WHO WAS MERLIN or, rather, what was Merlin?
This question has intrigued and vexed countless students of the Arthurian tradition for centuries. Was he someone who panicked and ran away from the Battle of Arfderydd? Who lost his sanity in the battle and lived like a wild beast in the woods? Had he really been a great bard of the chieftain Gwenddolau? If he were a madman, by what mechanism did his insane utterances become recognized as prophecies? Why was he also called Llallogan or Llallawg? Why was he dealt a triple sacrificial death akin to that meted out to the god Lugh (Welsh Lleu)?
These questions are important in and of themselves, of course. But for our purposes they take on a more profound significance. For by answering them to the best of our abilities in an objective way, can we say definitively that originally Merlin had belonged to a class of druidic priests? Or that he had performed some vital function for such a priesthood?
In Geoffrey of Monmouth's History of the Kings of Britain, Merlin (= the Welsh Myrddin) is associated with Amesburys Stonehenge on Salisbury Plain and with Mount Killaraus (= Killare next to the Hill of Uisneach, the center of Ireland), while in Geoffrey's Life of Merlin the great sage is placed atop a mountain in the Scottish Caledonian Wood.
Fragments of the Life of St. Kentigern tell of a madman/prophet named Lailoken, who is explicitly identified with Merlin, and who is found on a "rock" at Mellodonor (modern Molindinar Burn) within sight of Glasgow and at Drumelzier (modern Dunmeller) in Scottish Borders. Lailoken is said to have been buried near Drumelzier.
Before Geoffrey introduced Merlin into the Arthurian saga by substituting him for Ambrosius of Dinas Emrys (a hill-fort in Gwynedd, Wales) and Wallop, Hampshire, the madman/prophet had divided his time between Carwinelow (the fort of his lord Gwenddolau, near Longtown in Liddesdale, known now as the Moat of Liddel), nearby Arthuret (scene of the Battle of Arfderydd, in which his lord was slain and he went mad), the Lowland Caledonian Wood with its mountain and the court of King Rhydderch Hen/Hael. Rhydderch belongs at Dumbarton in Strathclyde, although Geoffrey makes him a Cumbrian king.
In Geoffrey the Caledonian mountain remains unnamed. This is unfortunate, in that by finding this mountain we might learn a great deal more about Merlins identity. And, incidentally, we would have a much firmer fix on the location of Arthurs seventh battle, which occurred in the Caledonian Wood.
Merlins Caledonian Wood mountain is mentioned in one other source: the 13th century French verse romance by Guillaume Le Clerc entitled Fergus of Galloway. The Fergus romance is distinguished by the authors knowledge of Scottish geography. To quote from Cedric E. Pickford in Arthurian Literature in the Middle Ages:
His [Guillaume's] Scottish geography is remarkably accurate... In the whole range of Arthurian romance there is no instance of a more detailed, more realistic geographical setting.
The modern translator of Fergus, the late D.D.R. Owen, has made similar remarks on this romance. The notes and synopses in his translation also remind the reader that various elements of the Fergus mountain episode were adapted from Chretien's Yvain and Perceval and the Continuations of the latter.
But it remains true that only Fergus actually names Merlin's mountain and purports to give us directions on how to get there. The hero Fergus starts his journey to the mountainn not as Nikolai Tolstoy (in his The Quest for Merlin) claims at the Moat of Liddel, where Merlin fought and fled in madness, but at Liddel Castle at Castleton in Liddesdale. Tolstoy uses 1) Guillaume's directions and the placement of King Rhydderch at Dumbarton 2) Merlin's affinity with the stag in Geoffrey's Life of Merlin 3) the incorrect identification of Merlin's Galabes spring/s (these are at Wallop in Hampshire, not in the Scottish Lowlands) and 4) the great height of the Black Mountain to select Hart Fell at the head of Annandale as Merlin's mountain.
There are marked problems with each of these "guidelines" used by Tolstoy. Firstly, the directions given are incredibly vague and hence can be used to chart a course from the Moat of Liddel to just about anywhere:
[Fergus] comes riding along the edge of a mighty forest... Fergus comes onto a very wide plain between two hills. On he rode past hillocks and valleys until he saw a mountain appear that reached up to the clouds and supported the entire sky...
Secondly, Fergus's mountain is given two names, neither of which match that of Hart Fell: Noquetran (variants Nouquetran, Noquetrant) and Black Mountain. The latter is obviously a poetic designation only, the primary name being Noquetran.
And thirdly, there is no edifice of any kind atop or on the flanks of Hart Fell which could have been referred to as Merlins Chapel. As described, this edifice must be an ancient chambered cairn. Such monuments are often associated with Arthurian characters.
The hill-name Noquetran looks to me like a Norman French attempt at a Gaelic hill-name, with the first component being cnoc, English knock, hill. As the French render English bank as banque and check as cheque, Cnoc/Knock became Noque-.
The secret to correctly interpreting the tran component lies in a closer examination of Professor Owens notes on the Fergus romance. For lines 773-93 he writes:
This adventure [of the Noquetran] is largely developed from elements in C.II [the Second Continuation of Chretiens Perceval]. There Perceval fights and defeats a Black Knight in mysterious circumstances. Earlier, he had found a fine horn hanging by a sash from a castle door. On it he gave three great blasts, whereupon he was challenged by a knight, the horns owner, whose shield was emblazoned with a white lion. Perceval vanquished this Chevalier du Cor and sent him to surrender to Arthur. At his castle he learned of a high mountain, the Mont Dolorous, on whose summit was a marvellous pillar fashioned long ago by Merlin.
lines 4460 ff, Owen writes:
In the Fergus romance, the Noquetran episode comes first. The horn hangs from a white lion (cf. the lion on the knights shield in the Perceval Continuation) in the Noquetran chapel, where Merlin had spent many a year. In front of the chapel is a bronze giant, apparently a statue, whose arms are broken off by Fergus, causing the giants great bronze hammer to fall to the ground. Later in the romance, Fergus goes to the Dolorous Mountain or the Eildons and encounters there a club-wielding giant in the Castle of the Dark Rock (reminiscent of the Black Mountain name applied to the Noquetran).
