in a bardic movement of some sort, and perhaps the hypothesis that they represented a national and semi-pagan reaction against Christian culture best fits in with all the facts ot the case.
The possibilities of the other suggestion are even more pregnant. Prof. Zimmer has always insisted upon the Viking period (800-950 A.D.) as forming a chasm in the social and intellectual development of Ireland. The intense and vigorous culture of the sixth-eighth centuries was wrecked and shattered, and the renaissance of the late tenth and eleventh centuries is a building anew the ancient fabric with the scattered fragments remaining, and also with much that had worked itself into the national consciousness during the years of storm and stress. It is, as a rule, the renaissance post-Viking recension of the monuments of early Irish culture that has been preserved to us, amongst others of the Lebor Gabala, the legendary pre-history of Ireland. But with the aid of Nennius, who at the end of the eighth century had access to an older form of the L. G. than any which has comedown to us, we can form an idea of the pre-Viking recension of this text. The section concerning the Tuatha de Danann was, so Prof. Zimmer asserts, much less detailed. The ordinary, postViking, recension describes them as addicted to "druidism, heathendom, and devil's lore, skilled in every art, wrapped in cloud caps and dark mists." Here we have the trace of stories concerning the spell-crafty Norsemen and the invisible-capped Siegfried. So at least it seems to Prof Zimmer.
The suggestion is thrown out casually, and is not followed up. But it is easy to see to what far-reaching consequences it might lead. The Tuatha de Danann represent what at first sight seems to be the only genuine mythological portion of Irish romance; the beliefs concerning them have practically survived to the present day as the fairy mythology of the Gaelic-speaking peasant. It would indeed be a triumph for the "revelationist" could it be proved that the vast structure of romance connected with the Tuath Dea had its basis in tenth-eleventh century amplifications of monkish imaginings drawn from biblical and classic fable with matter derived from the heathen Norsemen invaders. There would not be wanting peculiarities in the tradition of these stories to lend countenance to such a view. The fact which I instanced in discussing Mr. Coffey's monograph on the New Grange monuments, namely, that the historic connection with the kings of the early centuries of our era had faded from the popular memory, whilst the, according to the usual view, far older connection with the Tuatha De retained its full vitality, this fact would be explained at once; the alleged earlier set of traditions would be, as a matter of fact, hundreds of years younger than the other. Nor need we be puzzled, as we must be now, by the curious way in which considerable masses of the so-called mythological cycle stand aloof in literary tradition from any sort of connection with the oldest heroic cycle, that of Conchobor and Cuchulainn. Smaller difficulties,such as the curious parallelism between a passage in the Second Battle of Moytura and one in the Voluspa, to which I called attention in these pages (ante, iii, p. 391), would also disappear.
I may say at once that I do not think a theory, such as I have sketched, likely to be true. I believe it will be found that the Irish mythological cycle is made up of old and genuine Gaelic elements. None the less do I think that a searching examination, starting from the hypothesis of a late and largely foreign origin of this most Interesting and problematic portion of Gaelic legend, would throw much light upon it.
A passage in Prof. Zimmer's book is instructive, if the facts and inferences contained in it be admitted, as to the possibility of apparently genuine and archaic tradition being originated by late and erroneous views of history. In the Red Book Triads, in a poem of the Book of Taliessin, and in other mediæval Welsh texts, we find mention of Beli mawr ab Mynogan, obviously the Bellinus filius Minocanni of Nennius. Nennius obtained this personage from Orosius, who mentions a Minocenobellinus, which the Welsh scribe misread as Minocanni bellinus (i.e., Bellinus son of Minocannus). But the mention of Orosius rests upon a mistranscribed and misunderstood passage of Suetonius (Caligula 44) relating to "Adminio Cynobellini filio". Thus the carelessness of copyists and the ignorance of compilers have combined to invent a British worthy who might, had the literary conditions been favourable, have become the centre of a great romantic cycle.
