Whilst engaged on the Translations contained in these volumes, and on the
Notes appended to the various Tales, I have found myself led unavoidably
into a much more extensive course of reading than I had originally
contemplated, and one which in great measure bears directly upon the
earlier Mediæval Romance.
Before commencing these labours, I was aware, generally, that there
existed a connexion between the Welsh Mabinogion and the Romance of the
Continent; but as I advanced, I became better acquainted with the
closeness and extent of that connexion, its history, and the proofs by
which it is supported.
At the same time, indeed, I became aware, and still strongly feel, that
it is one thing to collect facts, and quite another to classify and draw
from them their legitimate conclusions; and though I am loth that what
has been collected with some pains, should be entirely thrown away, it is
unwillingly, and with diffidence, that I trespass beyond the acknowledged
province of a translator.
In the twelfth and thirteenth centuries there arose into general
notoriety in Europe, a body of “Romance,” which in various forms retained
its popularity till the Reformation. In it the plot, the incidents, the
characters, were almost wholly those of Chivalry, that bond which united
the warriors of France, Spain, and Italy, with those of pure Teutonic
descent, and embraced more or less firmly all the nations of Europe,
excepting only the Slavonic races, not yet risen to power, and the Celts,
who had fallen from it. It is not difficult to account for this latter
omission. The Celts, driven from the plains into the mountains and
islands, preserved their liberty, and hated their oppressors with fierce,
and not causeless, hatred. A proud and free people, isolated both in
country and language, were not likely to adopt customs which implied
brotherhood with their foes.
Such being the case, it is remarkable that when the chief romances are
examined, the name of many of the heroes and their scenes of action are
found to be Celtic, and those of persons and places famous in the
traditions of Wales and Brittany. Of this the romances of Ywaine and
Gawaine, Sir Perceval de Galles, Eric and Enide, Mort d’Arthur, Sir
Lancelot, Sir Tristan, the Graal, &c., may be cited as examples. In some
cases a tendency to triads, and other matters of internal evidence, point
in the same direction.
It may seem difficult to account for this. Although the ancient dominion
of the Celts over Europe is not without enduring evidence in the names of
the mountains and streams, the great features of a country, yet the loss
of their prior language by the great mass of the Celtic nations in
Southern Europe (if indeed their successors in territory be at all of
their blood), prevents us from clearly seeing, and makes us wonder, how
stories, originally embodied in the Celtic dialects of Great Britain and
France, could so influence the literature of nations to whom the Celtic
languages were utterly unknown. Whence then came these internal marks,
and these proper names of persons and places, the features of a story
usually of earliest date and least likely to change?
These romances were found in England, France, Germany, Norway, Sweden,
and even Iceland, as early as the beginning of the thirteenth and end of
the twelfth century. The Germans, who propagated them through the
nations of the North, derived them certainly from France. Robert Wace
published his Anglo-Norman Romance of the Brut d’Angleterre about 1155.
Sir Tristan was written in French prose in 1170; and The Chevalier au
Lion, Chevalier de l’Epée, and Sir Lancelot du Lac, in metrical French,
by Chrestien de Troyes, before 1200.
From these facts it is to be argued that the further back these romances
are traced, the more clearly does it appear that they spread over the
Continent from the North-west of France. The older versions, it may be
remarked, are far more simple than the later corruptions. In them there
is less allusion to the habits and usages of Chivalry, and the Welsh
names and elements stand out in stronger relief. It is a great step to
be able to trace the stocks of these romances back to Wace, or to his
country and age. For Wace’s work was not original. He himself, a native
of Jersey, appears to have derived much of it from the “Historia
Britonum” of Gruffydd ab Arthur, commonly known as “Geoffrey of
Monmouth,” born 1128, who himself professes to have translated from a
British original. It is, however, very possible that Wace may have had
access, like Geoffrey, to independent sources of information.
To the claims set up on behalf of Wace and Geoffrey, to be regarded as
the channels by which the Cymric tales passed into the Continental
Romance, may be added those of a third almost contemporary author.
Layamon, a Saxon priest, dwelling, about 1200, upon the banks of the
upper Severn, acknowledges for the source of his British history, the
_English_ Bede, the _Latin_ Albin, and the _French_ Wace. The last-named
however is by very much his chief, and, for Welsh matters, his only
avowed authority. His book, nevertheless, contains a number of names and
stories relating to Wales, of which no traces appear in Wace, or indeed
in Geoffrey, but which he was certainly in a very favourable position to
obtain for himself. Layamon, therefore, not only confirms Geoffrey in
some points, but it is clear, that, professing to follow Wace, he had
independent access to the great body of Welsh literature then current.