As it happens, the Eildons are noteworthy for having three major ancient monuments atop two of their three hills. On the Eildon North Hill is the largest hill fort in Scotland, the probable oppidum of the Selgovae tribe. Here also is a Roman signal station. On Eildon Mid Hill is a large Bronze Age cairn. The CANMORE database (of the Royal Commission on the Ancient and Historical Monuments of Scotland) has this on the cairn:
This cairn is situated on the SW flank of Eildon Mid Hill about 100ft below the summit, at a height of some 1285ft OD. It has been much robbed and now appears as a low, irregular mo und of stones, about 50ft in diameter, from which a few boulders protrude to indicate the possible former presence of a cist.
More remarkable was the presence below the cairn of the following (also from CANMORE):
A group of seven bronze socketed axes, found on the lower western slopes of Eildon Mid Hill in 1982, is now in the Royal Museum of Scotland (RMS). More can be found on the axes in the Proceedings of the Society of Antiquaries of Scotland, Volume 115 (1985), pp.151-158. An abstract of this article follows:
A group of bronze socketed axes
from Eildon Mid Hill, near Melrose, Roxburghshire
In 1982, a group of seven socketed axes was found on the lower western slopes of Eildon Mid Hill, Ettrick and Lauderdale District, Borders Region. Although recovered from redeposited soil, the axes probably represent a hoard of the Ewart Park phase of the late Bronze Age. The find reinforces what appears to be a significant local concentration of contemporary metalwork around the Eildon Hills.
CIRCUMSTANCES OF DISCOVERY
On 9 August 1982, several bronze socketed axes were discovered by Messrs W and A Wilson (uncle and nephew) on the margin of the rifle range on the lower western slope of Eildon Mid Hill, the central and highest of the three peaks which form one of the most conspicuous landmarks of the Scottish Border country (illus 1 -2). They were intending to use Mr William Wilson's metal detector slightly further uphill in order to search for shell-cases and cartridges in the area of the targets. While they were walking up the roughly trodden grassy path on the northern margin of the range, with their detector switched on but not consciously in use, their conversation was interrupted by a signal from the machine. The tone indicated a non-ferrous metal, and closer scanning with the detector suggested that there were at least four soil anomalies within a small area. The Wilsons decided to investigate the source of these signals: clearance with a trowel of a small patch of the tenacious turf revealed, to their surprise, not spent ammunition but a socketed axe of bronze (catalogue no 7); a second discrete signal was found to emanate from a further axe (7) also lying on its own just under the turf. The source of a third signal was revealed to be a cluster of three axes lying close together (3-5). Realizing the growing archaeological significance of their find, and not wishing to disturb the site further, the Wilsons responsibly left undisturbed in the ground the sources of what by then appeared to be two further signals of a similar nature. They replaced the disturbed turf as best they could to mask the site, and immediately reported their discovery to the Ancient Monuments Division of the Scottish Development Department, one of whose Inspectors, Dr N Fojut, in turn notified the National Museum. On 17 August, one of the writers (TGC) visited the finders to inspect the axes already recovered and to view the site of their discovery, and the following day, he and Mr lan Scott of the NMAS, with the assistance of Aidan Wilson, investigated the immediate area of the find (NGR NT 542325).
A small trench, 2m by 2m, was set out centred on the find-spot of the cluster of three axes already removed (cf illus 1, d). Following removal of the coarse turf, the points from which the five axes had been retrieved became clear: these showed up as irregular depressions in what appeared to be the natural subsoil of reddish clayey loam with plentiful stones (mostly the local felsite). Scanning of the trench with the detector relocated the positions of the two signals not investigated by the Wilsons the previous day. One of these emanated from a slightly darker patch of humic soil: on removal, this proved simply to be a deeper pocket of topsoil, occupying a slightly damper, clayier depression in what seemed once again to be the natural subsoil. A further socketed axe (6) lay at an angle on the side of the depression 10-15 cm below the present ground surf ace: perhaps on account of the damper matrix, the axe was in a noticeably more corroded condition than the others when found.
Surprisingly, the source of the final signal appeared to emanate from natural subsoil with no obvious trace of any feature on its surface. Removal of a small area of this supposedly natural 'subsoil' revealed the source of the remaining signal to be a further axe (2) and threw some light on the context of the group of axes as a whole. It became clear that this last axe was not, as first thought, lying in undisturbed ground, but rather was lying in compacted redeposited soil apparently occupying the side of a natural gully or channel in the hillside. In the time available, it was not possible to excavate the presumed channel nor determine its width or depth, but the circumstances which led to the incorporation of redeposited soil in natural features seem clear enough for the construction of the rifle range must have involved considerable smoothing out of the contours and irregularities of the hill-slopes. Churning up of the ground, the infill of erosional features such as gullies caused by water run-off, and the compaction of the area by machines could account for the formation of the deposit on and in which all seven axes lay. Indeed, gullies of the type envisaged can be seen elsewhere on the slopes of the Eildons (see illus 2). The axes are likely to have been moved bodily in a load of earth and soil, dumped and then slightly dispersed (axes 2 and 6 were separated by a distance of 1-5 m). Following the removal of the final two axes, the excavated soil and turf were replaced and the site restored to its original appearance as far as possible. Finally the surrounding area was scanned with the metal detector, but no further non-ferrous anomalies were noted
THE ORIGINAL DEPOSITION OF THE AXES
In view of their discovery in redeposited soil we cannot be absolutely certain how the axes were originally deposited. However, their number, their proximity and their similar condition all suggest that they came from a hoard, probably close to their eventual find-spot. Whether the seven axes recovered in August 1982 comprised the whole hoard remains uncertain. On the other hand, it is possible, though less likely, that more than one separate deposit was originally involved
I would see in Noquetran or Noquetrant a Gaelic cnoc or Anglicized knock plus one of the following:
G. dreann grief, pain (cf. Irish drean, sorrow, pain, melancholy)
G. treana, treannadh lamentation, wailing
In other words, Noquetran is merely a Gaelic rendering of the Old French Mont Dolorous!
The bronze hammer Fergus causes to be dropped near Merlins Chapel on the Noquetran is a folk memory of a bronze socketed axe being deposited on the slope below the Eildon Mid Hill cairn or, more probably, of such an axe being found on the site prior to Guillaume Le Clercs writing of the Fergus romance. Merlins Noquetran chapel is the Eildon Mid Hill Bronze Age cairn.