Beli the Great takes us from Ireland to Britain. Prof. Zimmer's work is chiefly valuable to the student of Welsh history and literary history; its importance for the student of romance lies in the insistence on the early and longcontinued relations between Gael and Cymry, relations which have suddenly been carried backwards in point of time and eastwards in point of territory by the unexpected discovery of an Ogham inscription at Silchester.[5] What Prof Zimmer says about the historic Arthur is sound, but neither novel nor concerned with the serious difficulties of the orthodox view.[6]
In the preceding Reports I sketched Prof. Zimmer's theory of the specific Breton origin of the Arthurian romance as we find it in the French romances of the twelfth and thirteenth centuries. That theory was complicated—and compromised—by connection with Prof Förster's attack upon M. Gaston Paris for arguing that the North-French romance writers received their material from Anglo-Norman intermediaries. Not a trace of these hypothetical intermediaries survives, urged Prof Forster; the French poets got their material from Brittany, urged Prof Zimmer. M. Loth, in the Revue Celtique for October 1892, has to my mind conclusively disproved the Forsterian side of the argument. His reasons can be appreciated by those who are unfamiliar with the minutiæ of historical phonology. He urges that the name Yvains in the French romances can only go back to a written Welsh Ywein. If the name had come to the French orally they would have attempted to render the sound of the Cymric y, which is something between a French e muet and a short o (Ywein = the modern Welsh Owen), and would have written something like Ewen; their retention of the y (which they undoubtedly sounded like a long e) conclusively shows that they only knew the name by sight, and not by ear. Again, the French romance writers, finding a written Caradoc Breich-Bras (i.e., in Welsh, C. of the strong arms), and misled by similarity of look between Welsh Bras = strong, and French Bras = arm, transcribed it as C. Brie-Bras (or, in French, C. short arms), which they never could have done had they heard the word pronouneed, for in accordance with the rules of Cymric phonology the initial consonants suffered change, so that the epithet was sounded Vreichvras. The demonstration seems conclusive as against Professor Forster, for it is obvious that the French romance writers had no access to Welsh MSS., and could only have derived the Welsh forms from Anglo-Norman sources; but Prof. Zimmer might retort that these Welsh written versions came into existence after the Norman Conquest had brought the Breton romances to the knowledge of the Welsh, but before the French romance writers knew of them. M. Loth, however, whilst cordially recognising, as every true student must do, that Prof Zimmer has successfully vindicated for Brittany many features of the Arthurian romance as we possess it, has little difficulty in showing that he has, more suo, driven his theory too hard, and altogether underrated the Welsh element in the romance. For the moment at least the centre of gravity of Arthurian study has been shifted back from Brittany to Britain. But little has been done towards that adequate solution of the Arthurian problem which must, I think, take into account the following factors: (a) the relation of the legendary account, preserved by the Welsh sources alone, to that found in the French romances; (b) the relation of both accounts to the substratum of fact connected with the historical Arthur; (c) the nature, whether in its origin racial and mythological, or borrowed and literary, of this legendary portion; {d) the relation of Cymric and Gaelic legend generally. Professor Rhys has made many acute suggestions under head {a) M. Gaston Paris, under head (c), has, in his study of the Lancelot story, made the most valuable existing contribution towards the explanation of the Arthurian romance; under head (d) there are scattered suggestions due to Prof. Zimmer, M. Loth, and myself, and I may claim to have clearly seen from the outset the importance of the factor. But much remains to be done, and no more fascinating field of study could be chosen.
I may here note a pamphlet on the Grail story, which I have unfortunately mislaid, sent to me from America by, I think, a Mr. Maclean. In addition to some spirited renderings from Wolfram von Eschenbach's Parzival it contained one ingenious suggestion—a comparison of Peredur's adventure with the Addanc of the lake, as told in the Welsh story of Peredur, son of Evrawc, and Sigurd's adventures with Fafnir and Sigrdrifa, as told in the Volsunga saga.
The foregoing Report has been largely concerned with critical questions, but I have, I trust, succeeded in bringing out the importance of what may appear at first blush to be mere dry-as-dust exhibitions of pedantry. It is only by the most exact and searching examination, conducted with all the appliances of the philologist, the palæographer, the historian, and the archaeologist, of all the remains written, figured, and oral of Celtic romance, that we can hope to trace its development and to set forth its true nature. The truth at which we thus arrive, by means which maybe deemed pedantic and wearisome, is far more beautiful than those lazy imaginings we spin out of our own consciousness. And meanwhile we have the spring of as fair and clear a stream of romance as ever welled forth from the imagination of man to cheer and refresh us in our march through the Sahara of criticism. Merely as a story-book Mr. O'Grady's Silva Gadelica is excellent reading, and if one takes it up side by side with the exquisite little volume devoted to the love-songs of Connaught which Dr. Hyde has just brought out, the oneness of the Celtic genius throughout the reach of centuries, as well as its unique and penetrating charm, are borne in upon the mind with irresistible force. The Celtic folk-muse greets us from Dr. Hyde's pages like one of her own heroines:
"The taste of her kisses is sweeter than the honey of the bees on the table,
And to be drinking it in berry-red brandy."
Whoso has tasted those kisses, whoso has heard her fairy-song, like Connla Ruad, will not stay afar from her, but if he may, will follow and dwell with her in the land where she is queen.
- ↑ A full and sympathetic account of Hector Maclean appeared in the Celtic Monthly for March 1893. To this I would refer the reader who wishes to know more of a singularly fine and brave character.
- ↑ This applies to the prose. Very considerable differences exist in the renderings of the verse.
- ↑ These extracts can as a rule be identified by their beginning with "Whence"; e.g. p. 512 (No. vii), "Whence Loch Con," etc.
- ↑ Fully summarised by me in The Academy, Aug. 12th and 19th, 1893.
- ↑ See Prof. Rhys in The Academy for Aug. 19.
- ↑ Difficulties which would be singularly lessened (though still graver ones would make their appearance) if Mr. Anscombe's startling ascription of Gildas' Epistola to the year 498 be correct. (A. Anscombe, Chron. Tracts, No. ii: St. Gildas of Ruys and the Irish Regal Chronology of the 6th Century.)