Sir F. Madden has put this matter very clearly, in his recent edition of
Layamon. The Abbé de la Rue, also, was of opinion that Gaimar, an
Anglo-Norman, in the reign of Stephen, usually regarded as a translator
of Geoffrey of Monmouth, had access to a Welsh independent authority.
In addition to these, is to be mentioned the English version of Sir
Tristrem, which Sir Walter Scott considered to be derived from a distinct
Celtic source, and not, like the later Amadis, Palmerin, and Lord
Berners’s Canon of Romance, imported into English literature by
translation from the French. For the Auntours of Arthur, recently
published by the Camden Society, their Editor, Mr. Robson, seems to hint
at a similar claim.
Here then are various known channels, by which portions of Welsh and
Armoric fiction crossed the Celtic border, and gave rise to the more
ornate, and widely-spread romance of the Age of Chivalry. It is not
improbable that there may have existed many others. It appears then that
a large portion of the stocks of Mediæval Romance proceeded from Wales.
We have next to see in what condition they are still found in that
country.
That Wales possessed an ancient literature, containing various lyric
compositions, and certain triads, in which are arranged historical facts
or moral aphorisms, has been shown by Sharon Turner, who has established
the high antiquity of many of these compositions.
The more strictly Romantic Literature of Wales has been less fortunate,
though not less deserving of critical attention. Small portions only of
it have hitherto appeared in print, the remainder being still hidden in
the obscurity of ancient Manuscripts: of these the chief is supposed to
be the Red Book of Hergest, now in the Library of Jesus College, Oxford,
and of the fourteenth century. This contains, besides poems, the prose
romances known as Mabinogion. The Black Book of Caermarthen, preserved
at Hengwrt, and considered not to be of later date than the twelfth
century, is said to contain poems only. {1}
The Mabinogion, however, though thus early recorded in the Welsh tongue,
are in their existing form by no means wholly Welsh. They are of two
tolerably distinct classes. Of these, the older contains few allusions
to Norman customs, manners, arts, arms, and luxuries. The other, and
less ancient, are full of such allusions, and of ecclesiastical terms.
Both classes, no doubt, are equally of Welsh root, but the former are not
more overlaid or corrupted, than might have been expected, from the
communication that so early took place between the Normans and the Welsh;
whereas the latter probably migrated from Wales, and were brought back
and re-translated after an absence of centuries, with a load of Norman
additions. Kilhwch and Olwen, and the dream of Rhonabwy, may be cited as
examples of the older and purer class; the Lady of the Fountain, Peredur,
and Geraint ab Erbin, of the later, or decorated.
Besides these, indeed, there are a few tales, as Amlyn and Amic, Sir
Bevis of Hamtoun, the Seven Wise Masters, and the story of Charlemagne,
so obviously of foreign extraction, and of late introduction into Wales,
not presenting even a Welsh name, or allusion, and of such very slender
intrinsic merit, that although comprised in the Llyvr Coch, they have not
a shadow of claim to form part of the Canon of Welsh Romance. Therefore,
although I have translated and examined them, I have given them no place
in these volumes.
There is one argument in favour of the high antiquity in Wales of many of
the Mabinogion, which deserves to be mentioned here. This argument is
founded on the topography of the country. It is found that Saxon names
of places are very frequently definitions of the nature of the locality
to which they are attached, as Clifton, Deepden, Bridge-ford, Thorpe,
Ham, Wick, and the like; whereas those of Wales are more frequently
commemorative of some event, real or supposed, said to have happened on
or near the spot, or bearing allusion to some person renowned in the
story of the country or district. Such are “Llyn y Morwynion,” the Lake
of the Maidens; “Rhyd y Bedd,” the Ford of the Grave; “Bryn Cyfergyr,”
the Hill of Assault; and so on. But as these names could not have
preceded the events to which they refer, the events themselves must be
not unfrequently as old as the early settlement in the country. And as
some of these events and fictions are the subjects of, and are explained
by, existing Welsh legends, it follows that the legends must be, in some
shape or other, of very remote antiquity. It will be observed that this
argument supports _remote_ antiquity only for such legends as are
connected with the greater topographical features, as mountains, lakes,
rivers, seas, which must have been named at an early period in the
inhabitation of the country by man. But there exist, also, legends
connected with the lesser features, as pools, hills, detached rocks,
caves, fords, and the like, places not necessarily named by the earlier
settlers, but the names of which are, nevertheless, probably very old,
since the words of which they are composed are in many cases not retained
in the colloquial tongue, in which they must once have been included, and
are in some instances lost from the language altogether, so much so as to
be only partially explicable even by scholars. The argument applies
likewise, in their degree, to camps, barrows, and other artificial
earth-works.
Conclusions thus drawn, when established, rest upon a very firm basis.
They depend upon the number and appositeness of the facts, and it would
be very interesting to pursue this branch of evidence in detail. In
following up this idea, the names to be sought for might thus be
classed:—