Mountain, Black Mountain and Castle of the Dark Rock are
all designations for the Eildons. The hill-name
Eildon is found in 1130 (_Place-Names of Scotland, 3rd
Edition, James Brown Johnston) as Eldunum and in 1150 as
Eldune. While various etymologies have been
proposed, the most commonly favored one is G. aill,
a rock, cliff, plus OE dun, a
hill. The Fergus romances
Castle of the Dark Rock (Li Chastiaus de la
Roce Bise) may stand for the hill-fort on Eildon North
Hill, with Eildon being perceived as composed of aill,
rock, plus not dun, hill, but instead - OE
dun, a colour partaking of brown and black; ME dunne,
donne, dark-coloured: Ir. Dunn, a dun colour: Wel. Dwn,
dun, swarthy, dusky: Gael. Donn, brown-coloured
So why were the Eldons identified with the Dolorous Mountain/Noquetran? The answer may lie in part with Nikolai Tolstoys astute observation that the lion Fergus thinks should be roaming over the mountain-top, but which he finds inside the chapel is an error or substitution for the god Lugos (Welsh Lleu, Irish Lugh). In Welsh, Lleus name could sometimes be spelled Llew, and the latter is the normal spelling for the Welsh word lion. Merlins associations with Lleu will be briefly discussed below. For now, suffice it to say that the Dolorous Mountain undoubtedly got its name because Lugos or Lugh was at some point wrongly linked to Latin lugeo, to mourn, to lament, bewail. Such mistakes in language could easily have occurred when going from Celtic to Old French.
The Dolorous Mountain is thus, properly, Lugos Mountain. And the Lugos/Lugh/Lleu mountain is particular is Eildon Mid Hill, the highest of the Eildons, with its Bronze Age cairn. Such an identification of the Dolorous Mountain has implications for the Dolorous Garde of Lancelot, especially given that Lancelot himself is a late literary manifestation of the god Lugh, something first discussed long ago by the noted Arthurian scholar Roger Sherman Loomis.
We know of five Lugh forts in Britain, four known and one unlocated. Of the former there is Dinas Dinlle in Gwynedd, Loudoun in East Ayreshire, Luguvalium or Carlisle in Cumbri and Lleuddiniawn or Lothian, land of the Fort of Lugh. Luguvalium has been interpreted as containing a personal name *Lugovalos, Lugos-strong, but I believe this name is instead a descriptive of the fort itself as being Strong as Lugh. Dr. Graham Isaac, a Celtic language specialist at The National University of Ireland, Galway, agrees with me that this could well be the case.
Then there is the Lugudunum or Hill-fort of Lugh of the Ravenna Cosmography. This place is (see Rivet and Smiths The Place-Names of Roman Britain) situated somewhere roughly between Chester-le-Street and South Shields. The only good candidate would seem to be Penshaw Hill, which the Brigantes Nation Website calls the only triple rampart Iron Age hill-fort known to exist in the north of England. Penshaw Hill is associated with the famous Lambton Worm, a monster not unlike the two worms or dragons of Lleus hill-fort of Dinas Emrys in Gwynedd, Wales.
The Eildons are noted for the stories of Canobie or Canonbie Dick and Thomas the Rhymer of Ercildoune. Canonbie is near to both the Carwinley of Myrddins/Merlins lord Gwenddolau and Arthuret Knowes, the scene of the Battle of Arfderydd in which Myrddin was driven mad. The 13th century Thomas is credited with meeting an elf-woman under the Eildon Tree (whose location is now marked by a stone) and being taken under the Eildons to the land of Faery. He is also credited with a prophecy concerning Merlins grave at Drumelzier:
Tweed and Powsail meet at Merlins grave,
[According to P.C. Bartrum in A Welsh Classical Dictionary, under his entry for Llallogan, this prophecy was first published by Alexander Pennycuick in 1715.]
The story of Canonbie Dick presents Thomas as a wizard from past days, and I will quote it in full:
long time ago in the Borders Region there lived a Horse
Cowper (trader) called Canobie Dick. He was both admired
and feared for his bold courage and rash temper. One
evening he was riding over Bowden Moor on the West side
of the Eildon Hills. It was very late and the moon was
already high in the night sky.
A similar story is told of Alderley Edge in Cheshire, only in that version the wizard is Merlin and the sleeping knights are King Arthur and his men. My guess is that in the case of the Canonbie Dick story, Thomas the Rhymer has taken the place of Merlin. This is not a new supposition, but combined with my identification of Myrddins Noquetran with Eildon Mid Hill as the Dolorous Mountain, the argument is significantly strengthened. For Fergus was written around 1200 AD, while Thomas is thought to have lived c. 1220-1298. At some point Thomas was substituted for Merlin at his chapel/cairn on Eildon Mid Hill.
If I am right and the Eildons are Merlins Mountain at the center of the great Celyddon Wood, then we can allow for the Celyddon as being thought of as the ancient woodland which covered much of the area surrounding the Eildons. When we combine this with the fact that Merlin was obviously wandering in the wood in the vicinity of Drumelzier when he was captured by Meldred, it is fairly obvious that the Celyddon, which in this context means merely a great forest of the Scottish Lowlands, extended at least to the Lammermuir Hills in the north and perhaps as far as the Cheviots in the east.
As the Roman Dere Street crosses the Teviot near Jedburgh and the Tweed at Newstead, and four of Arthurs other battles were fought on or near Dere Street, it is likely that Arthurs Celyddon Wood battle was fought in part of this ancient woodland on or near Dere Street somewhere between Teviotdale and Lauderdale.
Indeed, there were four great ancient forests surrounding the Eildon Hills: the Jedforest, whose Capon Tree oak is one of the oldest such trees in all of Britain; Teviotdale itself, which was covered by huge oaks and ash trees in the 12th century; the Ettrick Forest of Selkirkshire; the Lauder Forest, an immense forested track encompassing Lauderdale that still existed up until the 17th century. Apples, or rather crab-apples, the very species of tree Merlin takes refuge under in the early Welsh poetry, were also present in this region. The St. Boswells Apple is thought to be 150 years old and is the largest of its kind in Scotland. Thomas the Rhymer, taken to Fairyland at the Eildons, is given an apple by the Queen of Fairy.
