Shadows in the Mist: The Quest for a Historical King Arthur: COMPLIMENTARY COPY OF MY FINAL (!) ARTHUR NONFICTION BOOK

Here is my final Arthur book (at last). I have opted for a unification, so to speak, of my plotting out of the Arthurian battles with an adherence to Welsh tradition in regards to the most probable identification of Arthur's father. After some three decades of research, this is my best effort.

My apology for any remaining typos or format issues. Since writing and publishing this book, more supportive material for the theory has appeared on my blog site. I have not as yet incorporated the new material into the present volume. This is something I may or may not do sometime in the future.

THE BATTLE-LEADER

OF RIBCHESTER:

A DEFINITIVE IDENTIFICATION OF THE LEGENDARY KING ARTHUR

August Hunt

The Battle-Leader of Ribchester: A Definitive Identification of the Legendary King Arthur

Cover Photo: Sarmatians on Trajan’s Column.

ABOUT THE AUTHOR

August Hunt has a lifelong passion for the Arthurian stories and has been studying them since his youth. He has lectured extensively on King Arthur at colleges and for re-enactment organizations. His articles on British Dark Age topics are also featured on various award-winning websites.

Drawing on his considerable knowledge of folklore, heroic legend and myth, as well as place-name studies, history and archaeology, August is providing new and challenging material which illuminates many of the previously shadowy areas of the Arthurian tradition.

August holds a degree in Celtic and Germanic Studies and is a member of the International Arthurian Society. When he is not engaged in research and writing, he enjoys designing and building stone circles and other monuments that reproduce the celestial alignments of their ancient European counterparts.

His other Arthurian books include:

The Mysteries of Avalon: A Primer on Arthurian Druidism

A blog is maintained at:

https://mistshadows.blogspot.com/

THE BATTLE-LEADER OF RIBCHESTER:

A DEFINITIVE IDENTIFICATION OF THE LEGENDARY KING ARTHUR

AUGUST HUNT

TO SIR THOMAS MALORY

For All of the Adventures

TABLE OF CONTENTS

ACKNOWLEDGMENTS 9

MAP OF BATTLE SITES 12

THE KING WHO ONCE WAS 13

BEFORE ARTHUR: AMBROSIUS,

CUNEDDA AND VORTIGERN 24

ARTHUR’S ANCESTRY: RESTORING A

GENEALOGY 65

THE BATTLES OF ARTHUR 137

ARTHUR’S OTHER BATTLES:

MYTHOLOGICAL OR MISTAKEN 202

THE NORTHERN KINGDOMS 225

THE GRAVE OF ARTHUR 272

THE KING WHO WILL BE AGAIN 277

APPENDIX I: A NORTHERN PROTOTYPE

FOR THE ARTHURIAN GRAIL CASTLE 284

APPENDIX II: SEGANTII OR SETANTII? 287

APPENDIX III: LUCIUS ARTORIUS

CASTUS 296

BIBLIOGRAPHY 303

“It is all true, or it ought to be; and more and better besides.”

Sir Winston Churchill, on the legend of King Arthur

For friend and foe were shadows in the mist,

And friend slew friend not knowing whom he slew;

And some had visions out of golden youth,

And some beheld the faces of old ghosts Look in upon the battle; and in the mist

Was many a noble deed, many a base,

And chance and craft and strength in single fights

And ever and anon with host to host

Shocks, and the splintering spear; the hard mail hewn,

Shield-breakings, and the clash of brands, the crash

Of battleaxes on shattered helms, and shrieks

After the Christ, of those who falling down

Looked up for heaven, and only saw the mist;

And shouts of heathen and the traitor knights,

Oaths, insult, filth, and monstrous blasphemies,

Sweat, writhings, anguish, labouring of the lungs

In that close mist, and cryings for the light,

Moans of the dying, and voices of the dead.

‘The Passing of Arthur’ from Idylls of the King, Alfred Lord Tennyson

ACKNOWLEDGMENTS

I owe a great debt of gratitude to Dr. Richard Coates of the University of the West of England, Dr. Andrew Breeze of The University of Navarre and Dr. Isaac Graham of the National University of Ireland, Galway, for their acumen in treating of many word problems, tricky and obscure, and to Robert Vermaat, whose critical attention to many of my ideas often served to separate reasonable argument from mere fanciful construction.

My heartfelt appreciation also goes out to the following correspondents, whose kindness, patience and dedication helped me put the pieces of the Arthurian puzzle together: Elizabeth O’Brien, UCD Mícheál Ó Cléirigh Institute; Andrew Hawke, National Dictionary of Wales; Peter Wihl, Carmarthenshire place-name expert; Dafydd Hawkins, Powys place-name expert; Kevin Coyle, University of Ottawa; Paul Cavill, The English Place-Name Society; Chris Chandler of English Heritage; Andrew Deathe, Salisbury Museum; Hywel Wyn Owen, University of Wales, Bangor; Richard Coates, University of the West of England; Padraig O Riain, University College, Cork; Sigmund Eisner, University of Arizona, Emeritus; Gareth Bevans, National Library of Wales; Hoyt Greeson, Department of English, Laurentian University; Paul Acker, Saint Louis University; Gregory S. Uchrin, Catholic University of America; Jean-Yves le Moing; Christian Rogel, Director of the Bibliotheque du Finistere, Quimper; Helen McBurnie, Cramlington Parish Secretary; Neil Moffat, Reference and Local Studies Department, Dumfries and Galloway Libraries, Information and Archives, Dumfries and Galloway Council; Peter Drummond, Scottish Place-Name Society; Mark Douglas, Principal Officer for Heritage and Design, Planning and Economic Development, Scottish Borders Council; Nicola Hunt, Projects Officer of the Borders Forest Trust; Helen Darling, Part-Time Local Studies Librarian, Library Headquarters, St. Mary’s Hill, Selkirk; Jennifer Parkson, Map Library, Assistant for the National Library of Scotland; Henry Gough Cooper, Scottish Place-Name Society; Neil Bettridge, Archivist, Derbyshire County Council’s Record Office; John Reid, Scottish Place-Name Society; Beatrix Faerber, CELT Project Manager; Ceridwen Lloyd-Morgan, Assistant Archivist, Department of Manuscripts and Records, The National Library of Wales; Brynley F. Roberts, Centre for Advanced Welsh and Celtic Studies, University of Wales; Patrick Sims-Williams, University of Wales; Bruce Jackson, Lancashire County Archivist; Humphrey Welfare, Planning and Development Director, North, English Heritage; Richard Annis, Durham University’s Project manager of Archaeological Services; Tim Padley, Keeper of Archaeology, Tullie House Museum and Art Gallery, Carlisle; Georgina Plowright, Curator English Heritage Hadrian’s Wall Museums; Stephen White, Carlisle Library; Robert Collins, Newcastle Upon Tyne Museum of Antiquities; Kevan W. White of roman-britain.org; Gill Stroud, Sites and Monument Records Officer, Derbyshire County Council; Ken Smith, Cultural Heritage Manager for the Peak District National Park Authority; John Moreland, Reader at the University of Sheffield, Department of Archaeology’ Sue Palmer, Assistant Museums Manager of the Buxton Museum and Art Gallery, Oliver J. Padel, Cambridge University.

BIRTHPLACE, BATTLES AND BURIAL SITE OF ARTHUR WITH THE ROMAN DERE STREET

(Courtesy Hillarie Hunt)

INTRODUCTION

THE KING WHO ONCE WAS

What little we know of an ‘historical’ Arthur is contained in two early medieval works: the Historia Brittonum* or History of the Britons, ascribed to the Welsh monk Nennius, and the anonymous Annales Cambriae or Welsh Annals. These two sources supply us with the names of thirteen Arthurian battle sites. Twelve of these battles were supposedly fought against the invading Saxons, while one may have involved a conflict with another British chieftain named Medraut, the Mordred of later Arthurian romance.

The first twelve of these battles are all found in the HB immediately after mention of Aesc son of Hengist’s rise to the kingship in Kent, an event dated to 488 CE in the Anglo-Saxon Chronicle, and just prior to a section dealing with the Saxon kingdom of Bernicia and its king, Ida. Bernicia, coupled with Deira, comprised what became known as Northumbria, i.e. that portion of Britain that extends from the Humber River in the south to the Firth of Forth in the north. Ida began to rule, according to the ASC, circa 547 CE.

Camlann, the thirteenth battle, is found only in the AC, where it is dated to 537 CE. Thus the thirteen battles of Arthur are chronologically fixed within the period of 488 to 547 CE or from the latter part of the 5th century to the middle of the 6th. While several alternate chronologies have been proposed for the ASC and certain entries of the AC, for the sake of clarity the traditional dates will be allowed to stand.

The list of Arthurian battle sites, in the order that they occur in the HB and the AC, are as follows:

1) ostium fluminis quod dicitur Glein, mouth of the river Glein

2), 3), 4) & 5) flumen quod dicitur Dubglas, et est in regione Linnuis, river Dubglas in the Linnuis region 6) flumen quod vocatur Bassas, river Bassas

7) silva Celidonis, id est Cat Coit Celidon, Celidon Wood, Battle of Celidon Wood

8) castello Guinnion, castle of Guinnion

9) urbe Legionis, City of the Legion

10) litore fluminis quod vocatur Tribruit, river-shore Tribruit

11) monte qui dicitur Agned, mount Agned or monte qui nominator Breguoin, mount Breguoin

12) Badonis (AC), monte Badonis (HB), mount Badon (cf. Badonici montis of Gildas, who first mentioned Badon in his 6th century work, De Excidio Brittonum, The Ruin of Britain)

13) Camlann (AC), Camlan

In the HB, Arthur is called a dux bellorum or ‘leader of battles’, and is said to have fought alongside British kings against the pagan barbarians. It is from this bare listing of battle sites that the great body of Arthurian literature – the so-called ‘Matter of Britain’ – has grown. The consensus view among Arthurian scholars today is that the subsequent poems, stories, pseudo-histories and romances focusing on Arthur and his court are so heavily fictionalized, so overlaid with mythic, legendary and folkloristic elements, as to be worthless for the study of Arthur as a true Dark Age personage.

There are even those who dispense with the HB and AC Arthurian accounts as well, claiming that there is no way for us to substantiate the genuineness of either.

Some scholars go even further in refusing to accept as historically viable in entirety the HB or AC themselves. Indeed, to many the HB is no more than a hodge-podge of historical traditions which in all likelihood has little bearing on the actual events that transpired in Dark Age Britain.

A complication concerns the inability to clearly identify the place-names supplied in the battle list. The tendency has existed for some time to ‘make the places fit the theory’, rather than the opposite. Thus Arthur has been situated just about everywhere in Britain. Artificial geographical patterns have been sought for the battles in order to pinpoint Arthur’s power centre and shed dubious light on his origins. Sound philological principles have all too often gone by the wayside when treating of Arthurian place-names. It is precisely the inability to satisfactorily pin down Arthur’s battles that has led some scholars to give up the quest and join with those who insist on his non-historicity. For without firm battle site identifications, nothing of the historical Arthur can be known.

* Abbreviations: HB (Historia Brittonum), AC (Annales Cambriae) and ASC (Anglo-Saxon Chronicle) in future references.

To counter the argument that refuses to acknowledge the validity of the battle list, the two Arthurian entries in the AC have frequently been cited. These entries are typical, dry, bare-boned annalistic accounts of battles. Arthur, Medraut and the battle sites of Mt. Badon and Camlan are mentioned in the context of many other proper and place-names, all of which are demonstrably historical in nature. According to this line of reasoning, we need not doubt the veracity of the two entries.

516 an. Bellum Badonis, in quo Arthur portavit crucem Domini nostri Jesu Christi tribus diebus et tribus noctibus in humeros suos et Brittones victores fuerunt.

“The Battle of Badon, in which Arthur carried the cross of our Lord Jesus Christ for three days and three nights on his shoulders and the Britons were the victors.”

537 an. Gueith Camlann in quo Arthur et Medraut corruerunt, et mortalitas in Brittannia et Hibernia fuit.

“The Battle of Camlan, in which Arthur and Medraut fell, and there was plague in Britain and Ireland.”

Mt. Badon and Camlan are both, however, subject to the same kind of geographical shuffling as the other battle sites. Cases have been made for northern and southern Badons and Camlans. Few have been particularly convincing. Also, what may be legendary accretions similar to those present in the HB’s description of Arthur’s battle at Castellum Guinnion are to be found in the AC entry on Badon.

Octavum fuit bellum in castello Guinnion, in quo Arthur portavit imaginem sanctae Mariae perpetuae virginis super humeros suos, et pagani versi sunt in fugam in illo die, et caedes magna fuit super illos per virtutem Domini nostril Jesu Christi et per virtutem sanctae Mariae virginis benetreis ejus.

“The eighth battle was in Castle Guinnion, and in it Arthur carried the image of the holy Mary, the everlasting Virgin, on his [shield], and the heathen were put to flight on that day, and there was a great slaughter of them, through the power of our Lord Jesus Christ and the power of the holy Virgin Mary, his mother.”

Such embellishments have convinced many that the Badon entry in the AC should be disqualified as a record of a true Arthurian battle. In this case, it can be plausibly argued that the AC Badon entry has been contaminated by the HB’s account of Arthur’s battle at Castle Guinnion. This is not to say that Badon itself is denied status as an historical event; only that the placement of Arthur at Gildas’s Badon should be interpreted as an instance of hero-making and nothing more.

Gildas himself neglected to include in his work the name of the British commander at Badon:

26. … usque ad annum obsessionis Badonici montis, novissimaeque ferme de furciferis non minimae stragis…

“This lasted right up till the year of the siege of Mount Badon, pretty well the last defeat of the villains, and certainly not the least.”

Admittedly, in recent years there has been a sort of cautious reaction to the views set forth by proponents of a non-historical Arthur. While respecting the limitations imposed by the nature of the earliest Arthurian sources, limitations that the critical analysis of texts has largely defined, a handful of scholars have made significant headway in dealing with what they believe to be a fundamental over-statement of the problem of Arthur’s historicity. These scholars do not object to the actual process of critical analysis, but to some of the conclusions that have been drawn from the results of such analysis. The said conclusions, when treated of logically, can be revealed as arbitrarily formed and thus are reflections of expert opinion or even prejudice or bias, and not objective fact.

The ‘Arthur Problem’, put in the simplest terms, is this: is there sufficient reason for seeing the Arthur of the HB and AC as a plausible historical entity? Those who choose to see Arthur as a non-historical personage may strenuously object to this question. They would doubtless prefer that the problem be stated differently, e.g. is there sufficient evidence for seeing the Arthur of the sources as a historical entity?

Unfortunately, demanding evidence of the kind that would satisfy the proponents of the non-historical view automatically removes Arthur from the realm of historical study. Happening upon complimentary textual evidence from a source or sources deemed authentic and dependable seems a remote possibility. Archaeology, despite the ever-increasing light it sheds on Britain’s Dark Age past, has so far failed to yield anything substantive on Arthur. By refusing to allow for the possibility that Arthur may conceivably be historical, scholars engage in a sort of self-fulfilling prophecy, the fulfillment of which can only be that Arthur will continue to be found ineligible for historical status. Along with maintaining such perpetual ineligibility is a steadfast refusal on the part of scholars to allow interested parties to engage in research that might be deemed related even tangentially to Arthur as a possible historical phenomenon.

It may ultimately prove true that the only value in further analysis of Arthur’s battle sites might be an elucidation of the 9th century’s perspective on and attitude towards a reputed 5th-6th century British war-leader. Granted, there is some indication that the battle list as found in the HB is not an artificial construction undertaken by the monk Nennius, but instead preserves the content, rather than strictly the form, of a much earlier heroic poem originally composed in Arthur’s honor. If this widely held view is correct, then the battle list may reflect something other than a late traditional portrait of Arthur. It may be much more of a contemporary record of campaigns than the 9th century source in which it is embedded might otherwise suggest.

Still, if findings that arise from additional probing into the probable locations of Arthurian battle sites accomplish nothing other than to bring more into focus how the 9th century Britons interpreted their own remote past, then we will still have greatly advanced our knowledge of the period.

The burden of proof is just as much on the shoulders of those who dismiss Arthur as non-historical as it is on those who conditionally accept him as historical. Such an acknowledgment forces us to accept the possibility that Arthur existed without having to entertain the probability. If it can be demonstrated, based upon our knowledge of his battles, that an Arthur in the time period under consideration is a plausible phenomenon, then we can open a doorway into new areas of intellectual endeavor whose express purpose is to provide the impetus for the eventual discovery of evidence needed to historicize this British war-leader.

If there were to be an implied philosophy underlying this book, it would be that scholars of Arthuriana or Dark Age Britain ought not to view with disdain objective exploration of the potential historicity of Arthur. For as it may well turn out, the sources we do possess for the military career of this Dark Age figure may prove to have validity after all. While the means of providing such validity are currently not available to us, to state as fact that Arthur is not a historical entity or that we are not justified in seeing him as being even plausibly historical, is to risk making one of the biggest blunders imaginable in the annals of academic investigation.

It is the business of Arthurian and Dark Age scholars to consider possibilities. By possibilities is not meant, of course, wild theories that have no hope of ever being substantiated. Instead, possibilities in this context can best be defined as plausible historic scenarios that, while they may not be testable at the moment, may prove to be so in the future. Such scenarios must, needless to say, fit into the general, though wonderfully complex and interdependent tapestry created for us by universally accepted disciplines of study.

As more and more data comes in from these disciplines, and the resulting picture of the past is altered or refined accordingly, those scenarios that fail to conform in a manner deemed appropriate can be dispensed with. Eventually, with the aid of increasingly sophisticated scientific tools, our knowledge of Dark Age British history will be much greater than it is now. Most plausible scenarios will have been discarded. An historical Arthur might well be one of these casualties. Only a few scenarios – perhaps, if we are extraordinarily fortunate, just one – will remain solvent.

But until then, summarily deleting Arthur from the pages of our history books is not an ethical or reasonable solution to the ‘Arthur Problem’. It would be wiser and less shortsighted to include him, albeit with the necessary caveats. Any further evidence that supported Arthur’s historicity could thus be uncovered earlier, rather than later.

The present book, therefore, operates under the premise that precisely because Arthur may be historical, it would be intellectually prudent to apply more effort to the study of the only textual evidence we do have regarding this Dark Age British war-leader, i.e. the battles listed in the HB and AC. On the other hand, to ignore the battles themselves as possible historical events would be to intentionally turn a blind eye to evidence that has not yet been thoroughly evaluated. Potentially, a comprehensive examination of the battle sites, if undertaken with no agenda, nationalistic or otherwise, might harvest some new information on Arthur. And any new information, whether it ends up contributing arguments for or against a historical personage, is the proper goal of true scholarship.

The method employed by the author will be to utilize sound philological and geographical principles in the context of the Arthurian battles in order to arrive at several new site identifications. Included in this analysis, by necessity, will be a brief consideration of those past and present identifications deemed to be of a more respectable nature. But first will come some speculation regarding the period just before Arthur, with special emphasis on the figures of Ambrosius, Cunedda and Vortigern. Arthur’s origins will then be explored, utilizing the earliest versions of ancient Welsh genealogies, as well as the etymology of the name Arthur and its historical attestations in both Roman and Dark Age Britain. The remainder of the book will explore the Dark Age British kingdoms in the North and the power centres and grave of Arthur. These various investigations will produce a theoretical reconstruction of the life and death of ‘King Arthur’.

The reader should understand that many proper and place-name authorities have been consulted in the preparation of this book, either via personal correspondence or through their published works or both, and I have listed these generous and often patient contributors on my Acknowledgments page and in the Bibliography. Any conclusions I have drawn by relying on scholarly elucidation are solely my own and do not in any way reflect the opinions of the scholars themselves.

CHAPTER 1

BEFORE ARTHUR: AMBROSIUS,

CUNEDDA AND VORTIGERN

Ambrosius

Aurelius Ambrosius, said to be a Roman, is the most famous figure in Dark Age British history prior to Arthur. Why? Because he is credited with having united the Britons in a successful defense of the country against the Saxons, who from Vortigern’s time had, according to the traditional account, pillaged and conquered at will.

Ambrosius is important also because it has been fashionable to identify him with Arthur. As we shall see, such an identification is patently impossible.

To begin, Ambrosius was not a contemporary of Arthur. He was not, in fact, even a contemporary of Vortigern, who preceded Arthur by a century. And this is true despite the HB account, which brings Vortigern and Ambrosius (as the Welsh Emrys) together for a fabulous story that takes place at Dinas Emrys in northwestern Wales (see below).

There are major problems with accepting Ambrosius as a contemporary of Vortigern. First, he cannot have been a Roman and been in Britain during or after Vortigern’s rule. The withdrawal of the Romans is firmly dated at c. 409 CE. Vortigern’s ruling dates, depending on the sources consulted, are anywhere from twenty to forty years after the Roman withdrawal. If he were a Roman during or after Vortigern, then he came from the Continent and was not a native Briton. The argument could be made that ‘Romanized’ Britons continued to preserve the Roman way of life in southern England for a half century after the withdrawal of the troops. In this sense, a chieftain like Ambrosius might still consider himself to be ‘Roman’.

However, the HB tells us that Ambrosius fought a battle against a certain Vitalinus at a Guoloph or Wallop, thought to be the Hampshire Wallop. This Vitalinus is listed in the HB as the grandfather of Vortigern. This means that Ambrosius has wrongly been placed in the time of Vortigern. He actually belongs to the time of Vitalinus, who was probably of the 4th century.

The father of the famous 4th century St. Ambrose bore the name Aurelius Ambrosius. This man was, furthermore, the prefect or governor of Gaul (Gallia). Britain, Spain and Gaul were in the Gallic prefecture. So, we have here a historical figure named Aurelius Ambrosius who not only was a true ‘Roman’, but who could have had something to do with military operations carried out in Britain in the 4th century.

There is good reason to believe that St. Ambrose himself bore the name Aurelius. Jones' Prosopography of the Later Roman Empire gives no second name for the bishop of Milan and neither does Paulinus of Milan's Vita. Ambrose may have belonged to the gens Aurelia, as we know that he was related to Symmachus [Quintus Aurelius Symmachus]; an inscription refers to him as Aurelius Ambrosius. It is true that there is a debate over the Ambrose referred to in the inscription. Those who think it is Ambrose junior [St. Ambrose] point out that a dedication to St. Nazarius is involved. The point may be moot: if Ambrose senior belonged to the gens Aurelia, so did the son, and vice versa. Incidentally, St. Ambrose’s Milan was anciently called Mediolanum and there was also a Roman town of this name in the British Cornovii tribal territory.

One other factor strongly indicates that there is no good historical reason for accepting a 5th century Aurelius Ambrosius in Britain. Vortigern’s only interaction with Ambrosius, or Emrys Guletic (‘Prince Ambrosius’) as he is called in Welsh tradition, is in the Dinas Emrys folktale already alluded to above.

Other than Dinas Emrys, there appears to be no site in Britain which can be shown to contain the personal name Ambrosius. Still, this hero may even have been placed at Guoloph/Wallop because of the proximity of this stream to Amesbury. As Geoffrey of Monmouth did much later, Ambrosius's name was fancifully associated with Amesbury.

The town name does not, in fact, seem to contain the personal name Ambrosius. Its etymology is instead as follows:

Ambresbyrig, from a c.880 CE charter, then various spellings to Amblesberie in Domesday. Almost certainly a personal name Ambre or Aembre cognate with the Old German Ambri, hence Ambre's burgh, cf. Ombersley. All the early forms for Amesbury have the medial -b-, but no form has any extension that would justify derivation from Ambrosius. See https://mistshadows.blogspot.com/2018/06/ambirix-as-name-preserved-in-place-name.html.

Ambrosius as a Latin adjective means “the Divine or Immortal One”. As such, it was at some point taken to be a title for the Welsh god Lleu. Welsh tradition made Lleu the ancient ruler of Gwynedd, and this is the rank granted to Emrys or Ambrosius in the HB. Hence Dinas Emrys in northwestern Wales, the ‘[Hill-] fort of the Divine or Immortal One’, is actually the Fort of Lleu.

The Welsh also appear to have identified the youthful god Mabon with Lleu. That this is so is demonstrated by the placement of the two gods in death at the same place. According to the Mabinogion tale Math Son of Mathonwy, Lleu is found as the death-eagle in the oak tree at Nantlle (Nant Lleu) in Snowdonia not far from Dinas Emrys. And one of the Stanzas of the Graves reads:

“The grave on Nantlle’s height, No one knows its attributes – Mabon son of Modron the Swift.”

In Chapter 6, we will discuss Emrys’s Campus Elleti, supposedly a site in southern Wales, in the context of Camelot.

Geoffrey of Monmouth proceeded to further confuse the story of Ambrosius, a Roman governor of Gaul mistakenly identified with a Welsh god, by identifying both with the Northern Myrddin or Merlin. Hence we find Merlin or ‘Merlin Ambrosius’ in the Dinas Emrys story of Emrys/Lleu/Mabon.

In addition, Merlin is placed at the springs of Galabes, Geoffrey’s attempt at the Guoloph of the hero Ambrosius.

In conclusion, we can only say that there is no good reason for supposing that Vortigern and Ambrosius were contemporaries. Instead, the Ambrosius mentioned by Gildas as having military success in Britian must have been the 4th century Gallic governor of that name. This being the case, Ambrosius could not possibly have been the victor at the battle of Mount Badon, which is dated 516 CE. And, by extension, Ambrosius was not Arthur.

The Ghost Ambrosius or Why Arthur’s Predecessor Should be Stricken from the Annals of British History

Over the past several years, I've written a handful of articles on Ambrosius Aurelianus, a geographically and temporally dislocated figure in early British legend. Yet despite the evidence I've presented, Arthurian scholars, professional and amateur alike, continue to mistake him for a real personage of 5th century Britain. The idea that he might even be Arthur is still out there. I feel, therefore, that it is time for a summary treatment of this supposed military hero. The easiest way for me to do this is to itemize the points of my argument.

1) The name of A.A. matches perfectly that of the fourth century Governor of Gaul (whose territories included those of Britain) and his famous son, St. Ambrose. Vortigern's grandfather Vitalinus is said to have fought A.A. at Wallop in Hampshire. Such a battle reference puts A.A. well before Vortigern and negates the possibility that A.A. was a boy during Vortigern's reign.

2) St. Ambrose and his father lived at Trier on the Moselle. The Campus Elleti in Wales where Vortigern's men are said to have found the boy A.A. comes from a Welsh place-name Maes Ilid (see modern Llanilid) and this may be a substitute for the Moselle (Mosella/Mosellae). [For Llanilid as a scholarly error, see https://mistshadows.blogspot.com/2020/06/the-red-herring-of-llanilid-uther.html.] We also have a Mediolanum in what is now Powys, a town whose name is identical with that of St. Ambrose of Milan.

3) Dinas Emrys is a relocation for Amesbury, the latter thought (wrongly) to contain the name of Ambrosius. Dinas Emrys was placed in Eryri because this mountain range was fancifully connected to the Welsh word for eagle, and both St. Ambrose and Magnus the Tyrant (easily confused with Vor-tigern) are known to have been at Aquileia, a place-name that could have been incorrectly linked with the Latin word for eagle.

4) Trier was in Gallia Belgica, 'Gaul of the Belgae', and A.A.'s Wallop in Hampshire was in the ancient tribal territory of the British Belgae. Gallia could be used in medieval sources for both Gaul and Wales.

5) A.A. is said to have been given Dinas Emrys and the western kingdoms of Britain by Vortigern. This is impossible, as Gwynedd belonged to Cunedda and his sons. This is obviously a mistake for Amesbury, which was inside of what was to become Wessex, the kingdom of the WEST Saxons.

6) A.A appears to have been identified in folk belief with the god Lleu, styled Lord of Gwynedd,who was himself identified by the Welsh with the god Mabon. The Campus Elleti ballgame story is paralleled in the Irish story of Mac Og, the 'Young Son', the Gaelic version of Mabon.

7) A.A. was further identified with Merlin (Myrddin), himself possibly a form of Lleu or an avatar of that god.

8) The whole exhuming by A.A. of the two vases containing dragons (read “chieftains” or “warriors”) has its prototype in St. Ambrosius’ exhuming of pairs of saints. See https://mistshadows.blogspot.com/2018/01/st-ambrose-and-exhumation-of-saints.html.

In conclusion, the Ambrosius Aurelianus who first appears in the pages of Gildas is a purely legendary figure, based on the known historical Ambrosii of the Continent. He was mistakenly transferred to Britain during the normal course of folklore development, largely due to a confusion of place-names. There is no reason to believe that either Ambrosius - father or son - ever set foot on British soil. To concoct some famous war-leader of the Britons who happened to have been named after one of the Ambrosii is to ignore points 1-7 above.

But What About Gildas’s Ambrosius?

Although I have shown to my satisfaction why Ambrosius Aurelianus was not only wrongly placed in Britain, but put there at the wrong time, I've been asked a very good question by some of my readers: "That's all well and fine, if we're talking about the tradition recorded in Nennius and subsequent sources (like Geoffrey of Monmouth's pseudo-history). But what about A.A.'s appearance in Gildas? How do you account for that?"

As it happens, that is an excellent question. And not an easy one to answer. But I will take a stab at it, in any case.

A.A. was Prefect of Gaul (and thus of Britain as well) c. 337-340. We do not know when he died, but his son St. Ambrose (with whom he was conflated in Welsh legend) moved to Rome with his mother not earlier than 353 (https://books.google.com/books?id=sc49DwAAQBAJ&pg=PA7&lpg=PA7&dq=st.+ambrose+and+his+mother+went+to+rome&source=bl&ots=7w4smM9os3&sig=ACfU3U0AuKyqO3hjZIrPlxdpBvQVvfCZ5g&hl=en&sa=X&ved=2ahUKEwi_hL-SzqnpAhUOsp4KHZLYANQQ6AEwDHoECAgQAQ#v=onepage&q=st.%20ambrose%20and%20his%20mother%20went%20to%20rome&f=false). Some have thought A.A. may have fallen at the same time as his Emperor Constantine II, who died in 340.

In 343, Constantine's brother Constans, the new Western Emperor, visited Britain. It is not known precisely why (see http://www.roman-emperors.org/consi.htm#9), but the reason is hinted at in Ammianus:

Book XX

1 1 Lupicinus, master of arms, is sent with an army to Britain, to resist the inroads of the Scots and Picts.

Such was the course of events throughout Illyricum and the Orient. But in Britain in the tenth consulship of Constantius and the third of Julian raids of the savage tribes of the Scots and the Picts, who had broken the peace that had been agreed upon, were laying waste the regions near the frontiers, so that fear seized the provincials, wearied as they were by a mass of past calamities. And Julian, who was passing the winter in Paris and was distracted amid many cares, was afraid to go to the aid of those across the sea, as Constans once did (as I have told),1

1 In one of the lost books; it was in 343.

Let us suppose, for the sake of argument, that A.A. did not perish with Constantine. That although he was no longer serving as Prefect of Gaul, he accompanied Constans to Britain in some capacity. This is certainly not out of the realm of the possible. Granted, Constantine I/the Great had made the praetorian prefecture a civil, rather than a military post. But A.A. could have been replaced by another prefect, and found himself in another role as part of a major military expedition to Britain. It's also not inconceivable that A.A. fled to Britain after Constantine II's death, although had that been the case we would have expected him to take his family with him.

However it happened, if A.A. were in Britain at the time, how do we account for the sequence of events in Gildas?

Rather easily, I suspect. The problem has to do with a simple confusion of the two emperors named Constans - the one who was in Britain in 343 and the Constans II, son of the Constantine III who had been proclaimed emperor in Britain in 407.

A very puzzling line in Gildas has not, to my knowledge, been analyzed. It occurs in 25:2, and runs as follows: "After a time, when the cruel plunderers had gone home, God gave strength to the survivors." These survivors, and those who flocked to them, had as their leader A.A. On the surface, this would seem to be a nonsensical statement. The Saxons invited in by Vortigern did not, in fact, go home. Gildas had just previously told us that they had invited in more of their kind and proceeded to take over the island. We are told in Nennius that Vortimer pushed them to the Isle of Thanet, but that after he was slain they continued their depredations and conquest.

So who went home when A.A. showed up on the scene?

I would propose that Gildas' account is here hopelessly confused. The enemy that withdraws in this context was forced to do so by Constans I, accompanid by A.A., who may well have had the military command. We are probably talking about Scots and Picts, not Saxons. What we appear to have here is a simple jumbling of fourth and fifth century events.

Cunedda

The great Cunedda, called Cunedag (supposedly from *Cunodagos, ‘Good Hound’) in the Historia Brittonum, is said to have come down (or been brought down) from Manau Gododdin, a region around the head of the Firth of Forth, to Gwynedd. This chieftain and his sons then, according to the account found in the HB, proceeded to repulse Irish invaders. Unfortunately, this tradition is largely mistaken. To prove that this is so, we need to begin by looking at the famous Wroxeter Stone, found at the Viroconium Roman fort in what had been the ancient kingdom of the Cornovii, but which was the kingdom of Powys in the Dark Ages.

The Wroxeter Stone is a memorial to a chieftain named Cunorix son of Maquicoline. This stone has been dated c. 460-75 CE. Maquicoline is a composite name meaning Son [Maqui-] of Coline. The resemblance here of Cunorix and Coline to the ASC's Cynric and his son Ceawlin is obvious. Some scholars would doubtless say this is coincidence, and that the discrepancy in dates for Cynric and Ceawlin and Cunorix and (Maquicoline) are too great to allow for an identification. I would say that an argument based on the very uncertain ASC dates is hazardous at best and that if there is indeed a relationship between the pairs Ceawlin-Cynric and Coline-Cunorix, then the date of the memorial stone must be favored over that of the document.

There is also the problem of Cynric being the father of Ceawlin in the Anglo-Saxon tradition, while on the Wroxeter Stone it is (Maqui)coline who is the father of Cunorix. But such a confusion could easily have occurred simply by reading part of a genealogy list backwards.

While Ceawlin's father Cynric, the son of Cerdic of Wessex in most pedigrees, is capable of being derived quite well from Anglo-Saxon, the name could also be construed as an Anglicized form of the attested Celtic name Cunorix, Hound-king, the latter Welsh Cynyr.

Cerdic (= Ceredig) is not the only Celtic name in the early Wessex pedigree. Scholars have suggested that Ceawlin could be Brittonic.

Cunorix son of Maquicoline, based on an analysis of his name and the lettering employed on the inscription itself, is believed to have been Irish. It should not surprise us, then, to find Cunedda of Manau Gododdin, the reputed founder of Gwynedd, was himself actually Irish. There was an early St. Cuindid (d. c. 497 CE) son of Cathbad, who founded a monastery at Lusk, ancient Lusca. In the year entry 498 CE of the Ulster Annals, his name is spelled in the genitive as Chuinnedha. In Tigernach 496 CE, the name is Cuindedha.

The Irish sources also have the following additional information concerning St. Cuindid:

Mac Cuilind - Cunnid proprium nomen - m. Cathmoga m. Cathbath m Cattain m Fergossa m. Findchada m Feic m. Findchain m Imchada Ulaig m. Condlai m Taide m. Cein m Ailella Olum.

U496.2 Quies M. Cuilinn episcopi Luscan. (Repose of Mac Cuilinn, bishop of Lusca).

D.viii. idus Septembris. 993] Luscai la Macc Cuilinn

994] caín decheng ad-rannai, 995] féil Scéthe sund linni, 996] Coluimb Roiss gil Glandai.

trans: 'With Macc cuilinn of Luscae thou apportionest (?) a fair couple: the feast of Sciath here we have, (and that) of Columb of bright Ross Glandae'

The (later-dated) notes to this entry read: 'Lusk, i.e. in Fingall, i.e. a house that was built of weeds (lusrad) was there formerly, and hence the place is named Lusca ........Macc cuilinn, i.e. Luachan mac cuilinn, ut alii putant. Cuinnid was his name at first, Cathmog his father's name'.

Significantly, Lusk or Lusca is a very short distance from the huge promontory fort at Drumanagh, the ‘the ridge of Manach.’

Forgall's name is not really a name. It instead designates a rank, as is made clear by the entry in the eDIL for this word:

forgell, forgal(l)

Aire forgill a landed proprietor next in rank to a `ri': aire forggaill cid ara n-eperr? Ar is hé fortgella for na gráda doruirmisem, Críth G. 417 ( Laws iv 326 ) `testifies to [the character of] the grades'; `makes affirmation above the grades' Mac Neill, Law of Status 29 (see Críth G. p. 72 , B. Crólige p. 70 § 46 ). aire forgill, Críth G. 11 ( Laws iv 298 ). a. forggaill LL 29b22 . Called also: fer forgaill ZCP v 499 § 5 .

His epithet Monach is what I now believe lies behind the name of the Drumanagh fort. Again from the eDIL:

monach

adj o, ā. (mon) able to perform feats or tricks; dexterous, skilled. cach clessach na chanad cheilg | manach sein [i]sin Gædilg, LL 144b27 . ? fer manach craftsman (?), Hib. Min. 55.12 = f. manath, IT i 104.10 (see manath). n p. monaig ` equestrians' (= trick-riders?), Laws v 108.20 , with gl.: .i. bid ar monaib a n-each isna haenaigib `who stand on the backs of their horses (? perform feats of riding) in the fairs', 27 . As sobriquet: Forgall m.¤ the trickster (?), LL 144b26 . Forgull M.¤ .i. cleasuch ro bhoí, Cóir An. 205. ingen Forgaill manaich, LU 10176.

Forgall Monach is thus the 'skillful, high-ranking landed proprietor' or some such. As for his proper name, well, that is not difficult to determine. His fort at or near Lusk was called Luglochta Loga or Luglochtaib Loga. While the first part of this place-name has eluded specialists, Loga here is for the god Lugh. [I would render the place-name not as the 'Garden' of Lugh, which is suggested by an early gloss, but as Loc (log, lag, luic, lucc, etc.) + lucht, giving us a meaning of 'The Place of Those Belonging to Lugh.'] As Lugh was the skillful god and master crafsman in both Irish and Welsh tradition, it is fairly obvious to me that the Skillful, High-Ranking Proprietor is none other than Lugh himself.

The god Lugh was known as Lleu in Wales, and he was traditionally referred to as the Lord of Gwynedd. And it was Gwynedd which Cunedda and his sons (or teulu) came to control sometime after the Roman withdrawal.

Forgall Monach is credited with having another great fort, this one called the Bruidne Forgall Monach. Now, we can't know if this bruidne or 'hostel' is merely another name for Luglochta(ib) Loga or if it designates another nearby site. But I'm guessing the Ridge of Monach or the Drumanagh fort is this place.

Aeternus, Cunedda's father, is none other than Aithirne of Dun and Ben Etair just south of Lusca. Paternus Pesrudd (‘Red-Cloak’), Cunedda's grandfather, is probably not derived from Mac Badairn of Es Ruad (‘Red Waterfall’), since Es Ruad is in northwest Ireland (Ballyshannon in Co. Donegal). I think Paternus, from the L. word for ‘father’, is Da Derga, the Red God; Da, god, being interpreted as W. tad (cf. L. tata, ‘father’). The Da Derga's hostel was just a little south of the Liffey. Cunedda's great-great-grandfather is said to be one Tegid (Tacitus), while his great-great-great grandfather is called Cein. These two chieftains are clearly Taig/Tadhg and his father Cian. Cian was the founder of the Irish tribe the Ciannachta, who ruled Mag Breg, a region situated between the Liffey and either Duleek or Drumiskin (depending on the authority consulted). The Lusca and Manapia of Chuinnedha are located in Mag Breg.

According to the genealogy edited in Corpus Genealogiarum Sanctorum Hiberniae, the name of Mac Cuilind's father was Cathmug. He belonged to the descendants of Tadc mac Cian, otherwise called the Cianachta. There was a concentration of the saints of this family in the Dublin/Louth/ Meath area, corresponding roughly to the teritory of the Cianachta Breg.

It is surely not a coincidence that according to the Irish Annals Chuinnedha's other name was Mac Cuilinn. Obviously, Mac Cuilinn and the Maqui-Coline of the Wroxeter Stone are the same name and hence the same person. Gwynedd was thus founded by Chuinnedha alias Mac Cuilinn of the Manapii in Ireland, not by a chieftain of Manau Gododdin in Britain.

The Irish origin of Cunedda should not be a surprise to us, as there is the well-documented case of the Welsh genealogy of the royal house of Dyfed, which was altered to hide the fact that Dyfed was founded by the Irish Deisi. We know this because we have the corresponding Irish genealogy from a saga which tells of the expulsion of the Deisi from Ireland and their settlement in Dyfed. As is true of Cunedda's pedigree, in the Welsh Dyfed pedigree we find Roman names substituted for Irish names. There were other Irish-founded kingdoms in Wales as well, e.g. Brycheiniog.

What exactly the relationship was that existed between Cunedda and the British kingdom of Powys on the one hand, and Cunedda/Ceawlin of the Gewissei and the Saxons of southern England on the other, is something that can only be surmised once we plug Vortigern into the equation. While it is true that Bede called Ceawlin a Bretwalda, i.e. a preeminent ruler of Britain, we are not justified in equating him with Vortigern.

Vortigern

The name Vortigern or Gwrtheyrn, as found in the HB of Nennius, was once held to be a ruling title. It was thought to be represented by Gildas' Latin pun 'superbus tyrannus' or ‘Proud Tyrant’. However, we now know that Vortigern was a proper name and not a title. It is found recorded not only in several localities in Wales, but in Ireland as well.

Aside from the British Vortigern, whose name means ‘Over-lord’, we have records for the following Dark Age Irish Vortigerns or ‘Fortcherns’:

1) Fortchern, the smith of St. Patrick (Annals of the Four Masters Year Entry 448); as this Fortchern is paired with another smith, Laebhan, i.e. St. Lomman (?), this Fortchern may be:

2) Foirtchern son of Fedelmid, who was for a short time bishop of St. Lomman's Trim. Fortchern of Trim, who was of mixed Irish and British blood, is said to have later retired to Killoughterane/Cill Fortchern in the parish of Muinebeag, Co. Carlow. However, we are told in the ancient Irish sources that Fortchern the smith is the same as Foirtchern of Rath Seimhne (see below). It may not be a coincidence that there is a Gobbin's Cliff, the Cliffs of the divine smith Goban Saor, in Seimhne/Island Magee.

3) Vortigern of Ballyhank, East Muskerry, Co. Cork (inscribed stone).

4) Vortigern of Knockboy in Decies Without Drum, Co. Waterford (inscribed stone dated c. 700-900 CE).

5) Foirtchern of Monte Cainle (probably the Hill of Conlig/Coinleac in north Co. Down), a contemporary of St. Columba.

6)Foirtchern of Rath Seimhne (Island Magee, south Co. Antrim).

7)Fortchern, brother of Cathchern (a name cognate with British Cattigern, a supposed son of Vortigern in the HB narrative), son of Tigernach of the Meic Carthind of the Lough Foyle region.