We may be able to pinpoint the location of
the Coed Celyddon battle for precisely. Although
Geoffrey of Monmouth has been justly criticized for
producing stories of early British kings rather than
histories, we cannot discount the possibility that at
least occasionally his account of Arthurs reign may
preserve accurate historical traditions. When he
has Arthur face the Saxons in the Caledonian Wood, he
tells us that
This circular palisade Arthur constructs in the Caledonian Wood made me think of the semi-circular Catrail dyke. To quote the description of the Catrail from the Royal Commission on the Ancient and Historical Monuments of Scotland (courtesy Mark Douglas, Principal Officer for Heritage and Design, Planning and Economic Development, Scottish Borders Council):
This is the linear earthwork, known as the Catrail which runs from Roberts Linn to Hosecoteshiel and embraces the entire head of the Teviot basin from the Slitrig to beyond the Borthwick Water. It is not continuous, in the sense that is incorporates streams or woodland, and its engineering, in short sectors clumsily joined and imperfectly aligned, stamps it as a product of somewhat inexperienced communal effort. The date of its construction is doubtful, though it is certain that it fits only Anglian territorial dispositions. Oman notes the opinion that it might mark a political boundary after Aethelfriths victory at Degsastan in 603, and this is indubitably the earliest possible date for such a work.
As the date of the construction of
the Catrail is doubtful, could we not propose
that it was built slightly earlier, say in the 6th
century, the time of Arthur? The Catrail runs
through what was the great
The derivation of the name Catrail has been disputed. The most commonly disseminated etymology is Cymric cad or cat, battle, plus Cymric rheil or rhail, wrongly defined as fence, probably for a presumed wooden palisade that may have once existed atop the dykes bank. Unfortunately, as rheil or rhail does not mean fence. It is rail, track and is not attested in Welsh until the 18th century. According to Professor Richard Coates (Onomastics and Director of the Bristol Centre for Linguistics at UWE, and Hon. Director of the Survey of English Place-Names) rheil/rhailis doubtless a borrowing from English and so Catrail cannot be derived from Cat + rheil/rhail.
A cad/cat + rhigol (the rhill
mentioned below), the latter being Welsh for rut,
groove, (long narrow) channel; trench, furrow, ditch,
gutter, cannot have become Catrail even
through a process of Anglicization. Rhigolis
pronounced something like wriggle.
Again, Professor Coates has assured me that a
hypothetical Cad + rhigolcannot have become Catrail.
This is so despite the attractiveness of Battle-ditch
as a meaning for the earthworks name.
Dr. Andrew Breeze, the Celtic onomastic expert in
Helen Darling, Part-Time Local Studies Librarian, Library Headquarters, St. Marys Mill, Selkirk, was kind enough to send me the following on this feature-name:
KENNEDY, William Norman: Remarks on the ancient barrier called "The Catrail", with plans In Proceedings of the Society of Antiquaries of Scotland 1857-1860 pp117 - 121
Considerable diversity of opinion exists as to the derivation of the term, it being variously stated by different authors. Chalmers calls it "the dividing fence", or "the partition of defence"; Jeffrey, "a war fence or partition - Cat signifying conflict or battle, and Rhail a fence"; others from Cater a camp, and Rhail a fence, a dividing fence among the camps; others, again, from Cud a ditch, and Rhail a fence, the ditch fence or boundary; while another class call it "the Pictswork ditch", attributing the formation of it, and all other ancient artificial remains in the district through which it passes, to the Picts, - a race regarding whom very mythical traditions continue to float about and receive credence; but almost all writers concur in attributing its formation to the Britons, subsequent to the withdrawal of the Romans from this country.
CRAIG-BROWN, T. The
History of Selkirkshire or Chronicles of
Concerning the origin of the word "Catrail" there have been as many different suggestions as about its purpose:
CHALMERS. - Cad, a striving to keep; and Rhail, a division. Together - the dividing fence.
JEFFREY. - Cat, a struggle; and Rhail. Together - a war-fence.
MACKENZIE. - (Quoted by Gordon).
VARIOUS. - Cater, a camp; and Rhail. Cud, a ditch; and Rhail. Cad-rhill - war-trench.
MISS RUSSELL. - Cader (Welsh), a defence; and Rail.
An amateur philologist has ventured to suggest the Saxon words camp-trail as at all events a possible origin, but the probabilities are in favour of a British or Cymricderivation; and we incline to cater-rhail, a camp fence. Cater, Celtic for camp, as in the celebrated Cater-thun, is undoubtedly derived from the Latin, castra. Until the Romans came, there was no need for such a word in the native vocabulary, and it must be kept in mind that the diggers of the Catrail were Romanised Britons.
That there was along the Catrail a terrible struggle for supremacy between the Cymriand the Saxons may be inferred from the frequent occurrence of names beginning with the word "Cat", or "Cad", Cymric for battle or conflict. Nor far from its northern end is Cat-pair, and nearer, Cat-ha(?). Caddonwe have seen derived from Cad afon, battle river. Cat Craig is a hill on Tinnis farm in Yarrow, traversed by the Catrail. A little further up is Catslack, and there are pools near by known as Cat Holes. Finally, near the end of the trench is Catlee. It is necessary, however, to be on one's guard against etymological inferences, and these names are submitted with due reserve.
The noted Scottish place-name expert
Watson said that the Catrail name may be compared
with Powtrail [now Potrail], the name of a head-stream of
After noticing on the map that the Powtrail has many pronounced bends or turns, I would follow Professor Richard Coates (Onomasticsand Director of the Bristol Centre for Linguistics at UWE, and Hon. Director of the Survey of English Place-Names) suggestion of Welsh traill, turn, a turning, as the streams second component, supplying us with a meaning of stream that turns or the turning stream.
There is precedence for connecting traill with a stream or river-name: Chepstows earlier Welsh name was Ystraigl or Ys-traigl, a turn, this being a reference to the bend of the River Wye at the site of the town.
However, such a derivation for Powtrail does not help us with the Catrail name.