8)Fortcheirn son of Mael Rubae of the Ui Dicholla of the Dessi

9)Fortchern son of Iarlaith of the Ui Brigte of the Dessi

10)Fortchern son of Tigernach of the Ui Brigte of the Dessi

11)Clan Foirtchern in the Breadach genealogy on Inishowen, near the Lough Foyle Meic Carthind

These examples, some ‘in stone’, should be sufficient to dispell the notion that Vortigern is merely a title. Instead, Vortigern is a genuine Brythonic personal name.

In Wales, Radnorshire or Maesyfed (the ‘field of Hyfaidd’) was once known as Gwrtheyrnion, i.e. the kingdom of Gwrtheyrn. Gwrtheyrnion, roughly between the Wye and Ithon rivers, was a relatively small kingdom in southwestern Powys. Other places in Wales where Vortigern's name is preserved are Nant Gwrtheyrn in Lleyn, close to Gwyniasa (and surrounding Gwynus placenames), and a Craig Gwrtheyrn on the Teifi.

These three places are mentioned in Nennius's narrative, but only Gwrtheyrnion carries weight. The Lleyn and Teifi sites may represent the presence in these places of other Vortigerns, but in all likelihood it is merely the proximity to them of St. Garmon place-names that accounts for the ‘Over-lord’s’ association with them. In Nennius's story of Vortigern, the poor chieftain is literally hounded all over Wales by the saint. Thus wherever there was a known St. Garmon site, Vortigern was placed there. In my opinion, Vortigern was probably not in Lleyn, nor was he on the Teifi (despite the presence at nearby Nevern of a Vitalinus Stone; see below). He belonged instead to Gwrtheyrnion, which was merely one of several Welsh Dark Age sub-kingdoms.

Vortigern of Wales, who is said to have been the son of Guitaul (= Roman Vitalis) son of Guitolin (= Roman Vitalinus, a name found on a stone at Nevern dated by Charles Thomas between 466 and 533 CE – too late for Vortigern’s grandfather) son of Gloiu (Gloyw, the eponym of Welsh Caerloyw, modern Gloucester), is actually the British-Irish Fortchern son of Fedelmid son of Laeghaire. This Fortchern son of Fedelmid was of the right time to be the Vortigern of Nennius. Both Guitaul and Guitolin are substituted for the name Fedel-mid.

It was Robert Vermaat who first called my attention to the details surrounding this particular Fortchern. To quote extensively from his Vortigern Studies website article,

‘Scotnoe & Foirtgirn, the Irish Branch’:

“Foirtchern was the son of Fedelmid, son of Loguire, who was High King of Ireland throughout the period of the mission of St. Patrick (whose dates may be 428-462). Foirtchern’s mother was a daughter of the King of the Britons. The story goes that when St. Patrick’s nephew Lomman visited Trim (in Ireland), the boy Foirtchernn took him home to Fedelmid and his mother, who both spoke British and were delighted to see a visitor from his mother’s country. They made Lomman stay, who then subsequently converted the whole family. The mother might have been a Christian in the first place, for she ‘welcomed’ the saint. Maybe the fact that Lomman was a Christian made him more welcome than his being from Britain. Fedelmid may have embraced Christianity because the saint had just come from Tara Hill, where St. Patrick had defeated the druids of Fedelmid’s father the High King Loguire.

Foirtchern's date may be confirmed by the Annala Rioghachta Eirann:

Annals of the Four Masters, M432.0 – 4

The Age of Christ, 432. The fourth year of Laeghaire. Patrick came to Ireland this year, and proceeded to baptize and bless the Irish, men, women, sons, and daughters, except a few who did not consent to receive faith or baptism from him, as his Life relates. Ath Truim was founded by Patrick, it having been granted by Fedhlim, son of Laeghaire, son of Niall, to God and to him, Loman, and Fortchern.

These annals, though dating to 1616 in their youngest version, date back at least to 1172.

In any case, Fedelmid enthrusted Foirtchirnn to Lomman and founded the church of Trim, making St Patrick, Lomman and Foirtchirnn his heirs. But Foirtchernn was obdurate and did not want to accept his heritage, after which Lomman had to threaten him with taking away the blessing of the church, which is tantamount to incurring its curse. After Lomman's death, though, Foirtchirnn gave away his church within three days. This may be apocryphal, for Foirtchirnn was listed afterwards as the first episcopus (abbot) after Fedelmid and Lomman. He might have given it up later though, for he is also listed as a plebilis, a lay successor.”

Now, the question on my mind, after reading this account, was "Who succeeded Lomman at Trim?" The answer is in the Patrician Texts in the Book of Armagh:

He [Foirtchernn] held the abbacy for three days after his master's death until he came to Ath Truim, and then immediately handed his church over to the foreigner Cathlaid [Cathlaido perigrino].

I immediately recognized this ‘Cathlaid the Foreigner’ as a doublet for Catel Durnluc, the traditional founder of Powys, the kingdom that succeeded that of the Roman-period Cornovii.

Fortchern son of Fedelmid's mixed ancestry allows for the possibility that he possessed or inherited lands on both sides of the Irish Sea. We know that there were several Irish-founded kingdoms in Wales at the time: the Deissi established a ruling dynasty in Dyfed, Brycheiniog was of Irish foundation, and Cunedda of Manau Gododdin, founder of Gwynedd, was actually Chuinnedha/MacCuilind of Drum Managh in Ireland. Cunedda and his sons are said to have chased the Irish Ui Liathain out of Anglesey, Dyfed, Gower and Kidwelly, the Laigin were at Dinllaen and in the Lleyn Peninsula, and there is the possibility that Dinevor in Ystrad Tywi was named for an Efwr Llwydon, i.e. of the Irish Laithain. The Irish mercenary Cunorix son of Maquicoline/Cunedda was buried in the heart of Powys at Viroconium. There is no difficulty, then, in accepting a Gwrtheyrnion as a sub-kingdom named after Fortchern son of Fedelmid.

The only objection to a Gwrtherynion ruled by a chieftain of mixed British-Irish ancestry would be that such a king, with such a small sub-kingdom, could not possibly be the ‘superbus tyrannus’ of Gildas. But I offer this argument to account for how such a confusion could have taken place: any chieftain possessing a name such Vortigern, ‘the over-/super-/great- lord’ could easily have been misinterpreted as an over-king similar to the ardrigh or ‘high-king’ of Ireland. If I am right and Fortchern son of Fedelmid son of Laeghaire the high king is the British Vortigern of Gwrtheyrnion, then this kind of royal descent from an ardrigh could also have contributed to Gildas's misinterpretation of Vortigern's status in Britain.

In summary, then, what may have happened is this: a chieftain named Vortigern (or Fortchern), who was of mixed Irish-British ancestry, and whose grandfather was the ardrigh of Ireland, had established a small sub-kingdom in southwestern Powys in the 5th century. Gildas, attracted to the name because it seemed to denote a sort of British high king, laid the blame for the Saxon ‘invitation’ (i.e. the use of Germanic barbarian federates) in this presumed high king's lap. Further vilification continued after this identification of Vortigern as the offending monarch was made, until by the 9th century we have a fully developed story of Vortigern in the HB of Nennius.

Alternately, given that the Eliseg Pillar in what was the kingdom of Powys traces the descent of the Powys dynasty from Vortigern, and Catel Durnluc is in the various genealogies confused with Vortigern or made his near-descendent, it is possible that Fortchern son of Fedelmid, at least partly through his wife’s British blood, had managed to lay claim to the throne of Powys itself. His sub-kingdom of Gwrtheyrnion was, after all, part of Powys.

A final possibility, and one which calls into some doubt the notion that Vortigern was related to the Irish high king, is the proximity of Gwrtheyrnion to Brycheiniog. The latter, as Charles Thomas has shown in his ‘And Shall These Mute Stones Speak?”, was likely founded by the Irish-descended Dessi dynasty of Dyfed. We have seen above that fully three of the Irish Vortigerns hailed from the Dessi.

What can be said, with a fair degree of certainty, is that Fortchern son of Fedelmid (?) and the Irish Cunedda were contemporaries. Also, the son of Cunedda was buried in honor at the capital of Powys, Viroconium. Cunedda, his sons and their ‘teulu’ or war-band composed what the Saxons of southern England came to call the Gewessei.

Some More on the Gewissei

According to Richard Coates (see his 'On some Controversy surrounding Gewissae/Gewissei, Cerdic and Ceawlin', NOMINA 13, 1989-90), Gewessei is "a nominalization of the [Old English] adjective gewis, among the meanings of which were 'sure, certain, reliable.'" These were federate warriors, the 'sure or reliable' ones. The Old English name nicely matches in meaning Latin fidus, from which comes foedus and then foederatus.

However, while reading through different MSS. of THE ANGLO-SAXON CHRONICLE and related early sources, I happened to notice that the name Cerdic, which belongs to the founder of the Gewissei, is often spelled CERT-. A /t/ for /d/ substitution is common for the period, so such a change need not seem at all surprising. But what was surprising is that I happened to think of Latin

certus, certa, certi, etc.

For certi, we have the following in William Whitaker's WORDS:

cert.i N 2 2 GEN S N

cert.i N 2 2 LOC S N

certum, certi N N [XXXBO]

that which is fixed/regular/definite/specified/certain/fact/reliable/settled;

cert.i ADJ 1 1 GEN S M POS

cert.i ADJ 1 1 GEN S N POS

cert.i ADJ 1 1 NOM P M POS

cert.i ADJ 1 1 VOC P M POS

certus, certa -um, certior -or -us, certissimus -a -um ADJ [XXXAO]

fixed, settled, firm; certain; trusty/reliable; sure; resolved, determined;

What I'm going to propose, here and now, is that Gewissei was a term conjured by the English, who instead of interpreting Cerdic's name properly, as deriving from

caredig, ceredig [GPC]

[bôn y f. caraf: caru+-edig: < Clt. *karatīcos]

kind, loving, friendly; beloved, dear; diligent.

beloved one, kinsman; lover, friend, benefactor.

itself from

câr [GPC]

[Crn. car, ll. kerens, Llyd. kar, ll. kerent, H. Wydd. car(a)e ‘cyfaill’: < Clt. *karants o’r gwr. *kā- ‘annwyl, trachwantus’]

*kar-o- 'love' [Vb] [Matasovic's ETYMOLOGICAL DICTIONARY OF PROTO-CELTIC]

GOlD: OIr. caraid, -eara; cechraid [Fut]; carais [Pret.]

W:MWcaru

BRET: MBret. caret, MoBret. kared

CO: Co. care

GAUL: Caro- [PN]

CELTIB: Kara [PN]

PIE: *keh2-ro- 'love' (IEW: 515)

COGN: Lat. ciirus, OHG huor 'prostitute', Latv. kiirs 'lustful'

SEE: *karant- 'friend'

ETYM: This is a deadjectival verb formed from unattested *karo- 'dear,

beloved' < PIE *kh2-ro-. The root is PIE *keh2- (Skt. kdma- 'love', etc.).

REF: LEIA C-36, GPC I: 422, LIV 306, Ellis Evans 1967: 162, Delamarre

107,Deshayes 2003: 370, MLH V.l: l60f.

they related it to Latin certus.

If this is so - and I'm pretty sure (pun strictly intended) it can't be merely a coincidence - then we can no longer view the Gewissei as "allies" based on nothing more than an applied meaning of their name. Instead, the Gewissei are the descendants of Cerdic/Certic, and can no no longer be defined as allies of the English. At this point I'm not certain (sorry!) what they were, exactly.

Via private communication, Dr. Richard Coates responded to this idea by saying:

"I hadn’t thought of that but it’s intriguing."

Cunedda and his teulu fought alongside Saxons against other British in the area. We can assume that as had been the case with Roman federates, Cunedda and his followers were given lands in Gwynedd in return for rendering military service to the old Cornovii kingdom. Even if these lands had been granted in a de facto manner, a peaceful and supportive relationship could be sustained with Powys by the adoption of federate status. Doubtless this process had its origin in the Roman period.

Iusay Son of Ceredig

Many years ago I floated the idea that Iusay, son of Ceredig son of Cunedda, may be a form of the family/tribal designation Gewissae or Gewissei. While a proposed relationship between these names was not well-received (or, rather, for the most part ignored!), I would like to briefly revisit the possibility here.

The forms Gewissei and Gewissae are attested (see Richard Coates "On some controversy surrounding Gewissae / Gewissei, Cerdic and Ceawlin").

The later Welsh forms Iwys or Iwis for the Gewissae would appear to derive from the Anglo-Saxon form of this word. Simon Rodway has confirmed for me that "Iwys is the Welsh form of Gewissae (Armes Prydein, ed. Ifor Williams, English version by Rachel Bromwich (Cardiff, 1972), pp. 49-50)."

Alfred is king of the "giuoys", i.e. Gewissae, in Welsh Annal entry AD 900. Asser says in his LIFE OF ALFRED: "Cerdic, who was the son of Elesa, who was the son of Geuuis, from whom the Britons name all that nation Geguuis [Gewissae]."

Iusay (variant Usai) has not been successfully etymologized by the Celtic linguists. Recently, I sent queries to several, all of whom were forced to admit that they could not come up with an acceptable derivation. I myself have tried everything I could think of, including Classical and Biblical names. This attempt ended in failure. Although there are some forms of Biblical names as recorded in Irish texts (like Usai), the initial /I-/ of Iusay prohibits us from identifying such with the Welsh name. A Ius- might suggest a Roman name like Justus, but then we cannot account for the ending of Iusay/Usai.

Of course, it is possible Iusay and Usai are corrupt or that they represent some Welsh mangling of an Irish name. Neither I nor the language experts have been able to find such an Irish analog. This is not to say it does not exist, merely that we have been unable to find it.

All of which brings me back to this:

I have shown in previous research that the Anglo-Saxon Chronicle's Cerdic is Ceredig son of Cunedda, that the same source's Cynric is Cunorix son of Cunedda (as Maquicoline) and that Ceawlin, supposed son or successor of Cynric is, in fact, Cunedda (Maquicoline). Sisam and Dumville have aptly proven that Elesa (= the metathesis Esla) is a borrowing from the Bernician pedigree. Omitting Elesa, then, permits us to see Gewis, eponym of the Gewissei/Gewissae, as the immediate ancestor of Cerdic/Ceredig. As the genealogy in the ASC in the main runs backwards, it may be that Gewis/Gewissae/Gewissei is properly the son of Ceredig.

If so, we might be able to account for Iusay after all. It is well known that the /G-/ of Gewis or Gewissei/Gewissae came to be pronounced as a /Y-/. This is what accounts for the Welsh forms beginning in /I-/. /W/ and /U/ regularly substitute for each other, especially when going from Welsh to Latin (cf. gwyn and guin). If the terminal diphthong in Iusay/Usai represents the /-ei/-ae/ of Gewissae/Gewissei, then we need only allow for a lost medial small vowel /-i-/. Iusay would then be a Welsh form of not Gewis, but of the group designation Gewissae/Gewissei.

I feel this is a rather elegant solution to the problem posed by the name Iusay. Additional support of this idea has come from the following top Celticists:

From Professor Oliver Padel Honorary Research Fellow in the Department of Anglo-Saxon, Norse and Celtic, University of Cambridge -

"In fact I think your suggestion is not only ingenious but also quite convincing. The only difficult bit, I suppose, is how a tribal name came to be thought of as an individual personal name.

The I- for OE Ge- is fine, of course; as for its loss (Iu- becoming U-), one might think of the wider Welsh loss of I- in words beginning Iu-, such that original iudd (`lord') became udd (I'm using Modern Welsh spellings for clarity), and personal names containing that word as an element did likewise. (You will find details in Jackson's Language & History in Early Britain -- sorry I haven't got it to hand)."

From Dr Ben Guy, Research Associate, Latin Lives of the Welsh Saints Project, Department of Anglo-Saxon, Norse and Celtic, University of Cambridge -

"Your email was forwarded to me by Professor Russell, because I specialise in early Welsh genealogies (I completed a PhD on the subject last year). I'm happy to help if I'm able.

I think you're right that no etymology has been proposed for 'Iusay/Usai' before. What you propose is certainly an intriguing suggestion, but I think that you may encounter a couple of difficulties with it. Firstly, as you point out below, there appears to be one too few minims in Iusay for it to equate to Gewisse/Iwys. Welsh forms of Gewisse, of which the best known is in Armes Prydein Vawr, always appear as Iwis or Iwys (compare the examples listed in GPC online). There are also earlier forms that point to the same thing: 'Giuoys' in Annales Cambriae A, s.a. 899, and Asser's 'Geguuis'. But as you suggest, this is not an insurmountable problem - though the loss would be more readily explained on a palaeographical rather than phonological level. The greater problem is the '-ay/-ai' ending. Comparable endings appear in the English forms because they survive in Latinate contexts - chiefly Bede's nominative plural form 'Geuissae' and a genitive plural 'Gewisorum' (implying a Latin nom. pl. 'Gewisi') in some Anglo-Saxon charters (as mentioned in the Keynes-Lapidge Asser book, p. 229). I don't think that that kind of ending would be expected in an OE context, and it certainly wouldn't in Welsh - GPC takes Iwys as a plural or collective noun whose ending has been influenced by the plural noun ending -wys (< Lat. -enses) found in words like 'Gwennwys'. So in other words, for your proposed derivation to work, Iusay would have to be a version of a Latinate form such as Bede's 'Geuissae'. The question of how that got into the Ceredigion genealogy in the form 'Iusay' would then be all the more complex, and wouldn't be solely a matter of linguistics! That's not to say that you're necessarily incorrect, of course, but it would require a more elaborate, and therefore more speculative, theory of derivation.

There is one further thing you might consider though, if you wanted to pursue this further: the genealogy of St Cadog. This survives in two versions, one appended to the Life of St Cadog, the other in the Jesus College 20 genealogies. The former calls Cadog's great-grandfather 'Solor', the latter 'Filur'. Both of these names were probably copied ultimately from 'Silur' or the like. Given where St Cadog's cult centre is (Llancarfan), this can't be anything other than a representation of the pre-Roman tribe 'Silures', who were resident in that area. But the form 'Silur' is not the result of regular linguistic development from the 1st century AD; it is a form taken at a later stage from a Latin text, with the '-res' ending lopped off. This might help you envisage the kind of process that might have led to a Latinate form such as Bede's 'Geuissae' being included in the Ceredigion pedigree, but one has to make rather more leaps to get there!"

From Professor Doctor P.C.H. Schrijver, Department of Languages, Literature and Communication - Celtic, Institute for Cultural Inquiry, University of Utrecht -

"Linguistically, the first thing that comes to mind regarding the initial alternation Usai /Iusay is the pair OW iud, MW udd 'lord' < *iüdd. So OW word-initial j- disappears in front of ü (= MW u). As to your assumption that Iusay may be connected to Gewissae if there is a rule that states that medial -i- is lost, I can tell you that there is indeed such a rule: *wi > ü in non-final syllables (as in *wikanti: > MW ugeint, see my Studies in British Celtic Historical Phonology 159-60). This generates the ü that we need in order to later get rid of the initial j. The only remaining problem is connecting OE Ge- /je/ with OW j-. Barring that, I would say, yes, what you suggest is possible. That still leaves the origin and etymology of the name in the dark (the reconstruction leads to something like *iwissai- or *g/jewissai-), but first things first."

From Professor Doctor Stefan Zimmer, Department of Celtic, University of Bonn -

"Spontaneaously, your idea of interpreting "Iusay" as a W form of OE Gewisse seems quite attractive. One must, of course, check meticulously the palaeographic details. As I am, alas, not a palaeograher myself, I cannot say more. I see no "LINGUISTIC" problems."

From Professor Patrick Sims-Williams, Department of Welsh and Celtic Studies, The University of Wales, Aberystwyth -

"I suppose Ius- is the older form and became Us- like Iustic in Culhwch which becomes Usic. Forms of Gewissae are noted by Williams/Bromwich Armes Prydein pp. xv-xvi. One Welsh form is Iwys, which rhymes as I-wys, and as the diphthong wy can become w, you could get I-ws- which could be written Ius- in Old Welsh and then add -ae from Latin which almost gets you to Iusay."

From Professor Doctor P.C.H. Schrijver, Department of Languages, Literature and Communication, Celtic, Institute for Cultural Inquiry, University of Utrecht -

From Professor Doctor Stefan Zimmer, Department of Celtic, University of Bonn -

From Professor Patrick Sims-Williams, Department of Welsh and Celtic Studies, The University of Wales, Aberystwyth -

Elafius, Elessa

The identification of Elafius, an important man mentioned during the second mission of St. Germanus to Britain, with Elessa, father of Cerdic of Wessex, has long remained in doubt. Superficially, Elafius seems to be the Latin form of Greek Elafios, from elafos, “stag, hart”. However, I believe there is good reason for upholding the Elafius-Elessa identification.

As written, Elafius is a Latin name derived ultimately from Greek elaphos, ‘hind, stag.’ A son of Ceredig son of Cunedda is named Hyddwn, from Welsh hydd, ‘stag, hart.’ He was the grandfather of St. Teilo of the stags. It is possible, then, that Elesa is not from Aloc/Alusa of other Saxon royal genealogies, but is a corruption of Elafius, itself a Latin translation for Hyddwn. I've conclusively shown that the Gewissei pedigree runs backwards in the English sources, and so the Elesa presented to us as the father of Cerdic of Wessex would actually be the latter's son.

Ceredig Son of Cunedda and Cerdic of Wessex

According to the Anglo-Saxon Chronicle (or ‘ASC’), the most famous of Cunedda’s sons was Cerdic (Welsh Ceredig). This man founded the kingdom of Ceredigion in west Wales.

Cerdic of Wessex appears on the scene in 495 A.D. His death is marked in 534.

These are the battles of Cerdic as recorded in the ASC (interposed battles by other Saxon chief-tains are in brackets):

495 - Certicesora (Cerdic and Cynric arrive in Britain)

[Bieda of Bedenham, Maegla, Port of Ports-mouth]

Certicesford - Natanleod or Nazanleog killed

[Stuf, Wihtgar - Certicesora]

Cerdicesford - Cerdic and Cynric take the king-dom of the West Saxons

Cerdicesford or Cerdicesleag

Wihtgarasburh

537 - Cerdic dies, Cynric takes the kingship, Isle of Wight given to Stuf (of Stubbington near Port and opposite Wight) and Wihtgar

Ceredicesora or "Cerdic's shore" has been thought to be the Ower near Calshot. This is a very good possibility for a landing place. Howev-er, the Ower further north by Southampton must be considered a leading contender, as it is quite close to some of the other battles.

Natanleod or Nazanleog is Netley Marsh in Hampshire.

Certicesford/Cerdicesford is modern Charford on the Avon.

Cerdicesleag contains -leag, a word which origi-nally designated a wood or a woodland, and only later came to mean a place that had been cleared of trees and converted into a clearing or meadow. Cerdicesleag or "Cerdic's wood" I would tentative-ly identify with Lee (leag)-on-the-Solent. I pick this location because of the mention of Stuf (= Stub/b) both before and after the Cerdicesleag battle. Lee-on-the-Solent is just a little bit west of Stubbington, the settlement of the descend-ents of Stuf/Stubb. It is also just across the Solent from the Isle of Wight, which was given to both Wihtgar and Stuf.

Wihtgar as a personage is an eponym for the Isle of Wight. Wihtgarasburh is, then, the Fort of Wihtgar. Wihtgara is properly Wihtwara, 'people of Wight', the name of the tribal hidage. Wihtgarasburh is traditionally situated at Caris-brooke.

Two Unknown Battle Sites in the 'Cerdic Theater'

Aelle of Sussex's sons are all being used as eponyms in the ASC to map out the early boundaries of Sussex. Cissa is for Chichester (although see also Cissbury Ring not far north of Lancing), Wlencing for Lancing. The unknown Mearcredesburna or 'Mearcred's Burn' I make out to be the Adur's tributary, the Rother, once called the Limen, a British river-name. Limen was probably mistakenly connected to Latin limen, 'a limit', from limes, and so the name Mearcred (containing mearc, 'boundary') was substituted. It marks the eastern boundary of early Sussex. Aderitum or Pevensey, another Aelle conquest, is a bit further east.

Cymen of Cymensora is no different. The name is from Welsh cyman, meaning a host or army. It here stands for Wittering, the place of Wihthere’s people. Wihthere, rather than being ‘creature/being-host’, as Wittering sits right across The Solent from Wight, is probably ‘Wight-host.’ In other words, the here, ‘host’ = cyman, ‘host.’ One word is English, the other Welsh. As the meaning of the Welsh word was forgotten, it was personified into a son of the equally fictitious Port of Portsmouth.

A Note on the Grave of Vortigern

Robert Vermaat (see his award-winning Website Vortigern Studies) has elsewhere written about the Stanzas of the Graves which places Gwrtheyrn Gwrtheneu's ‘doubtful’ grave at an unlocated Ystyuacheu. To date, all efforts to locate this grave have failed. What follows is an attempt to both find this elusive burial site and to explore its significance in the broader context of just who Vortigern might have been.

The placename Ystyuacheu should be rendered in a more modern fashion as something like Styfacheu/Stafacheu/Stofacheu. Unfortunately, such a form is also not locatable. It is true, however, that MS. copyists frequently confused the letters u and n. This being so, I propose that perhaps the first -u- of Ystyuacheu might, in fact, have originally been an –n-. This would yield a Stynacheu/Stanacheu/Stonacheu.

In all of Wales, I found only one such Stynacheu/Stanacheu/Stonacheu site which made sense both etymologically and in terms of what we know of Vortigern. This is Stanage on the Teme River in Radnorshire. Stanage is from either OE stan + ecg, ‘stone edge’, or the ME stan + egge, with the same meaning.

The difference in the ending of Stanage and a hypothetical Stanageu/Stanagau may be accounted for in the same manner as the process by which the Cymracized English placename Stange became Stangau. These are the forms for Stange/Stangau:

STANGAU at SN761261 on map sheet SN72 900ft Parish of Llandeusant. 1948 OS 1:25000 First series.

STANGE 1840 OS 1" First edition ( David & Charles reprint).

STANGAU 1891 OS6" First edition.

STANGE 1805-12 OS2" Original Drawing Map.

RHIW alias STANGE 1808 Blaen Sawdde Estate Map. West Glamorgan Archives, Swansea.

As it turns out, Stange is a dialectal variant of stangau. The writers of some documents quite commonly 'corrected' the local pronunciation by inserting the standard form.

Stanage has a Welsh equivalent ‘y Fron-faen’ (modern spelling, ‘the stone breast/steep hillside’). The English and Welsh versions seem to have existed side-by-side among the relative speech-communities for centuries, but the Welsh version seems to have disappeared around the end of the 16th century, as the Welsh language became extinct in the area.

At Stanage there is an early medieval motte, a medieval mound and bailey castle and medieval Stanage Castle. In my opinion, the motte was thought to represent the ‘castle’ in which Vortigern burned to death on the Teifi (according to Nennius). The question then becomes; ‘Which tradition is correct’ - that which places Vortigern on the Teifi or that which places him on the Teme?

To begin with, the similarity in the two river names could easily have led to confusion. The oldest forms of the Teme are of the type _Temede_ (which appears in Welsh as _Tefaidd_ (though the name now appears to be lost), while the earliest spellings for _Teifi_ are _Te(i)bi_, with an earlier form in Ptolemy (2nd century CE) _Touegobios_ or _Touerobios. Teifi and Teme are etymologically related; cf. Thames, in Welsh Tafwys. In terms of etymology, Teme and Teifi are linked because -f- is the result of lenition of earlier Welsh/British -m-.

The truly interesting thing about the Teme site is that it is, as already mentioned, located in Radnorshire. That portion of Radnorshire between the Wye and the Ithon rivers, which lies west of Stanage, was once known as the cantref of Gwrtheyrnion, i.e. the land of Vortigern. Stanage lies in Maelienydd cantref, which bordered on Gwrtheyrnion.

It would seem, therefore, that the original story had Vortigern dying on the Teme near Gwrtheyrnion. This site was transplanted to the Teifi to take advantage of the St. Garmon place-name found there. Geoffrey of Monmouth later moved the site once again, situating it at Ganarew near his hometown.

The ‘doubtful’ quality assigned to Vortigern’s grave at Stange is appropriate, as this king was almost certainly buried at Viroconium.

CHAPTER 2

ARTHUR’S ANCESTRY:

RESTORING A GENEALOGY

The True Identity of Uther Pendragon

Later in this book I will treat of the Arthurian battles of the Welsh ‘Pa Gur.’ For now I wish to concentrate on that the poem's placement of the 'vythneint' ("predatory birds" or “wrathful ones”, a metaphor for warriors) at Elei (valley of the River Ely in Glamorgan). Why might this placement be so important?

Because one of them, viz. Mabon son of Modron, is called the servant (guas/gwas) of Uther Pendragon.

There is an implied sense in this passage that if Mabon is of Elei and he is the servant of Uther, then might we not infer that Elei belonged to Uther? In other words, Mabon was the servant of Uther at Elei. If this is not what is meant, then it if difficult to explain why it was felt necessary to tell us that Mabon was Uther's servant in this particular context.

So what, exactly, is in Elei? Well, there is the impressive Caerau hillfort (http://www.coflein.gov.uk/en/site/94517/details/caerau-hillfortcaerau-campcaerau-ely), an oppidum of the Silures tribe. Unfortunately, there is no evidence that this site continued to be inhabited after the Roman invasion and consolidation of the region.

However, just south of Caerau and only a couple of kilometers from the Ely River is the Dinas Powys hillfort. The story here is completely different. We have ample evidence for early medieval use (http://www.coflein.gov.uk/en/site/301314/details/dinas-powys-fort-previously-cwm-george-or-cwrt-yr-ala-camp). It "may have been established as late as the Roman period."

Now, of course, we have to be careful here. Mabon is a god - not a human servant. And the 'Pa Gur' is replete with battles against monsters and supernatural entities. These contests range all over Britain, and so are quite fabulous in nature. What weight, therefore, can we place on an oblique reference pertaining to Uther's residing at Dinas Powys in Ely?

There is nothing in and of itself that is marvelous about Dinas Powys - other than the fact that it was occupied during the Arthurian period. The real question becomes "Why would the poet have placed Uther there?"

One possible reason might be that the Cadoxton River runs at the eastern foot of the hillfort. This stream bears the name of St. Cadog, whose late 11th century (?) VITA includes a story about Arthur. Still, Arthur is nowhere in the Life said to be related to Cadog.

It would appear the only thing that can be said about Uther at Dinas Powys is that at the time of the composing of the 'Pa Gur' a tradition may have existed which knew of this hillfort as the fortress of Arthur's father. If this tradition is historically sound, then the 'Pa Gur' preserves the only extant notice of Uther's geographical whereabouts.

For a nice summary of the excavations at and theories regarding Dinas Powys, see

http://repository.uwtsd.ac.uk/378/1/Helen%20Anderson.pdf

In the VITA of St. Illtud , once the saint has visited the his cousin Arthur’s court he travels to King Pawl Penychen. Penychen was that part of Glamorgan between the Thaw and the Taff, wherein is situated Dinas Powys.

Illtud the soldier, b. c. 470 A.D. - eventually to become the famous St. Illtud - ended up becoming Pawl’s military commander. There is the strong possibility that Pawl's fort was Dinas Powys in Penychen.

A surprise awaited me when I looked into St. Illtud in more detail. He was referred to as 'farchog', "knight", and filwr, 'soldier/warrior', as well as 'princeps militie' (militum princeps) and magister militum (https://freidok.uni-freiburg.de/dnb/download/6837).

Chapter 2 of the VITA SANCTI ILTUTI calls the living Illtud "miles magnificus." We may compare this with Uther [Pen]dragon (where dragon has the usual metaphorical meaning of “warrior”):

W. uthr

fearful, dreadful, awful, terrible, tremendous, mighty, overbearing, cruel; wonderful, wondrous, astonishing, excellent.

L. magnific.us

splendid/excellent/sumptuous/magnificent/stately; noble/eminent; proud/boastful

But, even better, at the end of his VITA we are told of a 'terrible soldier/warrior', who though of a heavenly nature and left unnamed, is plainly the returning spirit of the old soldier Illtud, out to retrieve his stolen bell:

From the Life of St. Illtud:

In meridiana autem hora, dum rex quiesceret in tentorio campestri in planicie affixo, diuidereturque maxima predatio, uisum est regi quod quidam terribilis miles suum pectus lancea perforasset, atque post perforationem nemini uisum. [...] Timoratus imperauit sacrilego exercitui reddere Deo et sanctissimo Iltuto totam predationem, promittens deinceps emendationem, atque in honore eiusdem sancti edificauit templum, et seruentibus in templo concessit in quo stetit territorium. Hec emendation tamen profuit suo spiritui, recessit enim ab hoc seculo .ix.no die propter nequitie uindictam.

At the hour of noon, while the king rested in a field-tent put up on a plain, and the immense booty was being divided, it seemed to the king that some terrible soldier had pierced his breast with a spear, and after the piercing he was seen of none. […] Full of dread he bade his sacrilegious army to restore to God and to the most holy Illtud all the plunder, promising thereafter amendment, and in honour of the same saint he built a church, and to those serving in the church he granted territory in which it stood. This amendment, however, profited his spirit, for he departed from this life on the ninth day as punishment for his wickedness (VI, §25).

We may thus place Illtud as the terrible warrior and magister militum at Dinas Powys - the same place Uther [Pen]dragon, the Terrible [Chief-]warrior, is placed in the 'Pa Gur.' As Rachel Bromwich in a note to her TRIADS has rendered Pendragon as either ‘chief warrior’ or ‘chief of warriors’, concocting a terrible chief-warrior from the Illtud titles would have been a simple matter.

And why was Mabon in the Ely Valley? Well, some map work helped with that one.

I noticed that Llantwit ('Llan-illtud') Fardre was in Taff Vale

(https://www.visionofbritain.org.uk/place/6637). Why is this significant? Because Llanfabon, the Church of Mabon, was also in that river valley (https://www.visionofbritain.org.uk/place/6636).

Furthermore, Llantrisant, where Illtud was also worshipped, was on the Ely River only a few kilometers southwest of Llantwit Fardre. The Ely and Taff rivers share an estuary near the Dinas Powys hillfort, where Illtud was master of soldiers.

But perhaps better than these sites is Llanfabon-y-fro, the earlier name of Gileston. This town is located exactly between Dinas Powis and the Caerau oppidum of the Silures and Llantwit Major. The ‘fro’ of Llanfabon-y-fro is for the bro of Bro Morgannwg, the Vale of Glamorgan. The two forts and Gileston are all in the Vale of Glamorgan.

Llantrithyd, another Illtud church, is just a little north of the famous Campus Elleti of Ambrosius (see https://mistshadows.blogspot.com/2020/06/the-red-herring-of-llanilid-uther.html).

What that all means is that we cannot simply say that Mabon as a predatory bird (or war fury) of Elai/Ely was merely placed there because it was the scene of a battle or a monster fight in the 'Pa Gur.' Mabon (the so-called saint of Llanfabon and Llanfabon-y-fro could be a Christianized version of the earlier sun god, or perhaps a 'divine son' nickname for the Christ child) belongs to the Taff and Ely valleys, where Illtud churches are also located.

More importantly, in CULHWCH AND OLWEN we learn that Mabon was kept as a prisoner at Caer Gloyw, i.e. Glevum or Gloucester. The River Ely was in the Welsh kingdom of Glywysing, a place named for Glywys, an eponym for Glevensis, ‘a man of Glevum’. There can be little doubt, then, that Mabon is properly placed in the valley of the River Ely.

The question we must naturally ask now is whether Uther Pendragon/Illtud was actually Arthur’s father, or was this imposed upon the pedigree by Geoffrey of Monmouth or a predecessor? We know that there is no record of any children being born to Illtud, and he put his wife away when he became a religious.

As it happens, there is sufficient evidence to support the claim that Illtud was not, in fact, Arthur’s father. Instead, he appears to have been substituted for another personage with whom he was improperly identified.

Sawyl Benisel of the North

In our sources we appear to have a strange confusion over Uther/Illtud and a certain Sawyl, the Welsh spelling of the name Samuel. This last seems to be the Sawyl who ruled from the vicinity of the old Ribchester Roman fort, i.e. Sawyl Benisel (see below).

Before I seek to justify why I am opting for Sawyl rather than Illtud as Arthur's father, I wish to enumerate the 'Samuel Occurrences' as they relate to Illtud.

1) The best emendation for kawyl in the 'Marwnat Vthyr Pen' is Sawyl. For details on this, see https://mistshadows.blogspot.com/2023/01/how-expert-analysis-of-uther-pendragon.html. Uther is referring to his transformation by God in the midst of battle, and his becoming either 'like [the Biblical] Samuel' or a 'second Samuel.'

2) As we have seen above, in the PA GUR, Uther/Illtud is said to be a servant of Mabon at the Ely River in Glamorgan. Illtud served as master of the soldiers for a chieftain in the Ely Valley. The link with Mabon was established because there is an Illtud church in the valley of the Taff not far from a church of Mabon in the same valley. But we also have a giant Mabon whose castle was in the parish of Llansawel (Church of Sawyl) in Carmarthenshire. Maponus was worshipped at Ribchester in the Roman period and was specifically a Northern deity. The majority of the dedications made to Maponus were by soldiers belonging to the Sixth Legion at York.

3) In the Life of Illtud, the saint is chief over the soldiers who are punished for their transgression by Cadoc. But in Cadoc's Vita, the chief of these same soldiers is Sawyl Benuchel (a term later used for Benisel). This Sawyl is said to live by Llancarfan, and I have theorized that because the place was said to derive part of its name from 'stags', this man must be Sawyl of Ribchester, whose brother was Cerwyd[d], eponym for the Carvetii or 'stag-people', a Roman period tribe whose kingdom had bordered on the region ruled by Sawyl.

4) Geoffrey of Monmouth has Eldad (= Illtud) evoke Samuel and his actions when the former deals with the captured Saxon Hengist.

5) While reading up on St. Illtud's churches in Wales, I stumbled upon an interesting ancient cross carved with the names of saints, including those of Illtud and Samuel (Welsh Sawyl). The details of this cross I have pasted below.

It is thought the St. Samuel in question is the one of Llansawel in Carmarthenshire.

https://www.ucl.ac.uk/archaeology/cisp/database/

https://www.ucl.ac.uk/archaeology/cisp/database/picpage/ltwit_2_1.html

RCAHMW (1976): +SAM | SON | POSUIT | HANCC[.] | UCEM+ || [--] | PROA | NMIAEI | US+

Expansion:

+SAMSON POSUIT HANC C[R]UCEM+ [ORATE] PRO ANMIA EIUS+

Translation:

Samson (PN) erected this cross...for his soul.

CAHMW (1976): +ILT[U] | [..] || SAM | SON || RE | GIS || SAM | UEL | + || EBI | SAR | +

Expansion:

+ ILT[UTI] SAMSON REGIS SAMUEL + EBISAR +

Translation:

Of Illtud (PN), of Samson (PN) the king, Samuel (PN), Ebisar (PN).

While the following is true (passage from The Lives of the British Saints: The Saints of Wales and Cornwall and Such Irish Saints as Have Dedications in Britain, Volume 2, Sabine Baring-Gould, John Fisher, For the honourable Society of cymmrodorion, by C. J. Clark, 1908) -

"The genealogies of the Welsh saints give his [St. Asaph's] father's name as Sawyl Benuchel, the son of Pabo Post Prydain, but in the very early genealogies in Harleian MS. 3859 he appears as "Samuil pennisel map Pappo post priten," with epithet 'Penisel" (of the low head) or "Penuchel" (of the high head). The later genealogists confounded him with the Glamorganshire chieftain (dux), Sawyl Benuchel..."

- it also appears the St. Sawel of Llansawel in Carmarthanshire was confused with Sawyl of the North.

I say this because Llansawel is on the Afon Marlais. The southern Llangadog is by Allt Cunedda, where Sawyl Benuchel was supposedly swallowed up by the earth, but the northern Llangadog is very near another Afon Marlais. Thus it is likely that it was St. Sawyl, possibly the Samuel of the Illtud Cross, who was confused with Sawyl of the North. In other words, there never was a separate southern dux named Sawyl who came to figure in the Life of St. Cadog.

Sawyl Benisel of the North is known to have had strong family connections in Cumbria. What I hadn't thought about is the fact that Cunedda, in an elegy poem, is also placed in Cumbria:

https://mistshadows.blogspot.com/2022/01/the-propaganda-of-cunedda-as-romanized.html

This is interesting, in that Sawyl Benuchel's grave has traditional been placed on Allt Cunedda

not far from Cadog's monastery.

While Cunedda's presence in the North of Britain is based upon spurious tradition, the possible association of a Sawyl with ties to Cumbria and a Cunedda who supposedly fought in Cumbria would suggest once again that the southern Sawyl is merely a relocated version of the nothern Sawyl.

Putting all of that aside for a moment, I can remind my reader that while Illtud is not said to have had any children, and to have put away his wife when he became a religious, it is also true that both Uther and Sawyl are said to have had sons named Madog. When treating below of Eliwlad, son of Madog son of Uther, I will be able to show fairly convincingly that the name might well be patterned after the Ailithir title given to Madog son of Sawyl in the Irish sources. Ailithir means 'pilgrim', and another Welsh didactic poem that closely resembles the form of "The Dialogue of Arthur and the Eagle" has as its primary character a personage called merely creiriwr, 'pilgrim.'

There are other reasons for looking very favorably upon Sawyl as Arthur's father. Firstly, we can easily account for the presence of the Artorius name at Ribchester, given that fort housed the Sarmatian veterans. I will argue in the Appendices that these soldiers would have been employed by the 2nd century prefect of the Sixth Legion in an expedition to Armorica. Such a significant exploit might well have caused the name Artorius to have been preserved at Ribchester and to have been given to a royal son in the latter part of the 5h century.

More importantly, the Arthurian battle list of the HISTORIA BRITTONUM (see the next chapter) ends up revealing a marvelous pattern once it is arranged in the North. The battles trend all along the Roman Dere Street from York to Manau Gododdin. Badon fits extremely well as Buxton (the site of Badon according to 'The Dream of Rhonabwy'), and Camboglanna/Castlesteads on Hadrian's Wall answers perfectly for Camlan. Avalon, if we must have it as something tangible, could be the Aballava/Avalana Roman fort not far west of Camboglanna at Burgh By Sands. We even have a Dea Latis or 'Goddess of the Lake' at Burgh By Sands. The PA GUR poem fixes the Tryfrywd/Tribruit battle in the North, and so to try and relocate it to the South means we are going against the only tradition we possess for that battle.

A popular notion has it that the Sarmatians were particularly fond of the draco standard. If this were so, it may be that Sawyl was more easily conflated with Illtud because of the use of the term 'dragon' in the Pendragon epithet. However, my own research has proven that, in fact, there is not a shred of evidence that supports the claim that the Sarmatians had their own draco standard. For my study on this topic, see the following link:

https://mistshadows.blogspot.com/2024/02/there-is-absolutely-no-evidence-for.html.)