Henry Gough-Cooper of The Scottish
Place-Name Society has informed me that Alan James
prefers a different etymology for the Powtrail. In
his paper A Cumbric Diaspora? (in A
Commodity of Good Names, ed. Padel and Parsons,
Shaun Tyas 2008), James mentions the Catrail in a
footnote on Powtrail, which he speculates may be Cumbric
*polter-eil stream of a wattle fence of
woven hedge: for *eil see The Poems of
Taliesin, ed. I. Williams (
According to Gough-Cooper, *Polter is stream-naming word found in northern Cumberland, parts of the Solway basin and south-west Cheviots, e.g. see EPNS Cumberland I, 8, 24, 62; II, 373; plus Polterkened in the Lanercost Cartulary. Alan James suggests perhaps *pol + extension *duvr, the latter being the word for water. One might compare for this last element the river-names Kielder and Calder, both from Welsh caled, hard, and dwfr (OBrit dubro-), water, stream.
Williams, on the pages cited by James, says that:
The problem with the ail of Catrail being the word for fence is what to make of the first component, Catr-.
I would propose that the first component was derived from Gaelic cathair, 'fort'. Even though, as Dr. Richard Coates has made clear to me, the medial th in cathair is silent (cf. Welsh caer, 'fort') and should not have yielded Catr-, in Highland Scotland we do find the Brown and White Caterthun (cathair + dun) forts. If cathair could become cater in these names, certainly it could have become the Catr- of Catrail.
But if the Catrail is the Fort-fence, what fort are we talking about?
John Barber, Elaine Lawes-Martay and Jeremy Milln (in The Linear Earthworks of Southern Scotland; Survey and Classification, Transactions of the Dumfriesshire and Galloway Natural History and Antiquarian Society, Series III, Volume LXXIII, 1999), summarize their study of the Catrail thusly, bringing the dyke into connection with the fort next to Merlins Eildon Mid Hill, the center of the Celyddon/Caledonian Wood:
The natural segmentation
of the Scottish landscape into separate territories may
also account for the relative absence of sites used as
local-administration boundaries. These administrative
units are usually fairly clearly defined by the nature of
the terrain and separated from each other by distinct,
natural features, such as ranges of hills of mountains.
This factor may also account for the paucity of lengthy
earthworks. Of those examined in this study only the
Catrail seems likely to have functioned as a large scale
territorial boundary. In association with other natural
features, it may have functioned in respect of the
hillfort of North Eildon Hill in the same role as the
earthwork called The Dorsey seems to have
functioned in respect of the royal site of Eamhain Macha
In the same Transaction of the DGNHAS the Catrail is described in detail:
THE NAME MYRDDIN
The best derivation for the name Myrddin has been proposed by Dr. G. R. Isaac of The National University of Ireland, Galway:
"The attested name MYRDDIN reflects
an earlier, not directly attested
"The form Mórrígain with long vowel
in the first syllable, so 'Great Queen',
This Myr-ddyn > Myrddin is a much more satisfactory explanation than the previously offered theory that Myrddin was derived from the city name of Carmarthen, ancient Caerfyrddin, Roman Moridunum. This last notion derives from the fanciful HISTORY OF THE KINGS OF BRITAIN by Geoffrey of Monmouth. In this source, Myrddin is found as a boy at Carmarthen. The whole story is Geoffreys alteration of Nenniuss tale of the boy Ambrosius being found at Campus Elleti in Glamorgan.
THE CELYDDON WOOD AS THE LAND OF SPIRITS
The ancient Classical writer Procopius (in his HISTORY OF THE WARS, VIII, XX. 42-48) said:
Now in this island of Britain the men of ancient times built a long wall, cutting off a large part of it; and the climate and the soil and everything else is not alike on the two sides of it. For to the south of the wall there is a salubrious air, changing with the seasons, being moderately warm in summer and cool in winter But on the north side everything is the reverse of this, so that it is actually impossible for a man to survive there even a half-hour, but countless snakes and serpents and every other kind of wild creature occupy this area as their own. And, strangest of all, the inhabitants say that if a man crosses this wall and goes to the other side, he dies straightway They say, then, that the souls of men who die are always conveyed to this place.
From the Welsh poem "The Dialogue of Myrddin and Taliesin" (BLACK BOOK OF CARMARTHEN), we learn that at Myrddins Battle of Arfderydd:
Seven score chieftains became gwyllon
Gwyllon or "Wild Ones" is a word deriving from gwyllt, "wild". The Welsh epithet for Myrddin is, of couse, Gwyllt. Myrddin Gwyllt is Myrddin "the Wild".
But as Tolstoy pointed out, there is something odd about these two lines. The gwyllon or Wild Ones are equated with the warriors who died in the battle! The word died in the poems second line is Middle Welsh daruuanan. Modern Welsh has darfyddaf or darfod, which according to the authoritative Geiriadur Prifysgol Cymru (A Dictionary of the Welsh Language) has the following meanings:
To come to an end, end, conclude, finish, complete, terminate, cease; expire, die, languish, weaken, fail, fade, decline, perish
There is thus no ambiguity in the poetic passage we are considering. The warriors who became Wild Ones did not simply go away or disappear they died. In this context, then, to become gwyllon means to become a roving spirit that has left its battle-slain body behind. To exist as a Wild One is to exist in spirit-form after the death of the body.
The Christian medieval mind either could not accept this notion of wandering spirits or, just as likely, misunderstood it. The gwyllon were transformed into living madmen who leapt or flitted about the forest much as did their Irish counterpart, Suibhne Geilt.
In another Myrddin poem, "Greetings" (BLACK BOOK OF CARMARTHEN), we are told by Myrddin himself:
The hwimleian speaks to me strange
Hwimleian or "Grey Wanderer" is yet another word for a spirit or spectre.
Myrddin "the Wild" was thus the soul of a dead man, wondering the woods of Celyddon. We must presume that it was believed that certain individuals with necromantic skills could communicate with or perhaps even channel such spirits and thus extract from them information only available to the dead, e.g. prophecies.
Difficult as it is for us to believe in this modern age, ancestral spirits were once worshipped or placated with offerings. The 'Dis Manibus' or "For the Divine Spirits of the Dead" was a common heading for Roman period memorial stones. Morddyn the "Elf-Man" may have been such a spirit of the dead.
In the case of Merlin, then, we have a warrior-bard who is slain in battle at Arfderydd. His spirit wandered the Lowland Scottish forest, but had its proper home atop the Lugh Mountain of Eildon Mid Hill. There, in Merlins Chapel, the Bronze Age cairn whose interior symbolized the Otherworld/Underworld and which was also a portal to the Otherworld, the spirits of the dead who had worshipped Lugh in life congregated. Although thanks to Geoffrey of Monmouth and others Myrddin/Merlin came to be conflated with Lugh or perhaps seen as a human avatar of the god, the most judicious interpretation of the early sources does not warrant such an identification of ghostly warrior-bard with divine entity.
We need now address the problem of Myrddin's identification with the madman Lailoken, W. Llallogan or Llallawc. In his edition of Geoffrey of Monmouth's Life of Merlin, Basil Clarke proposes that Myrddin was originally a certain Llallawg or Llallogan who went mad at Arfderydd. Lailoken/Llallogan, suggests Clarke, may derive from a placename similar to the Gaulish village Laliacensis, which was named after the family of Lollius Urbicus, a second century Governor of Britain.
However, the word llallawc/llallogan (linked to W. llall, "other", a reduplicated form of ail/eil) is used in Welsh poetry, where it denotes something like "lord" or "friend" or "dear friend". W.F. Skene, in his The Four Ancient Books of Wales, translates the word as "twin brother". Ifor Williams translates it as "friend" in The Poems of Llywarch Hen. Egerton Phillimore and A.O.H. Jarmon make a case for it being a true personal name.
The crux of the issue is found in the Welsh poem, "The Prophecy of Myrddin and Gwenddydd, His Sister". There Gwenddydd addresses her brother Myrddin as both Llallogan (once) and Llallawc (several times). The question then is: are we to see Llallogan/Llallawc as merely a term of endearment applied to Myrddin or as a separate person with whom Myrddin was identified?
The MS. Cotton Titus A. XIX contains the tales of "Kentigern and Lailoken" and "Meldred and Lailoken". Jocelyn's 12th century Life of St. Kentigern also mentions Lailoken in its last chapter. In both, Lailoken is identified with Merlin.
In the story of Meldred and Lailoken, Merlin suffers a triple death at the hands of Meldred's shepherds. The triple death is a motif found in Celtic literature and was a special kind of death meted out to kings, heroes and gods alike. It is, in essence, a form a human sacrifice, made sacred by the use of three simultaneous methods of killing. In their book The Life and Death of a Druid Prince, Anne Ross and Don Robins hypothesize that each method of killing was sacred to a specific god. The three Celtic divinities in question have been tentatively identified with Taranis the thunder god, Esus the lord and master and Teutates, the overall god of the people. Doubtless there were regional counterparts to these deities.
Nikolai Tolstoy used the example of Lailoken's triple death to more closely link Merlin with the god Lleu/Lugh, as Lleu in Welsh tradition also suffers a triple death.
The true importance of Merlin's triple death on the Tweed near the Dunmeller of Meldred has been overlooked, however. For it clearly demonstrates that by the time of the composition of the Life of St. Kentigern, any memory of the poetic device which employed a state of madness to represent a post-death spectral existence had been lost. No one understood any longer that Merlin or, rather, Myrddin, had died at the Battle of Arfderydd. And because his madness was taken literally, a story was concocted which sought to demonstrate his divine nature by making him the victim of a three-fold human sacrifice.
That such a form of human sacrifice was actually practiced has been proven by examination of the Lindow Man, who was struck in the head by an axe, choked with a garrote and drowned in a pool (see again The Life and Death of a Druid Prince). By undergoing such a sacred killing, the Lindow Man was acting the role of a god like Lleu. And by acting such a role, he became the god a sun god who was seasonally killed and reborn.
The corollary of the three-fold death was, of course, resurrection. The god Lleu, who in death is represented as a Jupiter-eagle in an oak tree, was brought back to life by Gwyddion. The Lindow Man as Lleu or a similar sun god, by dying, helped insure the seasonal rebirth of the sun god. For as the sun died each year, so was it reborn. There cannot be life without death.
Merlin under his nickname Llalogan thus becomes the god in death. But in life, he was a warrior serving his lord Gwenddolau, who also perished at Arfderydd.
The model for Myrddins/Lailokens death at Dunmeller may have been borrowed from another elf figure. In Welsh tradition we find a character called Gyrthmwl Wledig (variants Gwerthmwl, Gwrthmwl, etc.). In Triad 1, he is mentioned as Chief Elder of Penrhyn Rhionydd in the North. Arthur is Chief Prince of the same place, while Cynderyn Garthwys [St. Kentigern] is the Chief Bishop. We have seen above that Lailoken was brought into connection with Kentigern.
In Triad 44, immediately after mention of the Arfderydd battle is made, Gyrthmwl's sons are said to have ridden onto Allt or Rhiw Faelawr or Faelwr in Ceredigion. This is the site of the fort known as Dinas Maelawr (modern Pendinas), the home of the legendary Maelor Gawr or Maelor "the Giant". Dinas Maelawr is also called Castell Maylor and was built upon a high hill or ridge beside the river Ystwyth. Gyrthmwls sons ride up the hill to avenge their father, whom Maelor is said to have killed.
We have seen that the Life of St. Kentigern places Merlin at Dunmeller, modern Drumelzier, on the Tweed. It is here that he is captured by the chieftain of Dunmeller, Meldred (cf. Maldred, the name of the early 11th century Scottish Lord of Carlisle and Allerdale in Cumbria), and it is Meldreds shepherds who kill him.
The chieftain Gyrthmwl in Triad 63 is referred to as "Ellyll Gyrthmwl Wledig", the word ellyll meaning "spirit, phantom, ghost, goblin, elf, fairy". In her note to Triad 63, Rachel Bromwich suggests that the word ellyll in this context may be related to that of the Gwyllt epithet used for Myrddin and the Irish Geilt epithet used for the madman Suibhne.