If the Sarmatians at Ribchester or their descendents did revere a draco standard, it would have been a Roman one – and all Roman soldiers held that particular standard in special esteem. It also holds true that while dragon in Welsh heroic poetry meant ‘warrior’, if we carry that metaphor back far enough the dragon itself may have its origin in the Roman draco standard.

Still, everything that looks so good if we go with Sawyl as Arthur's father vanishes if we instead opt to stick with Illtud as Uther. We cannot account for the preservation of the Artorius name, and trying to find the battles in the South is acutely problematic. We are forced to accept the Afon Gamlan next to Llanelltyd in Gwynedd for Camlan, and we have no Avalon by name. The other battles can be found, more or less, although the pattern thus produced is unsatisfactory and, indeed, highly improbable from a strategic sense. Our best bet is to try and identify them with a selection of the Gewissei battles as those are presented in the ANGLO-SAXON CHRONICLE and we can do that only by resorting to linguistic contortions.

At this point in my research, I am completely content with choosing Sawyl Benisel as Arthur's real father. It seems pretty clear to me (and, hopefully, will become so to my readers later on in this book) that Sawyl was wrongly identified with Illtud/Uther Pendragon, and that the latter was ultimately substituted for Sawyl. This may - and probably was - the work of Geoffrey of Monmouth, who also created Gorlois from the gorlassar epithet Uther gives himself. Illtud had family ties in Ercyng, which is where Geoffrey lived and wrote his HISTORY OF THE KINGS OF BRITAIN.

Eliwlad the Spirit in Eagle Form

When it comes to pre-Galfridian members of Arthur’s family, there are only two worth mentioning: Eliwlad and his father, Madog, son of Uther.

The two names are found in the Welsh didactic poem “The Dialogue of Arthur and the Eagle (Ymddiddan arthur a'r eryr).” Madog is rare as an early Welsh name. Eliwlad has proven to resist being etymologized.

Eliwlad/t is the earliest form of the name. P.C. Bartram (in his A CLASSICAL WELSH DICTIONARY) says "The older form of the name appears as Eliwlat, Eliwlad."

Eliwlad - Pen. 127, pp. 96-7 Pedwar Marchog ar Hugain Llys Arthur

Eliwlat, Eliwlad - Dialogue of Arthur and Eagle, Blodeugerdd No. 30 (extant in 14^th century or later manuscripts, but may be as early as the 12^th century, according to Patrick Sims-Williams in “The Early Welsh Arthurian Poems,” THE ARTHUR OF THE WELSH, ed. by R. Bromwich, A.O.H. Jarman and B. F. Roberts). Some late MSS. have Liwlod, where the E- has dropped. Dr. Simon Rodway said this is “perfectly predictable.”

Eliwlat - Bleddyn Fardd (13^th century)

cywydd poetry - Eliwlad, Liwlad (this last in Lewis Glyn Cothi of the 15^th century, showing the loss of the initial E-)

Only in Tudur Penllyn and Tudor Aled (both of the 15th century) do we have Liwlod. It has become customary to render the name as Eliwlod, as Bromwich does, for instance, for the Eliwlad of the late Triad in Pen. 127. But there is no reason whatsoever to assume that -lod should be preferred over -lad. I can’t see applying the lectio difficilior principle here, especially as the rare -lod spelling was not used universally even by the poets of the 15^th century. Furthermore, if we do opt for -lod being the exemplar, no workable etymology for the name is forthcoming. Granted, personal names can be opaque. But in addition, according to Dr. Simon Rodway, -aw- is not confused for -a- in the MSS. and the later -lod spelling demands an original -l(l)awd. Thus the -lod spelling appears to be anomalous.

I began exploring the possible derivation of Eliwlad by approaching Dr. Richard Coates with the suggestion that the original spelling may have been *Eilwlad.

Dr Coates responded:

“It looks perfectly possible to me that Eliwlad represents British *Aljowlatos 'other land'. Eliwlad/t is a plausible rendering of Eilwlad. One certainly finds occasional <e> for <ei> in MW, and metathesis is always possible. If it’s not from *aljo-, I have no idea.”

Dr. Simon Rodway added:

"<e> for /ei/ is quite common in Old Welsh, but not in Middle Welsh. Where it does occur in MW it is probably either due to scribal error or evidence for an OW exemplar which has not been correctly modernized."

I then went to the ancient Welsh MSS. themselves and found several instances in which eil was miscopied eli or vice-versa. This occurs because i and l can look very similar to each other. And such a metathesis may have been influenced by the many names spelled with El-, from Proto-Celtic *φelu-, ‘many.’ The best example in the MSS. was Eli spelled Eil in CULHWCH AND OLWEN of the Red Book of Hergest (Jesus College MS. 111, 818, 841).

Scholarly Support for Eliwlad as Deriving from Ailithir

"Your proposal to understand Eliwlad as a W 'translation' of Ir Ailithir looks quite attactive. Eli- might well stand for Ir ail(e), and tir is correctly translated as 'gwlad'. The respective range of meaning of both words is, of course, not identical.

If 'pilgrim' really is the "primary meaning" of ailithir, then this word is beyond any doubt a bahuvrîhi compound, designing somebody 'who is characterized by another [foreign] land', obviously in the sense that (s)he has visited a [remarkably] foreign land, is acquainted with it, etc.

We have to remind an alternative, however, viz. that the 'other land' referred to might be the 'Otherworld' , so that the bearer of the epithet may have been named so for assumed / desired magical qualities. Note that Rachel Bromwich, in her invaluable Trioedd Ynys Prydein (3rd ed., p. 428) has a Madawc m. Run y Kynnedvau. By the way, I trust that you have made already ample use of that magnificent book and the references found therein.

The whole story of the red Welsh Dragon (and its mischievous counterpart), including the epithet 'Uther Pendragon', may well be based on post-Roman misunderstandings of reminiscences of the Roman, originally perhaps Sarmatian, standard. But one should not overstress the Sarmatian-Alanian theory in discussing Arthurian matters. In case you read German, you may have seen what I wrote about in 'Die keltischen Wurzeln der Arthussage' (Winter: Heidelberg 2000)."

Professor Stefan Zimmer

"Irish aili- does not have a diphthong ai in the first syllable but a fronted low simple vowel [ae] (approximately as in Engl. back) followed by a palatalized -l´-. I find it quite plausible that this would have been borrowed immediately as W eli-."

Professor Doctor Peter Schrijver

“I don’t disagree with anything Zimmer or Schrijver say.”

Dr. Simon Rodway

"Given the context and manuscript tradition behind that, Eliwlad can be plausibly associated with Ailithir."

- Dr. Alexander Falileyev

“I think that -wlad cannot be anything else but gwlad 'country', and your idea that Eliwlad is a reinterpretation of Ailithir seems plausible to me. If Eliwlad developed directly from the British, we would expect *Eilwlad."

Professor Ranko Matasovic

"I see no philological objection to the proposal that the two names are probably similar in formation and may have an overlapping semantic range, though I think it's important to recognise the uncertainty as to whether either refers to one who is 'from' another land, or has travelled 'to' such a land."

Alan G. James, The Brittonic Language in the Old North

"I can see no reason to dispute any of this. I can see no reason to dispute any of this. For example, it looks perfectly possible to me that Eliwlad represents British *Aljowlatos 'other land'. Eliwlad/t is a plausible rendering of Eilwlad. One certainly finds occasional <e> for <ei> in MW, and metathesis is always possible. If it’s not from *aljo-, I have no idea.”

Professor Richard Coates

"If it is OK with Peter Schrijver, it is OK with me."

Associate Professor Emeritus Daniel F. Melia

"It is possible that Eliwlad REFLECTS ailithir, but it is no translation, but looks like a deliberately playful misinterpretation."

Professor David Stifter

“First it appears to me that you must be right in identifying gwlad as the second element. This is indeed the regular cognate of flaith in Irish, but the latter, a feminine i-stem, originally also had an abstract meaning ‘lordship, sovereignty’, and its application to a person is a secondary process in Irish (retaining the feminine gender!) for which there are several parallels, such as techt meaning not only ‘going’ but also ‘messenger’, cerd both ‘craft’ and ‘craftsman’, etc.

Your proposed adaptation of aili- to eli-, on the other hand, would have to have been purely formal, since Irish and British continue two different variants of the same word ‘other’, Ir. aile (also 'second') < *aljo- and e.g. Middle Welsh. all < *allo-. British ail, 'second', is from *aljo-.

But apart from this formal misgiving, I do admit that your derivation would make for a nice contextual fit!“

Professor Jurgen Uhlich

“Purely formally, the proposed derivation (more correctly, pace Uhlich, an "adaptation") would indeed appear to work.”

Professor Doctor Alderik Blom

“Associating the names Eliwlad ap Madog with Matóc Ailithir is ingenious and attractive.

It needs to be taken into account that Matóc Ailithir himself is a rather elusive figure, and that his epithet is only attested in some cases. The most recent presentation of the evidence known to me is Pádraig Ó Riain, Corpus Genealogiarum Sanctorum Hiberniae, where only a single manuscript mentions him on p. 112 par. 669.3, while only 2 out of 7 witnesses give him the epithet ailithir on p. 177 par. 722.76. This evidence might suggest that Matóc (Ailithir) is a relatively late figure; but without looking into the sources considerably more deeply it would be impossible to form a real judgement on the matter.

It's interesting that Eliwlad also appears a few times in later Welsh sources (Bromwich, Trioedd, 3rd ed., p. 346).

For the etymology that you propose to work, there would of course need to be a double strategy of derivation: Eli-, with its single l, would be borrowed from Irish (despite the existence of a transparent cognate in Welsh), while -wlad would substitute a Welsh word for “land” for Irish -thir (despite the existence of tir in as a common word in Welsh). It’s not clear to me why such a complex process would have been used.

I don’t think that any of these considerations refute your proposal.”

Professor John Carey

I would add that it may not be necessary to see in Eliwlad an actual translation or partial substitution for Ailithir. It may be that Eliwlad is a completely conjured name, less likely an actual given name, a son of Madog bestowed with a different name that had meaning identical to Ailithir.

The main theme of “The Dialogue of Arthur and the Eagle” has been compared to another similar poem. In the Dialogue, Arthur is instructed by the eagle in matters of the Christian faith. According to Dr. Oliver Padel, in ARTHUR IN MEDIEVAL WELSH LITERATURE”

“Barry Lewis has pointed out that a sixteenth-century dialogue between a creiriwr [crair + -iwr in the GPC] (‘pilgrim’) and Mary Magdalene of Brynbuga (the town of Usk) is remarkably similar in both form and content to the dialogue with the Eagle…”

So, in the Arthur poem, Eilwlad = Ailithir/pilgrim, and in the other example a pilgrim is also the second character.

Lewis (private communication) has elaborated somewhat on this point. I asked him “With Christian didactic poems such as 'The Dialogue of Arthur and the Eagle" and the one on St. Mary Magdalene of Brynbuga/Usk, I assume it is always true that the "instructor" is in one way or another a holy person? Either a saint or a priest, etc.?”

He replied, “So far as I know, that is the tendency, yes. I can't think of any counter-examples. The examples I know of all feature either saints, ecclesiastics or anonymous speakers. I can see no clear indications in the poem that Eliwlad is thought of as a saint. However, it appears that he has gone to heaven or at least has privileged access to information from there. That leaves the possibility open.”

How did this help with tracing Arthur’s genuine line of descent? Well, I had found in Irish sources a certain Matoc (= Welsh Madog) Ailithir son of Sawyl Benisel, a notable chieftain in North Britain first found in the Harleian genealogies. Ailithir, attested in Old Irish, is from aile, ‘other’, plus tir, ‘land’ (Professor Jurgen Uhlich, personal communication). The word meant ‘pilgrim’ in the context of Matoc, who was a saint who traveled from Britain to Ireland.

What we appeared to have here was Eliwlad/’other land’ son of Madog as a folk memory of Matoc Ailithir/’other land’. And this would mean, of course, is that Uther Pendragon, the ‘Terrible Chief-dragon’, was none other than Sawyl Benisel.

Proof That Eliwlad = St. Matoc

But one thing still bothered me: why is the ghost of Eliwlad portrayed as an eagle? For some time I got bogged down in specious connections with the god Lleu, who assumed eagle-form in death in MATH SON OF MATHONWY. Nothing seemed to help explain why St. Matoc would have been portrayed as a spectral eagle.

Until I happened to look up Matoc Ailithir's feast day in the "Martyrology of Tallagh" (https://books.google.com/books?id=--oCAAAAQAAJ&pg=PA492&lpg=PA492&dq=%22April+25%22%2B%22Matoc%22&source=bl&ots=XpVy9QuD-p&sig=ACfU3U1k35WnY4jEnWlFgdT2nwS4xIf4uw&hl=en&sa=X&ved=2ahUKEwif69PFk7vqAhUKpZ4KHQSMCJwQ6AEwAXoECAcQAQ#v=onepage&q=%22April%2025%22%2B%22Matoc%22&f=false). This fell on 25 April.

I then cross-checked for other more important saints who may have been associated with the eagle and who had the same feast day. And guess what I found?

While St. John is typically identified with the Eagle of the Tetramorph, the very important medieval Christian writer Irenaeus linked Mark with that bird. And Mark's feast day is 25 April!

Many sources discuss the significance of Mark as an eagle (see, for example, https://books.google.com/books?id=1TBFAQAAQBAJ&pg=PT1057&lpg=PT1057&dq=%22April+25%22%2B%22feast+day%22%2B%22eagle%22&source=bl&ots=afQr48Rv0j&sig=ACfU3U33tmqnuMbj9TGPrZmaOIzWUNIgpA&hl=en&sa=X&ved=2ahUKEwi2-O2alLvqAhXOpJ4KHRaVBbUQ6AEwC3oECAgQAQ#v=onepage&q=%22April%2025%22%2B%22feast%20day%22%2B%22eagle%22&f=false and http://www.thetextofthegospels.com/2017/10/mark-12-part-3-irenaeus-and-tatian.html).

But here is the actual passage from Irenaeus' "Against Heresies", 3.11.8 (http://www.thetextofthegospels.com/2017/10/mark-12-part-3-irenaeus-and-tatian.html):

"For the cherubim, too, were four-faced, and their faces were images of the dispensation of the Son of God. For, [as the Scripture] says, "The first living creature was like a lion," symbolizing His effectual working, His leadership, and royal power; the second [living creature] was like a calf, signifying [His] sacrificial and sacerdotal order; but "the third had, as it were, the face as of a man,"-an evident description of His advent as a human being; "the fourth was like a flying eagle," pointing out the gift of the Spirit hovering with His wings over the Church. And therefore the Gospels are in accord with these things, among which Christ Jesus is seated. For that according to John relates His original, effectual, and glorious generation from the Father, thus declaring, "In the beginning was the Word, and the Word was with God, and the Word was God." Also, "all things were made by Him, and without Him was nothing made." For this reason, too, is that Gospel full of all confidence, for such is His person. But that according to Luke, taking up [His] priestly character, commenced with Zacharias the priest offering sacrifice to God. For now was made ready the fatted calf, about to be immolated for the finding again of the younger son. Matthew, again, relates His generation as a man, saying, "The book of the generation of Jesus Christ, the son of David, the son of Abraham; " and also, "The birth of Jesus Christ was on this wise." This, then, is the Gospel of His humanity; for which reason it is, too, that [the character of] a humble and meek man is kept up through the whole Gospel. Mark, on the other hand, commences with [a reference to] the prophetical spirit coming down from on high to men, saying, "The beginning of the Gospel of Jesus Christ, as it is written in Esaias the prophet,"-pointing to the winged aspect of the Gospel; and on this account he made a compendious and cursory narrative, for such is the prophetical character."

We may compare this description of Mark the eagle, coming down from on high to preach the Gospel, with Eliwlad coming as an eagle to Arthur to teach Christian doctrine.

Now, Irenaeus was one of the Church Fathers. His writings were extremely influential (https://www.newadvent.org/cathen/08130b.htm). He most certainly would have been known by someone like the monkish scribe who composed "The Dialogue of Arthur and the Eagle".

While there will still be doubters, I do not think it is a coincidence that St. Matoc Ailithir/'other land', viz. "Pilgrim" and St. Mark share the same feast day, and the Eilwlad/'other land' son of Madog appears to Arthur as an eagle. This explanation makes a lot more sense in the Christian context of the dialogue poem than seeking to associate a saintly character with the pagan god Lleu.

If I'm right, then this bolsters my argument for Sawyl Benisel as the father of Arthur.

This discovery seemed very exciting to me, as some years before I had – with the assistance of leading toponymists – succeeded in placing Sawyl at Samlesbury near the Ribchester Roman fort in Lancashire.

Samlesbury or ‘Sawyl’s Fort’

For a nice history of Samlesbury, see

http://www.britishhistory.ac.uk/vch/lancs/vol6/pp303-313.

My own extensive treatment of the place-name’s etymology is as follows:

Sawyl (Samuel) Benisel ("Low-head"), another son of Pabo, is birth-dated c. 480. On the Ribble, not far south of “regio Dunutinga”, is a town called Samlesbury. The place-name expert Ekwall has Samlesbury as “Etymology obscure”, but then proposes OE sceamol, “bench”, as its first element, possibly in the topographical sense of “ledge”.

Mills follows Ekwall by saying that this place-name is probably derived from scamol plus burh (dative byrig). However, sceamol/scamol is not found in other place-names where a “ledge” is being designated. Instead, the word scelf/scielf/scylfe, “shelf of level or gently sloping ground, ledge” is used.

Watts says: “The early spellings with S-… may be due to AN [Anglo-Norman] influence and a native tendency to assimilation in the sequence of sc – s.”

The complete history of this place-name has been kindly supplied by Mr. Bruce Jackson, Lancashire County Archivist:

A D Mills: 'A Dictionary of English Place-Names'; Oxford University Press, 1991, page 284

'Samlesbury Lancs. Samelesbure 1188. Probably "stronghold near a shelf or ledge of land". Old English scamol + burh (dative byrig).'

David Mills: 'The Place Names of Lancashire'; Batsford, 1976 (reprinted

1986), page 130

'? burh on a shelf of land (OE sceamol + -es (possessive) + burh, in the

form byrig (dative)

Samerisberia 1179 (Latin)

Samelesbure 1188

Samlesbiry 1246

The original settlement was probably around the church which stands by the R. Ribble, at the foot of the 168 foot ridge to which the first element may refer. The derivation from OE sceamol, however, involves taking as base later forms of the name in 'sh-', such as Shamplesbiry 1246, which, though not uncommon, are far less frequent than forms in 's-'. If the 's-' forms are original, the etymology is less certain. There is much variation in the representation of the first element in early records - e.g. Sambisbury c.1300, Sammysburi 1524, Samsbury 1577. There is today no village around the church; the main settlement moved to the south, to SAMLESBURY BOTTOMS, (Old English botm, 'valley bottom', here referring to the valley of the R. Darwen in which the hamlet stands), where a community grew up around the cotton mill which was built there c1784.'

Eilert Ekwall: 'The Concise Oxford Dictionary of English Place-Names'; Oxford University Press, 1960, page 403

'Samlesbury La [Samerisberia 1179, Samelesbure 1188, -bur 1212, Schamelesbiry 1246, Scamelsbyry 1277. Etymology obscure. If the name originally began in Sh-, the first element may be Old English sceamol 'bench' &c. in some topographical sense such as "ledge".'

Eilert Ekwall: 'The Place-Names of Lancashire'; Manchester University Press, 1922, page 69

'Samlesbury (on the Ribble, E. of Preston): Samerisberia 1179, Samelesbure 1188, 1194, Samelesbur', Samelisbur' 1212, Samelesbiri 1238. Samelesbiry, Samelesbiri, (de Samlebir, Samlesbiry, Samplesbiry) 1246, de Samelesburi

1252, Samlisbyri 1258, Samlesbury 1267, 1311 etc., Samlisbury, Sampnelbiry, Sampnesbiry 1278, Samesbury 1276, 1278, Samlesbur' 1332, Samsbury 1577; Shamplesbiry, de Schamelesbiry, -byr 1246, Scamelesbyry, Shampelesbyri,

Shapnesbyri 1277.

The old chapel of Samlesbury stands on the S. bank of the Ribble, with Samlesbury Lower Hall some way off on the river. I take this to be the site of the original Samlesbury. The etymology is much complicated by the variety of the early spellings. The forms with S- are in the majority, but there are a good many with Sh-, and it is not easy to see why S- should have been replaced by Sh-, whereas S- for Sh- is easily explained by Norman influence. If the original form had Sh-, I would compare the following names: Shamele (hundred Kent) 1275; Shalmsford (Kent); Shamelesford 1285, Sahameleford 1275, perhaps Shamblehurst (Hants): Samelherst, Scamelherst' 1176, Schameleshurste 1316. All these may contain Old English sceamol "bench, stool," or some derivative of it.....The meaning of this word in topographical use is not clear, but very likely it may have been something like "ledge, shelf".....In this case the word might refer to a ledge on the bank of the Ribble. In reality, Samlesbury Lower Hall stands on a slight ledge (c 50 ft above sea level), which stretches as far as the church.

If the spellings in Sh- are to be disregarded the etymology is much more difficult. The first element is hardly the personal noun Samuel. It it is a personal noun, as the early forms rather suggest, it may be a derivative of the stem Sam- found in German names. This stem is not found in English names, but the related stem Som occurs in Old English Soemel and perhaps in the first element of Semington, Semley, Wilts. Burh in this name, as in Salesbury, may mean "fortified house, fort" or "manor"...’