To quote the relevant Triad in full:
Tri Tharv Ellyll Ynys Brydein:
Three Bull-Spectres of the Island of
Ellyll is cognate with Irish Ailill, a personal name with the same meaning. Rather remarkably, the best etymology for ellyll and Ailill is yet another reduplicated form of ail/eil, other. To quote from Professor Daniel Melia of the University of California, Berkeley (personal communication):
Im citing CELTICA3, 1956,
which reads as follows in its entirety:
Dr. Ranko Matasovic of The Etymological Lexicon of Proto-Celtic (again via private correspondence) says
Welsh ellyll is indeed cognate with Mir. Ailill, but these names cannot be related to English elf, which is frfom Germanic *albiyo-. It is certainly possible that these words contain the stem *al- (actually *h2el-, in a more modern notation), the plain pronominal stem that meant other, different (Lat. alius, Gr. allos, etc.). I would add that Alladhan, the name given to Llallogan in the Irish Suibhne Geilt story, would appear to be from Irish allaid, wild. The most likely etymology for allaid is the same al- root, as an Other is someone who lived beyond the civilized world and was hence barbarous or wild, a stranger or foreigner or an enemy, i.e. someone deemed dangerous because he did not belong to ones native land. The evolution of meaning would be similar to the development from Latin silvaticus, belonging to woods to French sauvage.
Dr. Graham Isaac of The National University of Ireland, Galway, and Professor of Celtic Thomas Charles-Edwards of Jesus College, Oxford, both agree with this derivation for ellyll/Ailill.
In my article Camelot and Other Arthurian Centres (Vortigern Studies Website), I showed how Camelot derives via the Old French from an earlier Romano-British Campus Alletio. Alletio is the name of a god whose name contains the same ail/eil root. Anne Ross has suggested Alletio may, therefore, be the God of the Otherworld. If so, it is rather remarkable that this divine name would appear to be similar is not identical in meaning with both Llallogan/Llallawg and ellyll/Ailill (and with Irish Alladhan; see below).
The first version of the Triad 44 in Peniarth MS. 27 adds that a giant Maelwr was slain in his fort by Gyrthmwl's sons in revenge for the murder of their father. Lailoken, as already mentioned, is killed by the shepherds of Dunmeller. Gyrthmwl, given that Penrhyn Rhionydd has been very plausibly identified with the Rhinns of Galloway (see Bromwich's note to Triad 1), looks to be the 8th century Bishop of Whithorn, Frithuwald.
The discrepancy in date between Frithuwald and Kentigern (d. 603) is probably due to Nennius's statement that the Bernician chieftain "Friodo(l)guald", i.e. Frithuwald, was ruling in the last quarter of the 6th century. This Friodo(l)guald is mentioned along with Deoric son of Ida, Hussa, Urien Rheged, Rhydderch Hen, Gwallawg and Morgan. Interestingly, in one of the Wessex royal pedigrees a Frithuwald is made the father of Woden.
One might hypothesize that the Bernician chieftain was captured by British enemies based at Dunmeller. As often happened with war captives, he was selected because of his high political status to be given us as a triple sacrifice. Probably the sacrifice was intended to bring good fortune in battle with the Saxons. By dying a triple death, Frithuwald became an incarnation of the god Lugh, who appears to have undergone a seasonal death and rebirth.
Now Frithuwald in OE almost certainly means either "Peace-ruler" or "Protection-ruler". But Gyrthmwl Wledic (wledic is cognate with the Germanic wald) may well have been interpreted by the Welsh as "Ffridd"-ruler, i.e. "Forest-ruler". Irish has frith, "a wild, mountainous place, a forest, a deer forest", Welsh has ffridd, "wood, wooded land, a mountain pasture, a sheep walk" and ME has frid, "deer park", from AS fyrhth(e), "wood, woodland".
Triad 63, which refers to Frithuwald/Gyrthmwl as a tharv ellyll or "bull-spectre", also has a variant, wherein the phrase charw ellyll or "stag-spectre" is substituted.
If "stag-spectre" is the more proper form, then we would have yet another reason why Gyrthmwls death story was borrowed for that of Myrddin, as Geoffrey of Monmouth tells us that Merlin took the form of a stag.
To sum up: it seems fairly certain that the story of Gyrthmwl the Ellylls death at Dinas Maalwr on the Ystwyth was borrowed in order to provide Myrddin with a death-story at Dunmeller on the Tweed. We should remain content with the notion that Myrddin is simply Myrddin, the Elf-man. After all, we have plenty of evidence for Aelf- or elf names among the Anglo-Saxons. Aelfric is a good example. The proper name Ailill (see above), the early Irish cognate of Welsh ellyll, elf, was quite common. There would appear to be no reason, therefore, to propose that Myrddin only became the Elf-man after his death at Arfderydd.
According to Dr. Graham Isaac (personal correspondence), the name of Myrddins sister (called Ganieda in Geoffrey of Monmouth) is perfectly transparent as it stands, meaning Bright like the Day, with gwen, the feminine form of gwyn white, bright, and dydd day. I have found, however, an Old Irish feminine name Findad/Findadh/Findath, mother of the 6th century St. Fintan of Clonenagh. Findadh, which means fair, blonde lady, looks to me to be cognate with Welsh Gwenddydd.
Geoffrey has her build Merlins house or astronomical observatory with its 70 doors and 70 windows, where he can watch Phoebus (Apollo the sun god) and Venus, it has been customary to see in her the goddess Venus herself. One thinks of Virgils description of the goddess in his AENEID as dea candida, the Bright or Shining Goddess. This megalithic house is the prototype for the stone, i.e. capstone of a cromlech or dolmen, under which the Lady of the Lake entraps Merlin in later romance.
Are there grounds, then, for seeing Gwenddydd as a goddess? As she is presented to us in Geoffrey of Monmouths Life of Merlin, she would appear to be mortal and to function as a priestess. But as I have written elsewhere, the Lady of the Lake was none other than the Irish goddess Nemhain (Frenzy). So if the Lady of the Lakes entrapment of Merlin derives from Gwenddydds building of a megalithic observatory for her brother, might not we tentatively identify her with Nemhain? Could not Gwenddydd be an epithet for the goddess, rather than just a name?
Gwenddydds building of an astronomical observatory for her brother has often been compared with Merlins own building of Stonehenge. And, indeed, authorities have interpreted Gwenddydds building of Merlins house as a reference to Stonehenge. The problem, of course, is that Stonehenge does not have 140 gaps between its circle of stones or even 70, if we take the doors and windows to be the same openings duplicated. Furthermore, the context of the Life of Merlin, which describes the building of the observatory by Gwenddydd, leaves little doubt that this particular stone circle is somewhere in the Scottish Lowland Caledonian Forest not on Salisbury Plain in England.