Henry Cecil Wyld and T Oakes Hirst: 'The Place Names of Lancashire'; Constable, 1911, page 226

'Samlesbury

1178-79 in Samesberia

1187-88 de Samelesbure

1189-94 Samlisburi

1227 Samlesbiri

1228 Samlesbyr

1246 Samelesbiri

1259 Samelebir

The first element is undoubtedly the Hebrew personal noun Samuel. This does not appear to have been popular amongst the English in early times.....It is not recorded by Bjorkman [Erik Bjorkman: 'Nordische Personennamen in England'; Halle, 1910] as having been adopted by any Norseman in this country, but Rygh mentions a Norwegian place name Samuelrud ["Norske Gaardnavne Kristiana", 1897, volume ii, page 201]. In volume i the same writer records Samerud (pp 7 and 9), but says that this is possibly a Modern name.'

John Sephton: 'A Handbook of Lancashire Place-Names': Henry Young, 1913, page 23

'A parish 4 miles east of Preston. Early forms are Samerisberia, Samelesbure. First theme is the scriptural name Samuel. Ancient Teutonic names are also found from the root Sama....' .

I would suggest as a better etymology for Samlesbuy “Sawyl’s fort”. There are, for example, Sawyl place-names in Wales (Llansawel, Pistyll Sawyl, now Ffynnon Sawyl). Richard Coates, of the Department of Linguistics and English Language at The University of Sussex, says of Samlesbury as “Samuel’s Burg”:

“After a bit of extra research, it seems that all the spellings in <Sh-> and the like are from just 2 years in Lancashire assize roll entries (1246 and 1277). That makes them look more like the odd ones out and <S-> more like the norm. I'm coming round to preferring your interpretation, even though Ekwall in PN La (p. 69) simply rejects the idea it might come from "Samuel". Brittonic *_Sam(w)e:l_ (<m> here is vee with a tilde - nasalized [v]) is a good etymon for the majority of the forms, including the modern one, of course.”

Dr. Andrew Breeze of Pamplona, another noted expert on British place-names, agrees with Dr. Coates:

“I finally looked up _Samlesbury_ last night and feel sure you are right. The form surely contains the Cumbric equivalent of Welsh _Sawyl_<_Samuel_. Your explanation of this toponym in north Lancashire is thus new evidence for Celtic survival in Anglo-Saxon times.”

Brythonic place-name expert Alan James also has signed on to the idea that Samlesbury is from the name Samuel:

"Checking LHEB p415 confirms that the medial consonant in Proto/Old Welsh *Saμuil would have remained audibly nasal through the 6th – 9th centuries, and could have been adopted into OE as *Samɪl or *Samel, and that name could very well be the specific in Samlesbury.

However, we need to be clear that *(æt) Samelesbyrig is an English p-n formation, not a Celtic one, formed with anglicised *Samel, not *Saμuil, and certainly not Middle-Modern Welsh Sawyl. The fact that the specific is an anglicised version of a Brittonic-influenced latinate form of a Hebrew name is no more evidence of Celtic survival than it is of Hebrew survival. The eponymous *Samel need not have known a single word of Welsh. It only tells us that there had been some transmission of personal names from Brittonic to OE, presumably by bilinguals (probably mothers?) at some point during the period of anglicisation, and there’s plenty of evidence to support the judgement that, in what became Lancashire, that was quite a long and gradual process.

You're probably right in your suggestions that Samel could have originated as an anglicised, Old English, form of proto-Welsh *Saμuil, but as *(æt) Samelesbyrig is an English p-n formation, not a Celtic one, we can't assume that Samel of Samlesbury was, or spoke, Welsh (Brittonic, Cumbric). By the time the place was named, Samel could have become current as a personal name among monolingual Old English speakers. So Samel of Samlesbury can't be identified with any certainty with anyone named Sawyl.

As for Samlesbury being an anglicized version of a place originally named for a 5th century Sawyl, I wouldn't rule that out, but it would mean Sawyl had been adopted into OE not only a personal name but as a figure of local legend. I'm always doubtful about place-names being 'translated', it does happen, but it's not normal. But if Sawyl > Samel featured in folklore among English speakers, they might have associated the site with him and given it that name.

When it comes to the family of Sawyl being placed as you have suggested, I would agree that some kind of 'legendary mapping' went on in the central middle ages, maybe starting during the time of the Cumbrian kingdom in the 10th century and continuing through the 11th and into the 12th, where figures in local folklore and poetry were associated with particular places, though whether any of those identifications relate to what really happened half a millennium earlier is, at best, unproveable."

I did receive one dissenting opinion on my proposed etymology for Samlesbury being the traditional home of Sawyl Benisel. This came from Professor John Insley of Heidelberg, who is in charge of the English Place-Name Society's Lancashire project. Prof. Insley sent me this:

"Mats Redin, Studies on Uncompounded Personal Names in Old English (Uppsala, 1919). 143 f., links the Soemel, -il/Seomel of the genealogies, Nennius and John of Worcester with the stem of OE sōm f. ‛arbitration, agreement, reconciliation’, ġesōm ‛unanimous, peaceable, friendly’. The [o:] oft he stem syllable would have been from to [ø:] (represented by <oe> and <eo>) as a result of i-mutation induced by [i] in the diminutive suffix –ila-. I would interpret the first element of SAMLESBURY as a variant of *Sœ:mel formed with the diminutive –ul-suffix and with a different ablaut grade (zero-grade) in the stem syllable. I would render this form as *Sāmul. Cf. ON Sámr, OSax Samuko. I think this is more satisfactory than the Welsh version of the biblical Samuel."

Dr. Richard Coates, in commenting on this alternate etymology, concluded –

“I would say Redin was cautiously inclined to go for a Germanic solution where he could, and I think John inclines the same way. The name is undoubtedly difficult, as we’ve seen. For the moment I’d personally say it’s 50/50, or maybe 53/47 in favour of the Welsh.”

Personally, given all the other evidence which points strongly towards the Ribchester area for Sawyl, I think Insley’s derivation is not a good one. In fact, I think it is quite strained and, indeed, inventive. As an example of what I'm talking about, we can take a brief look at the family of Sawyl.

Pabo in the earliest, most reliable genealogy (the Harleian) is the son of Pabo son of Ceneu. Later pedigrees intrude a Arthwys and Mar, but P.C. Bartram is surely right in saying that the earlier line of descent is "more correct, being chronologically more satisfactory."

A map shows the geographical relationship of Pabo Post Prydain (of the Papcastle Roman fort in Cumbria) and his "sons" Cerwyd(d), an eponym for the Carvetii tribe, Dunot of Dentdale (see https://mistshadows.blogspot.com/2020/02/why-sawyl-benisel-is-listed-between.html) and Sawyl of Samlesbury near the Ribchester Roman fort. Note that Papcastle is on the Derwent, and the fort there was called Derventio. This matches the name of the Darwen River at Samlesbury, another Derwent.

Sawyl's son St. Asa belongs at Llanasa and Llanelwy/St. Asaph just a little to the SW of Samlesbury in Flintshire, while a church to his son Sanctan can be found at Kirksanton in southwestern Cumbria:

For more on Sanctan, see https://books.google.com/books?id=a_1fAgAAQBAJ&pg=PA47&lpg=PA47&dq=chendisil&source=bl&ots=hy_TsaWT3F&sig=ACfU3U0PeMe8M8291HhlEH9pgQ5zJssvpg&hl=en&sa=X&ved=2ahUKEwjL2tLCmebgAhUSip4KHaB1A-MQ6AEwB3oECAgQAQ#v=onepage&q=chendisil&f=false.

A daughter of Sawyl is said to have married Maelgwn of Gwynedd, a kingdom that bordered on Sawyl’s.

The headwaters of the Rivers Dent and Ribble are literally right next to each other.

Granted, in all of this we are talking about a sort of legendary mapping out of the region surrounding Sawyl’s kingdom (which probably approximated that of the ancient Setantii tribe). Pabo was not Sawyl’s real father; the name appears to be from ON papa, papi, ‘hermit.’ Neither was his son the Carvetii! But the localities selected help us determine where Sawyl belonged. And if this certainly seems to be Samlesbury.

A Top Celtic Language Specialist Weighs in on My Identification of Samlesbury as 'Sawyl's Burg'

I had long since obtained agreement from the majority of place-name experts on my proposed derivation of the Samlesbury, Lancashire place-name from the Welsh personal name Sawyl. But, I had neglected to take my case to at least one of the world's most renowned Celtic language scholars. To redress that oversight, I decided to send the following query to Prof. Dr. Peter Schrijver of Utrecht University. This man had been very generous in the past when it came to helping me with linguistics problems I encountered during my Arthurian research.

“Hi, Peter.

Please find attached an entry for a place-name found in Victor Watt's THE CAMBRIDGE DICTIONARY OF ENGLISH PLACE-NAMES.

My question to you is whether you think the first element of this place-name could come from the Welsh (or Cumbric) form of the name Sawyl, at some point in its development from the Latin form of the Biblical Samuel.

Yes, Victor Watts disagrees. But without meaning to bias you from the outset, several place-name experts agree with me that some form of Samuel through the Welsh would best account for this place-name.

I have also suggested that there may be good reasons embedded in Welsh tradition to support the idea. Sawyl Benisel of the North is said to descend from Pabo Post Prydain. Pabo is here being used as an eponym for the Papcastle Roman fort in Cumbria, which is on Derwent Water. Samlesbury and its related sites are on the Darwen in Lancashire, an identically named river. Sawyl's son Asa/Asaph is a well known saint in nearby NE Wales, while his son San[c]ton has a church in extreme southern Cumbria. All of the other relatives of Sawyl can be easily (and firmly) traced to Cumbria and immediately adjacent territories. The Irish sources give Sawyl an Irish princess as wife, one hailing from the Dal Fiatach along the coast in NE Ireland. Indications, therefore, are that Sawu ruled from what had been the Roman period tribal region of the Setantii (or Segantii).

What I haven't done thus far is ask a Welsh/Celtic specialist how we might have ended up with the early forms for Samlesbury, assuming for the sake of argument that the name Samuel is involved. I'm hoping you can please fill in that gap.

Yes, I understand all too well that you may disagree with my proposed etymology for Samlesbury, and that's fine. As always with these kinds of questions, I have my own doubts. We are dealing with a name in English territory, one that may have been subjected to influence from English, Anglo-Norman and even Norse.

Thank you very much for your help with the issue."

His response:

"That seems quite possible to me. The forms with Sc-, Sh-, Sch- are later and point to a secondary Anglo-Saxonized interpretation. An ancient (i.e. earliest medieval) British or British Latin spelling <Samuel> or <Samel-> would be expected, or a later (say 700-1000 CE) Welsh spelling <Samuil, Samuel>. Hard to say exactly when the name became Sawyl. Old Welsh was still [Savuil], but a spelling <sauuil> may well have been avoided because of double <uu>. Approximately by the 11th century (so Middle Welsh) the -v- may well have been lost. But this is very rough chronology. "

As a follow-up, I wanted to make sure and ask Schrijver what he thought of yet another idea for the etymology of Samlesbury. Via personal communication with Professor John Insley of Heidelberg, who is responsible for the Lancashire portion of the EPNS, I had the following on a proposed derivation for the place-name:

"I would interpret the first element of Samlesbury as a variant of *soe:mel formed with the diminutive -ul-suffix and with a different ablaut grade (zer-grade) in the stem syllable. I would render this form as *Samul."

Schrijver's opinion of that idea?

"It’s a concatenation of unsupported assumptions: diminutive -ula instead of the -ila of Soemel, Seomel, ablaut of the root, arbitrary etymology. Pretty desperate. Maybe he’s unaware of the fact that if he would like to hang on to the connection with OE so:m ‘agreement’ its Celtic counterpart *sa:m- (> Ir. sám ‘peace’) is much closer to the vocalism of Samlesbury…"

Sub-Roman Occupation of the Ribchester Fort?

An excavation project within the Roman fort at Ribchester has only recently been undertaken by the archaeology department of the University of Central Lancaster:

http://www.uclan.ac.uk/news/ribchester-roman-dig-bbc.php

When I wrote to Dr. Duncan Sayer, one of the directors of the dig, and asked if they had yet found any evidence for sub-Roman use of Bremetennacum Veteranorum, he replied with this exciting news:

“Yes, I believe we have identified some evidence of sub-Roman occupation within the fort at Ribchester. Certainly the abandonment date of AD370 is no longer really tenable and at this early state in the project we are reasonably convinced they have structures and workshops that relate to a later-Roman and sub Roman phase of activity.”

Why Sawyl Benisel is Listed Between Welsh Kingdoms in the Harleian Genealogies

Something odd happens in the earliest Welsh genealogies. Here, for example, is a section from the Harleian:

Dunoding

[C]uhelm map Bleydiud map Caratauc map Iouanaul map Eiciaun map Brochmail map Ebiau[n] map Popdelgu map Popgen map Isaac map Ebiau[n] map Mouric map Dinacat map Ebiau[n] map Dunaut map Cuneda.

Meirionydd

[C]inan map Brochmail map Iutnimet map Egeniud map Brocmail map Sualda map Iudris map Gueinoth map Glitnoth map Guurgint barmbtruch map Gatgulart map Meriaun [map Typiaun] map Cuneda.

[C]atguallaun liu map Guitcun map Samuil pennissel map Pappo post Priten map Ceneu map Gyl [Coyl] hen.

Rhufoniog?

[?I]mor map Moriud map Ædan map Mor map Brechiaul.

Here's the odd thing: Samuil pennissel and his 'branch' are not assigned anywhere. Yet he is sandwiched between the Welsh kingdoms of Meirionydd and Rhufoniog. Prior to those two kingdoms come Dunoding ("people of Dunawd") in Wales. When you look on the map of early Wales, you see that Dunoding falls between Meirionydd and Rhufoniog.

Now, Samuil is made the brother of a Dunod! And we know there was this regio Dunutinga somewhere in northwestern England. So what seems to have happened here in the Harleian MS. is that the two Dunodings were confused. This possibility adequately explains why Samuil/Sawyl and his pedigree were situated between Rhufoniog and Meirionydd.

The nothern Regio Dunutinga has routinely been identified with Dentdale, but the linguistics don't seem to work. As explained by Brythonic place-name expert Alan James (personal communication):

"With regard to Dent, it's depressing how the same old rubbish keeps being trotted out. In BLITON I deal with it thus: Discussion of this place-name has ... been persistently muddled by the identification of this place with the lands in regione Dunutinga granted to Ripon according to VW17, and associated in turn with the semi-legendary chieftain Dunawd (< *Dönǭd < Donātus, see Morris (1973) p. 214n4). Early forms give no support for this identification; whatever the correct etymology for Dent, it certainly has nothing to do with Dunawd. If the *regio Dunutinga was around Dent, the name in VW is very garbled. If, on the other hand, the *Dunutingas were named after any Dunawd, their regio was not Dent.

Indeed, evidently highly intelligent, expert historians who are not historical linguists seem to have a problem with the simple point, that the Dunutingas may well be the descendants of Dunod, or they could be garbled *Dinetingas, people of Dinet, Dent, but they cannot be both! Dinet can't be equated with or derived from Dunod.

Myself I think they were descendants of some Dunod, but we can't know whether he was the 'famous' one, nor can we know where their regio was."

Continuing with what he believes to be the proper etymology of Dent, he passed along this from BLITON:

"A hypothetical Brittonic cognate of OIr dind, primarily meaning ‘a sharp point’, but associated topographically with ‘a height, a strong point, a notable place’.”

And added this via correspondence:

"I think Dent is named from the fell [Dent Fell, at the headwaters of the river]. It's quite a likely *dïnned. And note the other possible examples, in Cmb, listed in BLITON, as well as the one in Abd. It seems to have been a pretty common Brittonic (and Pictish) hill-name."

James is pretty convincing here: Dentdale would not seem to be regio Dunutinga. But if not, then where is the latter place?

James concludes "somewhere in the general area of Craven, Bowland, the Howgills, Mallerstang Common is a plausible guess, but again no more than that."

If we go to the land grants given to Ripon -

https://books.google.com/books?id=k9cGAAAAYAAJ&pg=PA6&lpg=PA6&dq=%22ripon%22%2B%22donatus%22&source=bl&ots=s78lGSBmqc&sig=ACfU3U3Afx4XWddYsYaWVfDRxA_739C4pA&hl=en&ppis=_e&sa=X&ved=2ahUKEwjX593Svd7nAhXPJTQIHYdxC8gQ6AEwAHoECAYQAQ#v=onepage&q=rippel&f=false

- we find much useful information. Note that Gaedyne is Yeadon, and Caetlaevum is Catlow.

Notice also that Amounderness is intruded into later lists of these places, as is the Mersey district (Marchesiae). Alan James assures me that

"Amounderness was north of the Ribble: the Amounderness Hundred comprised the lands north of the Ribble up to the Cocker, beyond Lancaster, i.e. the Fylde (OE gefilde 'plain'), and east to the fells and the border with YWR. Samlesbury, south of the Ribble, was in the Hundred, and former Northumbrian 'shire', of Blackburn. Small portions north of the Ribble (notably Ribchester) were later annexed to Blackburn Hundred, but Amounderness never extended south of the Ribble."

But must we abandon Dentdale entirely? I don't think so.

From Melville Richards' Early Welsh Territorial Suffixes:

"DUNODING [Dunawd] a cantref which was later divided into Eifionydd and Ardudwy. Dunawd was a son of Cunedda (dunaut Cy ix 183). Cf. 1283 'Cantredo de Dinnedin qui habet duas commotas videlicet Euyonith et Hardidew' LW 155. See G s.n. Dinoding, HW i 238."

So, the *Dinetingas, people of Dinet, Dent, could well derive from an earlier Dunutinga/Dunoding, it seems to me. If we can have forms like Dinnedin for Dunoding, I don't see any reason why we can't allow Dentdale to be a Dunoding of the North.

Such an identification for the land of Dunawd, reputed brother of Sawyl, lends credence to the idea that Sawyl himself ruled from the Ribble.

Still, we have a problem, best explained by Alan James:

"... even if Dinet was a form of the personal name, it can't also be a place name. Either they were the people of a place called Dinet, or of a guy called Dinet. If a place were named after a guy named Dinet, it wouldn't just be Dinet."

However, we could propose that something like this happened: we begin with a Cumbric name like afon (if the river) or dol (if the dale) or whatever, followed by Dunawd. This would be standard place-naming formula. You can find Caer Dunod, etc., in Wales, for example.

We then have the English adding to Dunawd their own -inga as a designation for the people who live in this Dunawd place or who descend from the said Dunawd.

This would not change the name of the place. It would merely designate the inhabitants of the place in the terms of the land grant.

I asked Alan James if this would this satisfy the requirements of Dent being from Dunawd. He responded:

"I suppose if the place were something like *Cair Dunod, and the English speakers picked up the idea that Dunod was the eponymous ancestor or hero figure of the local Britons, they might have called those people Dunodingas. But then the generic was forgotten, and the place-name became Dinet."

Sarmatians at Ribchester

Professor Roger Tomlin of Oxford imparted to me his opinion on the most likely disposition of the Sarmatian troops in Britain:

"...we must assume that this very large number of men – and their horses – were sent across the Continent to Britain. The logistics would have been enormous. And they are attested in only one place, Ribchester. The evidence here is stronger than you imply: not only the two (lost) tombstones of the ala Sarmatarum, but the mid-3rd century dedication to Apollo (RIB 583) for the numerus equitum and the Notitia's evidence (Occ. 40.54) of the cuneus. Whatever form the unit took, it was there a long time, and it is surprising (as you say) that no other Sarmatian unit has left a trace. Did Dio exaggerate the numbers – or was there 'wastage' en route?

These are possibilities, but I don't think 'illiterate barbarians' is an explanation. The same consideration would apply to any auxiliary unit: inscriptions would most likely be erected by Roman officers, as indeed these are, or would be (say) tombstones which an illiterate simply paid someone else to write.

As for the question of how they were organised ... The rarity of double-strength 'milliary' alae is surely due to the expense and difficulty of co-ordinating even one thousand cavalry. I can't believe that 5,500 Sarmatians were organised as a single unit. I would expect them to be organised as 500-strong alae, or drafted piecemeal into existing alae and part-mounted cohorts to fill gaps. They would then disappear without trace within a generation. But it looks as if an ala was organised at Ribchester. The analogy of such units as the Suebi at Lanchester (RIB 1074) – who, after all, are also illiterate 'barbarians' from outside the Empire – would suggest that the Sarmatians were mostly organised into smaller contingents – numeri – under Roman officers. Perhaps, like the Suebi, they then took their names from where they were stationed.

This is an unsolved problem. I can only suggest various possibilities which would have reduced their 'presence'. No one knows the answer."

Ribchester ‘along the line of the wall’ and the Sarmatian Cuneus

Peter Verburgh (personal correspondence) has reminded me that the Ribchester fort of the Sarmatian veterans, which I've recently decided upon as the home and base of the famous 6th century Arthur, is listed among the 'per lineum ualli' in the NOTITIA DIGNITATUM. This is an important classification of Ribchester, as Arthur operated militarily at the Wall just south of Corbridge and at Castlesteads.