Given that it was Geoffrey who first identified Merlin with the Ambrosius (Divine or Immortal One) who was said to be Lord of Gwynedd and who can be shown to be not a historical personage, but the god Lleu, and given that Lleu and Mabon the youthful sun god were identified in Welsh tradition (they are both placed in death at Nantlle, the Bright Stream or Stream of Lleu, in Gwynedd), the only real possibility for Gwenddydds stone circle is that of the Lochmaben Stane (originally Clach-mabon, the Stone of Mabon) on the north side of the Solway. While now only a single standing stone remains, a credible mid-19th century witness claims there was a large stone circle on the site whose stones were cleared to make room for cultivation.
Currently, there is no stone circle in the Scottish Lowlands with 70 stones. This does not mean there never was such a circle, as some have been denuded of many of their stones and others are known to have been completely removed. The closest such circle to the Lowlands would be Cumbrias Long Meg and Her Daughters at Little Salkeld, sometimes called the Maughanby Circle. According to Ekwall, Maughonby preserves the Old Welsh personal name Merchiawn. There was a famous Dark Age Merchiawn, father of the Cynfarch or March who left his name at the Mote of Mark hill-fort in southern Dumfriesshire. It was this Cynfarch who was the brother of Uther, the father of Arthur (see my Shadows in the Mist: The Life and Death of King Arthur). Whether the Maughanby Merchiawn was Arthurs grandfather is not, alas, something we can establish.
One wonders, however, if the 70 doors and 70 windows are to be taken seriously as gaps between the upright stones of a circle. After all, Geoffrey claims Venus is to be watched from the house Gwenddydd builds, and we know this planets greatest elongations take place approximately 70 days before and after inferior conjunction, i.e. Venus passes from greatest elongation as an evening star to greatest elongation as a morning star in about 140 days.
Lastly, it might be revealing to investigate the name of Myrddins lord, Gwenddolau. As it turns out, Gwenddolau is a Cymracized form of another name which is present in several early historical sources. Gwenddolau, who perished c. 573, represents the Pictish king Cindaeladh/Cennalath of the Irish Annals (Tigernach and Ulster, respectively), who died. c. 580. In the Pictish Annals, Cindaeladh/Cennalath is called Galam cennaleph, cennaleph being a byname.
Galam Cennaleph ruled alone 552-3
G. Cennaleph ruled with Brude son of Maelcon 553-4
Brude son of Maelcon ruled 554-84
Peredur (Praetor, a title for the Roman governor at York) and his brother Gwrgi were present at the Battle of Arfderydd in 573, but according to the Welsh Annals they did not die until 580 exactly the year that saw the death of Cindaeladh/Gwenddolau.
NOTE: SOME OTHER PLACES ASSOCIATED WITH MERLIN
There are some other Merlin sites whose locations are uncertain. Four of these are the Fountain of Barenton, the Forest of Broceliande, a tomb of Merlin and Merlin's springs of Galabes.
The Fountain of Barenton is none other than the mineral springs of Berrington (Berinton) near Tenbury Wells in Herefordshire.
Broceliande is usually put in Brittany where, however, the name has not survived. Chretien says absolutely nothing about his knight leaving Britain. In this Chretien was right. Broceliande, also called Brecheliante, has been improperly identified with Breckland in Norfolk, the land of the brakes. I say improperly because the name Breckland was coined in 1925 by the archaeologist W. G. Clarke (information courtesy Edwin Rose, Norfolk County Council archaeologist).
Merlin's Forest of Broceliande (in Old French Briosque + "land") should instead be connected with Dumfries. While Dumfries was once thought to be the "Fort of the Frisians", authorities beginning with Chalmers (see Watson) correctly identified the <fries> component of the place-name with Gaelic preas, Angl. pres(s), gen. phris, Angl. -fries, gen. pl. preas, (b)p(h)reasach, "bush, copse, thicket". Niviane is Nefain mother of Urien Rheged and, quite possibly, a borrowing of the Irish goddess Nemhain.
A previously unlocated grave of Merlin is said by the Prose Lancelot to be in the Perilous Forest of Darnantes atop a mountain. Darnantes or Dar-nantes is the River Dore, which flows through the Golden Valley in the Black Mountains. Dore is either from French D'ore, "golden", or W. dwr, "water", while -nantes is from W. nant, "stream, brook". The Perilous Forest of the Dore River must be in this area, which is still forested to this day. Only a couple miles west of the Dore is Mynydd Merddin, "Myrddin's Mountain", one of the traditional Welsh sites of Merlin's tomb. Watling Street ran down the Golden Valley to Caerwent (information courtesy Chris Chandler of the National Monuments Record, citing James G. Wood's 1993 article "Primary Roman Roads Into Herefordshire and Monmouthshire and the Crossings of the Severn"). However, as Mynydd Merddin is an outlier of the Black Mountains, this could well be a relocation of Merlins mountain in the North (see above).
As for Merlin's spring or springs of Galabes, Geoffrey of Monmouth places this site in the region of the Gewisse. In a note to his The Quest for Merlin (pp. 270-271), Tolstoy suggests that Geoffrey may have substituted the Gewisse for Nennius's Guunessi. This would mean, of course, that Galabes would be found in Guunessi. Tolstoy would appear to be mistaken here. Merlin's Galabes is pretty plainly Nennius's Guoloph, i.e. Wallop, the site of a battle between Aurelius Ambrosius and Vitalinus. Now there is a Wallop stream in the Shropshire of Vortigern, but there is another in Hampshire, the Wallop Brook, site of the villages known as the Wallops. Hampshire is within the territory of the Gewisse.
The Cair Guorthirgin of Guunessi has been identified with a site at Nant Gwrtheyrn near the northwest coast of Llyn between Yr Eifl and Nefyn (see the Gwrtheyrn entry in P.C. Bartram's A Classical Welsh Dictionary). Guunessi (Gwnnws, Gwynnys) is now a farm two miles south of Nant Gwrtheyrn.
The Spirit of the Wood is Copyright © 2008 and 2009, August Hunt. All rights reserved. Used with permission.
Comments to: August Hunt
VortigernStudies is copyright © Robert Vermaat 1999-2009. All rights reserved