I'm quoting here the relevant portion of the N.D.:

Occ. XL:

DUX BRITANNIARUM*

17. Sub dispositione uiri spectabilis ducis Britanniarum:

18. Praefectus legionis sextae

19. Praefectus equitum Dalmatarum, Praesidio

20. Praefectus equitum Crispianorum, Dano

21. Praefectus equitum catafractariorum, Morbio

22. Praefectus numeri barcariorum Tigrisiensium, Arbeia

23. Praefectus numeri Nerviorum Dictensium, Dicti

24. Praefectus numeri uigilum, Concangios

25. Praefectus numeri exploratorum, Lauatres

26. Praefectus numeri directorum, Uerteris

27. Praefectus numeri defensorum, Braboniaco

28. Praefectus numeri Solensium, Maglone

29. Praefectus numeri Pacensium, Magis

30. Praefectus numeri Longovicanorum, Longouico

31. Praefectus numeri superuenientium Petueriensium, Deruentione

32. Item per lineam ualli:

33. Tribunus cohortis quartae Lingonum, Segeduno

34. Tribunus cohortis primae Cornouiorum, Ponte Aeli

35. Praefectus alae primae Asturum, Conderco

36. Tribunus cohortis primae Frixagorum, Uindobala

37. Praefectus alae Sabinianae, Hunno

38. Praefectus alae secundae Asturum, Cilurno

39. Tribunus cohortis primae Batauorum, Procolitia

40. Tribunus cohortis primae Tungrorum, Borcouicio

41. Tribunus cohortis quartae Gallorum, Uindolana

42. Tribunus cohortis primae Asturum, Aesica

43. Tribunus cohortis secundae Dalmatarum, Magnis

44. Tribunus cohortis primae Aeliae Dacorum, Amboglanna

45. Praefectus alae Petrianae, Uxelodunum or Petrianis

46. 'Luguuallii'

47. Praefectus numeri Maurorum Aurelianorum, Aballaba

48. Tribunus cohortis secundae Lingonum, Congauata

49. Tribunus cohortis primae Hispaniorum, Axeloduno

50. Tribunus cohortis secundae Thracum, Gabrosenti

51. Tribunus cohortus primae Aeliae classicae, Tunnocelo

52. Tribunus cohortis primae Morinorum, Glannibanta

53. Tribunus cohortis tertiae Neruiorum, Alione

54. Cuneus Sarmatarum, Bremetenraco [Ribchester]

55. Praefectus alae primae Herculeae, Olenaco

56. Tribunus cohortis sextae Neruiorum, Uirosido

From http://www.vortigernstudies.org.uk/artsou/notitialist.htm

Observant readers will notice two things immediately. One, the unit garrisoning Ribchester is referred to as a cuneus. This distinction is not found for any other unit in the N.D.

Two, no commander is listed for this cuneus. Again, this is the only instance in the entire text of the N.D. in which a unit commander is omitted.

I will explore the reason for this exception and apparent oversight below. But first, here are the relevant definitions associated with a cuneus:

cuneus (m. pl. cunei)

1. Literally ‘a wedge’, a type of formation in that shape used in battle and countered with the forceps (Fest. s.v.; Veg., DRM 3.19; Aul. Gell. 10.9.1). 2. An irregular auxiliary unit, often mounted (ND Or. XXXIX.1–9. RIB 882; 1594). See also caput porcinum and forceps [Cowan 2007]

caput porci(num) (n. pl. capita porcorum)

Literally ‘pig’s head’, an informal term referring to the battlefield tactic known as the cuneus. Veg., DRM 3.19; Amm. 17.13.9. [Cowan 2007]

forceps (f. pl. forcipes)

Literally ‘pincers’, a battlefield formation used to counter the cuneus by mirroring its shape (centre held back and flanks advanced). Aul. Gell. 10.9.1. See also forfex [Cowan 2007]

forfex (f. pl. forfices)

Literally ‘shears’ or ‘claw’, a battlefield formation used to counter the cuneus by mirroring its shape (centre held back and flanks advanced). Fest. s.v.; Veg., DRM 3.19. See also forceps [Cowan 2007]

From https://perlineamvalli.wordpress.com/2018/05/24/the-roman-army-a-to-z-cuneus/

A good description of cunei can be found at:

http://lukeuedasarson.com/Cunei.html

The Sarmatian cuneus would have been composed of somewhere between 200-300 heavy shock troops, so-called "cataphracts", a term borrowed from the Greek and meaning (from the Lewis and Short Dictionary at PERSEUS):

"I. mailed, in mail (of soldiers and their horses), Sall. ap. Non. p. 556, 16 sq. (id. H. 4, 57 Dietsch).—As subst.: cătă-phracti , ōrum, m., mailed soldiers, Sisenn. ap. Non. 1. 1.: “loricatos, quos cataphractos vocant,” Liv. 35, 48, 3; 37, 40, 5 al.; Prop. 3 (4), 12, 12; Serv. ad Verg. A. 11, 770; Front. Princ. Hist. 5, p. 247 Nieb.; Inscr. Orell. 804."

It was the job of such heavily armored units to smash through enemy lines with a tightly grouped cavalry charge. To quote from Richmond's study of the Sarmatians at Ribchester ("The Sarmatae: Bremetennacum Veteranorum and the Regio Bremetennacensis", THE JOURNAL OF ROMAN STUDIES, Vol. 35, Parts 1 and 2, 1945, pp. 15-29):

"They [the Sarmatians] were remarkably equipped for war, men and horses being clothed from head to foot in mail after the manner of Persian catafractarii, and carried a heavy spear requiring the use of both hands... It was without doubt these heavy cataphracts whom the Imperial staff, ever anxious to develop their cavalry arm, desired to acquire; for Roman experience had shown that, while the Sarmatian cavalry were at a disadvantage when off their gaurd or hampered by snow or mud, no ranks could withstand their charge in the battle-line. In the Roman service many defects could be cured or remedied by drill and discipline, and it became the view of later Roman military experts that cataphracts rendered the best service both in breaking a battle-line and in pursuing broken infantry."

But why for ONLY the Cuneus Sarmatarum is no commander listed? Well, the answer to that question may be answered here:

"Interestingly, the officer recorded in RIB 583 as commanding the unit is a centurion deputised from Legio VI victrix, so it may be the case that the lack of a (formal?) commanding officer recorded in the Notitia is not an inadvertent scribal omission, but may reflect a more-or-less permanent state of affairs in actuality."

From http://lukeuedasarson.com/NDDuxBritCunSarm.html

The inscription the author of that Website is alluding to is to be found at https://romaninscriptionsofbritain.org/inscriptions/583. It reads:

"To the holy god Apollo Maponus for the welfare of our Lord (the Emperor) and of Gordian's Own Unit of Sarmatian cavalry of Bremetennacum Aelius Antoninus, centurion of the Sixth Legion Victrix, from Melitene, acting-commander and prefect, fulfilled his vow willingly, deservedly. Dedicated 31 August in the consulship of the Emperor Our Lord Gordian for the second time and of Ponpeianus."

Now, if in the late period (that recorded by the Notitia Dignitatum) a lower-level officer of the Sixth Legion at York was designated as the commander of the Cuneus Sarmatarum, we can postulate an even closer connection between a 6th century Arthur at Ribchester with the memory of the prefect and dux Lucius Artorius Castus from York.

Finally, if the cuneus of Sarmatian cataphracts were commanded by an officer detached from the Sixth Legion at York, this may have been because this allowed the unit to be used anywhere at anytime - in short, whenever an elite "shock' heavy cavalry was needed to break a battle-line. It goes without saying that the Sixth Legion would have been responsible for determining where and for what reason this force was sent into action.

However, Dr. Benet Salway adds (via personal correspondence):

"As for the cuneus of Sarmatae in Britain, I don’t think you can build anything on the lack of mention of commander. It seems to be consistent scribal practice across the ND (east and west) not to list commanders for cunei (see http://www.intratext.com/IXT/LAT0212/Y.HTM), even in lists, such as this one, which give commanders for other types of units.

I am not sure what this means for the command structure of cunei."

The following was sent to me by Professor Roger Tomlin:

"In Dietrich Hoffmann's great book on the Notitia and the late-Roman army, the author analyses the evolution of cunei from larger cavalry units. It strikes me there isn't much epigraphic evidence of what their commanders were called: I suppose praepositi. For what it's worth, the centurion Aelius Antoninus at Ribchester is said to be praepositus, which you might press to mean acting-commander at that date; and I expect you could find centurions who are acting-commanders of part-mounted cohorts, if not cavalry alae.

I took a quick look at Michael Speidel's Denkmäler, and there are centurions active there, even though the equites singulares were entirely cavalry. And what about No. 27, an altar dedicated by Ulpius Marcellus ex decurione factus (centurio)? I think we should be cautious about thinking cavalry and infantry careers were entirely distinct. After all, every equestrian prefect of an ala in the early Empire had previously commanded an infantry cohort.

Arthur and the Irish

One of the biggest and as yet unsolved problems for Arthurian researchers has been this unsettling fact: all Arthurs subsequent to the famous 6^th century one belong to Irish-descended dynasties in Britain.

How do we account for this?

Firstly, let me say that St. Illtud/Uther Pendragon did not have an Irish wife.

This is not the case with Sawyl Benisel. For Sawyl’s wife was none other than Dechtire, daughter of Muiredach Muinderg of the Ulaid. Assuming she was Arthur’s real mother, our hero would himself be half-Irish.

It would be interesting, in this context, to see where the Irish Dal Fiatach king Muiredach Muinderg had his chief court.

Fortunately, the Irish place-name and national monument folks have done their homework. The case has been made for two rulings centers for Muiredach, both in Co. Down: Drumnabreeze and Dromorebrague. I have posted below links and excerpts from the relevant Websites.

***

http://www.placenamesni.org/resultdetails.php?entry=11215

Drumnabreeze, County Down

Show the Map

Origin

perhaps Ir. Dromainn Brís ‘ridge of (the) high ground’

Background

7km E of Lurgan

par: Magheralin bar: Iveagh Lower, Upper Half

Drumnabreeze was part of the O’Lavery district of Moira in 1609 (CPR Jas I 395a), and although it was let to Stephen Haven of Dromore by 1631 (Inq. Ult. (Down) §16 Car. I), Laughlin Roe O’Lawry was still resident in 1663 (Sub. Roll Down 273). It is on rising ground east of the Lagan valley, with a hill in the centre 297 feet high, from which the view ‘ranges considerably over the flat ground adjoining the Lagan’ (OSNB). The first element seems to be one of the words for ‘ridge’, droim or its derivatives dromainn or droimne. John O’Donovan suggested that the final element was brí ‘upland’ (gen. plural breg later brígh). As noted by the editor Margaret Dobbs, a name like Droim na mBrígh invites comparison with Druimne Breg ‘ridge of the high ground’, the location of Uachtar, the earliest chief site of Dál Fiatach in east Ulster (Descendants Ir xiii 324, 336, 338). However, there is no archaeological evidence for an important historical settlement in this area, and Dromorebrague in the parish of Aghaderg is a more likely location.

If the element brí is accepted, it is possible that the final [z] of the modern pronunciation is due to the addition of English plural -s, which is usually pronounced as [z] after a vowel. There are forms without -s in the Census of 1659 and the Down Subsidy Roll (1663). A plural -s has been added to this element in English spelling in other place-names, for example Mac

Muiris na mBrígh,‘MacMorris of Brees’ Co. Mayo (Misc. Ir. Annals 1406). Dean Mooney mentioned Brees Castle from Caisleán na mBrí, Brees in Clanmorris, Mayo, referred to as (is) na Brighibh hi ccloinn Muiris in 1595 AD (AFM vii 1988). Another possibility is that [z] represents the genitive ending of bríoghas, an unattested collective form of brí (cf. Ó Máille 1989-90). This would be bríos in modern spelling.

***

NOTE: There once existed a fort at Drumnabreeze. I found this information here:

https://www.jstor.org/stable/20566485?seq=1

The argument in favor of Dromorebrague is unsupportable, in my opinion, as one must opt for a place-name with an intrusive element (mor, 'big, great'). Drumnabreeze is, on the other hand, quite perfect for our needs.

Still, for the sake of full disclosure, the reader will find below the relevant entries on Dromorebrague and its hillfort.

***

http://www.placenamesni.org/resultdetails.php?entry=11800

Dromorebrague, County Down

Show the Map

Origin

Ir. Droim Mór Breá ‘big ridge of the high ground’

Background

Dromorebrague must be the townland of Shankill district called in the early 17th century Ballydromore, although it is now known locally just as Brague. There is a hill-fort at 489 feet near the north-east boundary of the parish (ASCD 148). The townland is high ground and Droim Mór ‘big ridge’ would need no explanation as the basic name of the place. The two place-names in this parish including the element brague have been understood as containing bréige, the genitive of bréag ‘a lie’, and thus having the sense of an adjective ‘false’ (Glancy 1956, 78). The term buachaill bréige ‘false lad’ was often used for a standing stone which might look like a person when seen from far off (Joyce ii 435; Ó Ceallaigh 1952-3(b) 36-7). The -brague element has also been studied by Dean Mooney. He thought that Dromorebrague derived from Droim Mór Bréagtha ‘big ridge for playing games’, where -brague represented the genitive of bréagadh, ‘deceiving’ and thence ‘beguiling’, used in this case of a hill where people once assembled to play games (Mooney 1956(a), 26-7) (Mooney MS 11). However, this would strain the meaning of bréagadh, and bréagtha would be unlikely to be anglicized brague. It is possible that -brague could represent the genitive singular or plural breagh (standard breá) of brí ‘height, plateau’, with final -gh delenited to [g] in the modern form. Dromorebrague might then be part of the district Droim Breá or Dromanna Breá ‘ridge (or ridges) of the height (or heights)’, which was the earliest settlement of Dál Fiatach in east Ulster.

https://apps.communities-ni.gov.uk/nismr-public/Details.aspx?MonID=7762

On top of a hill with excellent views. The site is impressive, although some 52m of the N extent have been removed, where the bank & ditch would have been situated. The site, 71m E-W diam., stands proud of the field surface & is surrounded by an arc of waterlogged ditch running round the outside of the bank, which varies in height. At WSW, the bank is 1.25m above the interior, 5.4m wide & 2.35m above the ditch, which is 1.4m wide & 1m deep. The interior gradually slopes to N. See SM7 for further details. An archaeological evaluation was carried out on a proposed development site NE of the hillfort. Two test trenches were excavated across the development area. No artefacts or features of archaeological significance were uncovered [SOC 20/09/04].

SMR Number DOW 034:032 view on map

Edited Type: HILLFORT

Specific Type Specific Period

Townland: DROMOREBRAGUE

Council: ARMAGH CITY BANBRIDGE AND CRAIGAVON

County: DOW

Grid Ref: J1344041630

Protection: Scheduled

Parish: AGHADERG

Barony: IVEAGH UPR;UPR HALF

Town:

General Type: HILLFORT

Condition: SUBSTANTIAL REMAINS (Vast majority definable)

General Periods:

Submit

IRON AGE

PREHISTORIC

https://apps.communities-ni.gov.uk/nismr-public/docs/DOW/DOW_034/DOW_034_032/Public/SM7-DOW-034-032.pdf

https://apps.communities-ni.gov.uk/nismr-public/docs/DOW/DOW_034/DOW_034_032/Public/SM7-DOW-034-032A.pdf

https://apps.communities-ni.gov.uk/nismr-public/docs/DOW/DOW_034/DOW_034_032/Public/SM7-DOW-034-032B.pdf

The Trace from Arthur, son of Sawyl, to Artur son of Conaing son of Aedan

While re-reading Bannerman (STUDIES IN THE HISTORY OF DALRIADA), I chanced to bother to consult the footnotes. The text on p. 89 read:

"... it is said that [Maith] Gemma [from Irish mathgamain, 'a bear'], the saint's mother, was a daughter of Aedan and a niece of a British king which implies that Aedan's wife was of British extraction."

And the footnote:

"This is probably the Maithgemm of Monad [the Dalriadan capital], daughter of Aedan... It seems she married a certain Cairell, grandson of Muredach Munderg, of the Dal Fiatach."

What this means (for those who don't see it immediately!), is that Muiredach marries a daughter to Sawyl Benisel, and then, later, a grandson to Aedan's daughter. A son Arthur of Sawyl and the Irish princess would be roughly contemporary with Gemma and Cairell. The Dalriadan Arthur would have been named after the famous British one via obvious family connection.

I will note that there is a problem in the Irish sources regarding the Dalriadan Arthur's father. He is called both Conaing (an Irish version of the English word cyning, 'king') and Aedan, with Conaing being said to be a son of Aedan. Bannerman discusses this conflict as follows:

"There remains the problem of whether Artur was Adan's son or grandson. The combined evidence of Adomnan and AT is strong. However, it should be remembered that the original compilation of the Senchus, if it is to be accepted as an historically authenticated document, should be dated to c. 650, some fifty years earlier than Adomnan or the Scottish annals underlying AT. Furthermore, Aedan was in his seventies when he died and it would be strange indeed, considering that he had seven sons, if, towards the end of his life, none of his grandsons was of an age to take a part in the political activities of his time. Artur, grandson of Aedan, might easily have been twenty years of age, or more, by 590. It is at least possible that Adamnan and the compiler of the Iona Chronicles, when referring to those of Aedan's descendents who died before him, assumed that they were all his sons, especially as some were so in fact."

Allowing, for the sake of argument, Conaing as the Dalriadan Arthur's father, we can draw out the following comparative genealogies for the two Arthurs:

Sawyl-Dechtire

Arthur Conaing Gemma-Cairell

Artur

To me, this is incredibly convincing. And, indeed, it provides the one piece of real evidence I was missing that would allow me to feel confident about Sawyl as Uther.

Arthur Son of Bicoir and the Slaying of Mongan Son of Fiachna

Long ago, I demonstrated that the Briton Arthur son of Bicoir was, in fact, Arthur son of Petuir of Dyfed.

https://mistshadows.blogspot.com/2016/09/bicoir-father-of-artuir-and-beccurus.html

https://mistshadows.blogspot.com/2016/12/old-notes-on-arthur-son-of-bicoir-as.html

What I had not fully explored is the reason why Arthur son of Petr/Pedr of Dyfed was linked to Kintyre in the Irish sources.

I think this is pretty obvious: the ancient and, presumably, Dark Age capital of Dyfed was Arberth in Penbro. Penbro, like Irish Kintyre, means Land's End. Pen and Kin are cognates, and Welsh bro is found in its Irish form as OIr. mruig, MIr. bruig.

We have god reason for thinking the Dark Age Dyfed center was at or very near Arberth because of the survival of the Voteporigas stone:

https://www.genuki.org.uk/big/wal/CMN/Castell-Dwyran

"CASTEL-DAUYRAN (CASTELL-DWYRAN or DYRAM,) a chapelry, partly in the parish of KILMAENLLWYD, lower division of the hundred of DERRLYS, county of CARMARTHEN, partly in the hundred of DUNGLEDDY, county of PEMBROKE, SOUTH WALES, 4 1/2 miles (N.E.) from Narberth

https://www.ucl.ac.uk/archaeology/cisp/database/

Rhys/1896, 108--110, records that the stone had originally stood at Castell Dwyran...

Voteporigas was the great-grandfather of Arthur of Dyfed. The latter is thought (see Bartrum) to have been born c. 560. Mongan son of Fiachna was supposedly killed by Arthur son of Pedr in 625. The stone (elsewhere referred to as a dragon stone or piece of pitchstone, which was restricted geologically to the Isle of Arran next to Kintyre) Arthur uses to kill Mongan may be a folk reflection of the name Pedr or Petrus.

Two questions need to be answered. First, if Arthur son of Pedr has been wrongly placed in Kintyre, how did this error come about? And, two, how did the name Arthur end up in Dyfed among the Deisi-descended royal dynasty?

It is possible, I suppose, that the name Arthur was confused with the name of the location of Mongan's grave:

U625.2

Áedán son of Cumuscach and Colmán son of Comgallán migrate to the Lord; and Rónán son of Tuathal, king of Ind Airthir, and Mongán, son of Fiachna of Lurga, die.

1] Remarkable are the four over whom it has closed without recall,

2] The earth of Cluain Airthir churchyard today:

3] Cormac the Handsome,

4] And Illann son of Fiachu.

1] The other two—

2] Many territories do service to them—

3] Are Mongán son of Fiachna of Lurga,

4] And Rónán son of Tuathal.

https://celt.ucc.ie/published/T100001A/text195.html

Cluain Airthir has been identified:

c. airthir

7 bps. of, Ll. 374, I. 110 b 1; ¶ 7 sts. of, Lb. 24; ¶ Fionntan of, Ai. 150 b; ¶ now Magheracloone, c. Mon., Ch. 78, Au. i. 94, Mi., Cri.; ¶ Cailcú of, Mt. 35, Ll. 363; ¶ = Caelchú ó Lúi Airthir, Fg. 182; ¶ Mongan mac Fiachnae, lord of C. Airrthir, Hb. 64

https://research.ucc.ie/doi/locus/C

Airthir (airther) means "eastern."

But this seems a bit far-fetched.

Instead, I think we must look to the second question: why did Pedr of Dyfed name his son Arthur?

I have recently shown that the name Arthur apparently originated with Sawyl Benisel of Ribchester, and that it found its way into the Dalriadan royal house via Dal Fiatach connections:

http://mistshadows.blogspot.com/2024/07/a-startling-discovery-brings-back-sawyl.html

Now Mongan belonged to the Dal nAraide in NE Ireland. To the north of the Dal nAraide were the Dal Riata, who founded Dalriada in Scotland. To the south of Dal Araide were the Dal Fiatach. These tribal groups were variously allied with each other or fighting against each other. The best recent account of the interactions of these kingdoms during the floruit of Mongan comes from pp. 4-5 of John Bannerman's STUDIES OF THE HISTORY OF DALRIADA.

The poem on the death of Mongan tells us the warriors who kill him were in Kintyre. Or, at least, they came from Kintyre:

T627.6

Mongan son of Fiachna Lurgan, stricken with a stone by Artur son of Bicoir Britone died. Whence Bec Boirche said:

Cold is the wind over Islay;

There are warriors in Cantyre,

They will commit a cruel deed therefor,

They will kill Mongan son of Fiachna.

https://celt.ucc.ie/published/T100002A/index.html

If Mongan's killers really originated from Kintyre, then they were Dalriadans - that much is clear. Yet this scenario is impossible to square with what appears to be a steady alliance between the Dal nAraide and the Dal Riata (as treated of by Bannerman).

But what if Kintyre in the Irish Annal is an error for Penbro? What if Pedr of Dyfed had married a Dal Fiatach princess, just as Sawyl Benisel had? And the name Arthur had come from the Dal Fiatach into the Dyfed royal house?

The year dates for the death of Mongan and the death of his father Fiachna mac Baetain fall within a year of each other. One finds, when going over the annals, that sometimes the same event is entered for different years. Suppose Mongan died with his father at Leithat Midind, and Arthur son of Pedr was on the side of the Dal Fiatach king Fiachna mac Demmain? An Arthur from Penbro could easily have been confused for the various Dalriadan Arthurs (Arthur son of Aedan or Conaing and Arthur grandfather of Feradach).

This solution to the Arthur son of Bicoir problem is particularly elegant in that it both accounts for how a Dyfed king could have been involved in warfare in the North as well as accounting for how the Arthur name came to be present in southwest Wales. Such a solution must, of course, remain purely speculative as we lack any genealogical information regarding Pedr of Dyfed's wife and queen.

Sawyl Benisel, Father of Arthur, and the ‘Irish Connection’

I have just pointed out that Sawyl’s wife, according to Irish sources, was a princess, none other than the daughter of Muiredach Muinderg of the Dal Fiatach of Ulster. But I did not bother to delve more deeply into possible Ulaid influences present in the Arthurian legends. I would like here to remedy that deficiency.

My readers will notice first that on the map above, the Dal Fiatach were neighbors of the Dal nAraide. Well, Fiachna son of Baetan, whose son Mongan is sired upon Caintigern by Manannan mac Lir in a story that perfectly parallels the begetting of Arthur, was of the Dal nAraide. The Dal nAraide, in turn, were neighbors of the Dal Riata who settled in Scotland and who named one of their princes Arthur. Arthur son of 'Bicoir' (= Petuir of Dyfed) is said to have slain Mongan off the island of Islay, a part of Dalriada.

But this is not the only fascinating connection between Arthur and Ulster.

Fergus Mac Roich, whose sword Caladbolg is the prototype for Arthur's Caledfwlch, was an Ulster king. [See https://mistshadows.blogspot.com/2020/07/king-arthurs-sword-excalibur-lightning.html and https://mistshadows.blogspot.com/2019/09/the-sword-in-anvil-and-sword-in.html.]

Findabair, the prototype for Arthur's wife Guinevere, was of Connacht, the enemy of Ulster in THE TAIN. [Most Arthurian scholars have failed to observe that in Geoffrey of Monmouth's pseudo-history, Arthur must marry Guinevere before he can conquer Ireland. This is at least literary evidence that Findabair was acknowledged as an Irish Goddess of Sovereignty, a role she most certainly played in the Irish sources.]

I have elsewhere pointed out that Sawyl's Irish wife bore the same name (Deichtine) as the mother of CuChulainn. Lugh was CuChulainn's father, and this god came to play a huge role in Arthurian story, first as Geoffrey of Monmouth's 'Lucius Hiber[n]us', and then as Lugh/Llwch Lamhcalad/Llawcaled or Lancelot of the Lake. Although the case for the Setantii of Lancashire being etymologically related to CuChulainn's original name of Setanta has been challenged (see Appendix II below), the marked resemblance of the British tribal name and Irish personal name is intriguing. They may well have been connected by medieval writers who were less well-informed about Celtic linguistics.

Although Arthur's 'Avalon' is certainly the Aballava Roman fort at Burgh-By-Sands in Cumbria, we have record of Aedan son of Gabran (variously father or grandfather of the Dalriadan Arthur) fighting Baetan, father of Fiachna father of Mongan, on the Isle of Man. As Mananann was linked with the Isle of Man, it is often identifed with his Emain Ablach. I have elsewhere shown that the goddesses of Geoffrey of Monmouth's Avalon are Irish in origin.

Given all of the above, I feel that an identification of Sawyl as Arthur's father is considerably strengthened. Indeed, I can find no other possible candidate for Arthur's father that will better explain the Irish associations so prevalent in the story of our great British hero. It makes sense that Ulster traditions would have infiltrated the court of Sawyl, given his marriage to an Ulster princess. These traditions then found their way through the usual channels of transmission to Geoffrey of Monmouth.

Cadguallaun Liu Grandson of Sawyl Benisel = Caedualla Rex Brettonum?

In 2004, Alex Woolfe published an article in NORTHERN HISTORY that may have considerable bearing on my theory that the legendary Arthur came from a royal family based at Ribchester:

https://www.academia.edu/313147/Caedualla_rex_Brettonum_and_the_passing_of_the_old_north

In this piece, Dr. Woolfe very plausibly suggests that a great chieftain of the 7th century named Cadwallon, usually assigned to the Gwynedd dynasty, should instead be attached to the family of Sawyl Benisel:

"It remains now to identify an alternative and better placed candidate for Bede's Caedualla rex Brettonum. What is sought is a northcountryman, a neighbour to both Deira and Bernicia, who was appropriately placed to be over-king of both and who had afloruit in the early to mid seventh century. Happily a candidate readily presents himself. Harley pedigree 19 reads:

Catguallaun liu [1] map Guitcun map Samuil penissel map Pappa post Priten map Ceneu map Gyl hen"

I've read his entire paper - more than once now. And it seems to me that if he is right, this would reinforce my own idea that Arthur was a son of Sawyl. How? Well, only because it would be proof that Sawyl's family (and, presumably, kingdom) was extremely powerful in the North during the Dark Ages. And, that by being so, it would have been fertile ground for the production of a significant hero prior to Sawyl's grandson.

The case could even be made that Arthur's earlier victories against the Saxons in the North, including that of Badon, had set the stage for the later military successes of Cadwallon.

[1]

Dr. Woolfe argues, for instance, that the Ynys Glannauc (Enislannach in Gerald of Wales) battle of Cadwallon was a relocated event. I agree. In my opinion the proper location is the Llan Lleennawc, "Lleenog's enclosure", probably here a fort or fortified settlement and not, as later, a church. This place is found mentioned in the Book of Taliesin and incorporates the name of the father of Gwallog of Elmet. The Kingdom of Elmet lay directly to the east of the old tribal territory of the Setantii, which was ruled over by Sawyl.

XI.

GWALLAWC

[BT 29]

"Ef differth aduwyn llan lleennawc"

On the identification of Heavenfield, where Cadwallon met his death, see https://mistshadows.blogspot.com/2020/05/where-exactly-was-hefenfeldheavenfield.html

CHAPTER 3

THE BATTLES OF ARTHUR

The First Battle: The Mouth of the River Glein

It has long been recognized that there are only two extant Glen rivers which conform philologically to ‘Glein’ and which could have been subject to Saxon attack from the Continent in the 5th-6th centuries CE, the Age of Arthur.

These are the Glen of Lincolnshire and the Glen tributary of the Till in Northumberland.

The Glen of Lincolnshire has no distinctive features or strategic fortifications which would make it of any value to an invading force. On the other hand, the Northumberland Glen is hard by the Yeavering Bell hill-fort, which prior to becoming a Saxon stronghold was the British Gefrin. Gefrin is from the Welsh word gafr ‘goat’ or a compound containing gafr plus Welsh bryn (mutated fryn), for ‘Goat-hill’. I would remind the reader, however, of a Gaulish god conflated with

Mercury called Gebrinius. It is possible that Gefrin represents a British counterpart of this divine name.

The Yeavering Bell hillfort is 12.8 acres in size and encloses the two summits and the saddle between of a hill that rises to a height of 1181 ft above sea level. There is a single stone rampart

13 ft wide, with entrances midway along the north and south sides, and a third on the northeast.

At the east and west ends are small, crescent-shaped annexes, the latter with an entrance at its mid-point. The centre of the fort was the site of about 130 circular huts. The eastern summit is ringed by a trench which held a wooden palisade nearly 164 ft in diameter. Archaeologists do not know whether there is any relationship between the hillfort and the Anglo- Saxon royal town of Ad Gefrin (‘at Gefrin’) that succeeded it at the foot of the hill.

Other hill-forts abound in the region: Wooler, Kyloe Hills, Dod Law forts at Doddington, the Old Bewick hill fort and the Ros Castle fort and settlement between Chillingham and Hepburn. And, of course, the Roman road known as the Devil’s Causeway, a branch off of Dere Street, passes only a couple of miles to the east of the mouth of the Glen.

Scholars who argue in favor of the Lincolnshire or ‘Lindsey’ Glen do so primarily because the following battle, that of the Dubglas, is put in a Linnuis region by the HB. Linnuis, as we will see, is wrongly thought to represent the later regional name Lindsey.

An actual battle at the mouth of the Lindsey or Lincolnshire Glen is scarcely possible, unless it were a battle of reconquest by Arthur and not a successful defensive engagement. This is because we have archaeological evidence for Saxon cemetaries well north, west and south of the Lindsey Glen as early as c. 475 CE.

The Next Four Battles: The River Dubglas in

the Linnuis Region

Philologists have long recognized that Old Welsh

Linnuis must derive from Br.-Lat. *Lindensis, *Lindenses, or *Lindensia, and the identification

with Lindsey works fine on purely linguistic grounds. Lindsey, of course, was the early English name for what we now think of as Lincolnshire.

The root of Lindensis is British *lindo-, ‘pool, lake’, now represented by Welsh llyn, ‘pond, lake’. The Roman name for the town of Lincoln –

Lindum – is from the same root. The ‘pool’ or ‘lake’ in question is believed to have been on the Witham River near the town.

The problem is that there is no Dubglas or ‘Black Stream’ (variants Douglas, Dawlish, Dowlish, Divelish, Devil’s Brook, Dalch, Dulais, Dulas, etc.) in Lindsey. This has caused other place-name experts to situate the Dubglas battle either near Ptolemy’s Lindum of Loch Lomond in

Scotland or near Ilchester in Somerset, the Roman period Lindinis, as there are Dubglas rivers in both places. We might even look to the Douglas River in Lancashire, not far west of the Roman Ribchester fort. Unfortunately, none of these candidates is satisfactory, because Arthur would not have been fighting Saxons at these locations in the time period we are considering.

A site which has been overlooked, and which is an excellent candidate for Arthur’s Dubglas, is the Devil’s Water hard by the Hadrian’s Wall fort of Corbridge, which has upon it a place called Linnels. Almost a century ago it was proposed that Linnels was from an unrecorded personal name. But modern place-name expert Richard Coates, upon looking at Linnels on the Ordnance Survey map, observed the remarkable double elbow in the Devil’s Water with a lake nearby and concluded that Linnels was from a British *lindo-ol:in, "lake-elbow".

In http://www.historiabrittonum.net/wp-content/uploads/2018/09/The-Arthurian-battle-list-of-the-Historia-Brittonum.pdf, Keith J. Fitzpatrick-Matthews argues against this identification, saying

“August Hunt (2005), quoting personal correspondence with Richard Coates, concludes that it survives as Linnels (>*lindolīn) on the Devil’s Water in Northumberland (Hunt 2012, 98): while both names share the element *lindo-, they are emphatically not the same name; nor can linnuis derive from *lindolīnenses, which would be necessary if the identification of the name were correct.”

The fact is that if a lake were present, there could once have been residents on that lake. The Lake-elbow name might then well be secondary. And, indeed, river courses change, sometimes fairly radically over time. A widened portion of a river or a large pool/lake may have developed bends, leading to the creation of the Lake-elbow place-name. Thus 'Lake-elbow' follows naturally from the obvious fact that a lake existed.

But we can go further with this. On maps, there is literally a Linnel Lake, as well as several river-bend or 'oxbow' lakes present. I have checked the earliest maps, and these lakes are all natural. They are not reservoirs or water-filled gravel pits.

It was once thought that the Devil’s Water stemmed from a Dilston Norman family, the D’Eivilles. But going by the earliest spelling of the Devil’s Water (Divelis c. 1230) leads recent authorities to state uncategorically that this etymology is incorrect and the Devil’s Water is certainly of the Dubglas river-name type.

The Devil’s Water at Linnels is thus the only extant Dubglas river-name associated with a demonstrably Welsh lake-name that is geographically plausible as a battle site against Britons and Saxons during the period of Arthur.

Worth noting is the fact that the Roman Dere Street road at Corbridge splits immediately north of the Wall, the eastern branch or ‘Devil’s Causeway’ continuing North-NorthEast, straight to the Northumberland Glen.

As an aside, I would mention that the Battle of Hexham was fought at Linnels on May 14, 1464.

The Sixth Battle: The River Bassas

The Bassas river is the most problematic of the Arthurian battle sites, as no such stream name survives and we have no record other than this single instance in the HB of there ever having been a river so named. We can only say that the location of the Bassas may be somewhere in the same general region as the Glen and Devil's Water battles. We will see below that the locations of subsequent battle sites will support this notion.

Some Arthurian theorists have opted for very questionable identifications of Bassas. They have pointed to Bass place names such as Bass Rock in the Firth of Forth, the Bass at Inverurie in Aberdeenshire and Bass Hill at Dryburgh.

Alas, the etymology for bass is fairly recent. In the Scottish National Dictionary there is an entry under 'bass' as follows: “A workman's tool basket; also a basket for carrying fish – known in Banff and Fife: on Lothian coast ‘bass’ is a square straw basket about 2' by 2' used for carrying fish.”

Bass Rock and similar formations would have been named by fisher folk due to their resemblance to such a basket.

The Bass Burn or Bass ‘stream’, a tributary of the Scar or Scaur Water approximately 15 miles North-West of Dumfries and just south of Auchenhessnane, was originally called the Back

Burn. Both the 1st edition (1861) and 2nd edition (1899) Ordnance Survey maps name it as Back Burn. The 1955 edition names it as Bass Burn. It is possible that either the original surveyors simply misheard what the local people called it, or that later surveyors did. As there are other Back Burns in Lowland Scotland, the chances that this stream’s original name was Bass is slim.

An acceptable, and perhaps preferable, explanation for the name Bassas is that it records an OE personal name found in place-names, i.e.

Bassa. This is the view of Graham Issac.

The ending -as in Bassas would appear to have no explanation in either Latin or Welsh grammar. But it does have an explanation in Old English grammar. The name could thus be Old English. Just as Baschurch (Shropshire) is from Old English 'Basses cirice', i.e. 'Basse's church' (Eglwyseu Bassa in the Old Welsh poems), and Basford (Nottinghamshire) is Old English 'Basses ford', and Baslow Derbyshire) is Old English 'Basses hlaw', i.e. 'Basse's burial-mound'; so 'flumen quod uocatur Bassas' is easily understood as 'the river which is called Basse's', i.e. 'Basse's river'. There is a Basingbourne in Cambridgeshire, Old English Basingeburna, which is 'the stream of Basse's people', 'Basse's kin's stream'.

There are two genuine OE Bassa place-names further north in Northumberland. Bassington in Cramlington parish was a farmstead a litte over a mile north-west of the village. It appears on a map of 1769 and is probably a much older site. In the present day town of Cramlington the site of Bassington Farm is on the Bassington Industrial Estate. However, other than this Bassington's proximity to the Devil's Water at Linnels (approximately 20 miles as the crow flies), there is little to recommend it as the site of Arthur's Bassas River battle. Most damaging, there is no stream here.

The other ‘tun of Bassa’s people’ is at the confluence of the Aln and the Shipley and Eglingham Burns not far east of a Roman road that connects Dere Street and the Devil’s Causeway. This Bassington is also roughly equidistant between the Northumberland Glen and the Devil’s Water/Dubglas near Hadrian’s Wall, and near the Roman fort of Alauna on the Aln at Low Learchild.

Once again, however, there is no stream at Bassington bearing the Bassa name.

In the East Riding of Yorkshire, near Bridlington, there is a place I originally overlooked. This is Bessingby, the by or ‘farmstead, village, settlement’ of the people of Bassa. The important thing about Bessingby is that there was a Romano- British settlement here (http://www.pastscape.org.uk/hob.aspx?hob_id

=1191551) and a Bessingby Beck or stream nearby.

A Roman road ran from Stamford Bridge to Bridlington

(http://www.pastscape.org.uk/hob.aspx?hob_id

=1029959&sort=4&search=all&criteria=Orby&ra tional=q&recordsperpage=10) and some believe (see Rivet and Smith) this to be the territory of the Gabrantovices, probably “cavalry fighters” and not “goat fighters”. The Gabrantovicum Sinus of Ptolemy would then be Bridlington Bay.

It is quite conceivable that Bessingby Beck was once known simply as “Bassa’s Stream”.

There is a problem with this placename, however. The –by terminal is Norse, and it is likely, therefore, that the entire name is not from Bassa, but from Bessi. Or, that if Bassa is the name recorded, it would not have become a place-name until fairly late. Here is what Alan James had to say on this subject:

“A. H. Smith's PNERY, 100, which says: 'The first element may be a patronymic formation, "the people of Basa or Besa", but there is little or no evidence for such -inga- formations with OScand by. It is therefore more likely to be a patronymic Basing or Besing with an uninflected genitive. Each name is well recorded... the fo mer may be from OE Bassa or OScand Bassi, the latter from OScand Bessi (a variant of Bersi....).

As there is no clear evidence for a change of a to e in ME, Besing- seems more probable, and in that case the less frequent but earlier Basing- forms would be Anglo-Norman spelling variants... Besing's farmstead'.

Subsequent work, especially by Barrie Cox, has demonstrated that the patronymic '-inga-' formations in S and E England (as far north as Yorks) date from the pre-Christian period, so such formations would have been long since obsolete by the time OScand by was introduced.

Smith's etymology would imply an Anglo- Scandinavian formation from the late 9th - early

11th ct.”

Thus this would seem an unlikely candidate for Arthur’s Bassas battle.

The conventional thinking on the Bassas name is to derive the first component from W. bas. Kenneth H. Jackson first discussed this possibility.

According to the Gieriadur Prifysgol Cymru, bas means ‘shallow, not deep, fordable; shallows, shallow water’. This would make a great deal of sense for a river-name – or even merely a DESCRIPTION of a river or stretch of a river.

Alan James of BLITON was kind enough to send me the following on bas in place-names:

Bas Late Latin *bassus adopted as Late Brittonic *basso-/ā- > Middle - Modern Welsh bas, Cornish *bas (in compound and place-names, see CPNE p. 18), Breton bas The Latin origin is reasonably certain, though the Late Latin ancestral form seems elusive. 'shallow', adjective. (a2) Bazard Lane Wig (stream-name, New Luce) PNGall p. 34 bas- + -ar, which see [+ SW Scots lane < Gaelic leana, 'a slow, boggy stream']. (c2) Bazil Point Lanc (Overton) PNLanc p. 175 ?bas- + linn, proposed by R. Coates, CVEP p. 318. Oliver Padel Cornish Place-name Elements Nottingham 1985: *bas 'a shallow', as a noun, 'shallows': only in basdhowr glossing vadum 'a ford' ... the verb occurs, ppp basseys 'abated'... Welsh and Breton bas... [occurs in Cornwall in:] C2) [= specific in name-phrase] Carn Base, coast[al name]; ?Park an Bays f[ie]ld[-name]

Alan James again came to the rescue when I asked how Bassas may have developed out of Late Latin or Late Brittonic:

“By the time the Latin word was adopted by Britt speakers, its inflectional forms were probably quite reduced at least in "vulgar" speech, and the Britt inflextions likewise. So your hypothetical form would be, for practical purposes *bassas. The -as suffix is nominal, noun-forming, it would be 'a shallow, shallows'. I suppose that might be a stream-name, more likely a name for a stretch of a river or a point on a river or estuary, a strategic location where a battle might well be fought, though of course there must be scores of possible candidates.”

Long ago the antiquarian Skene suggest Dunipace ner Falkirk in Stirlingshire for Arthur’s Bassas. The idea has not been thought well of by scholars over the years. However, recently place-name expert John Reid has tentatively proposed that Dunipace might be rendered Dun y Bas, the ‘Hill of the Ford.’

Commenting on this possibility, Alan James shared this with me:

“It ought to be *din-y-bas, not **dun-y-bais (that's what misled me); it would mean more correctly 'fort of the shallow', which is apparently okay topographically; the changes din > dun, /b/ > /p/, and /a/ > long /a:/ could all be explained in terms of adoption by Gaelic speakers. 'Hills of death' [a local, traditional etymology] would be G *duin-am-bais, which I wouldn't rule out, though I'm uneasy with /mb/ > /p/.”

If we may allow for bas here to be for a shallow ford, something rather remarkable occurs: we find ourselves directly between the Dumyat and Myot Hill hillforts which delineate the territory of the ancient Pictish Maeatae. According to the Life of St. Columba by Adomnan, Arthur son of Aedan of Dalriada died fighting the Miathi.

Furthermore, on the Carron was the remarkable Roman monument called Arthur's O've from early on. For more on this building, see

https://spns.org.uk/wp-content/uploads/2019/12/John-Reids-East-Stirlingshire-place-name-data-v2019.pdf

https://canmore.org.uk/site/46950/arthurs-oon-stenhouse

https://falkirklocalhistory.club/around-the-area/castles/arthurs-oon/

https://web.stanford.edu/group/archaeolog/cgi-bin/archaeolog/2012/03/21/arthurs-oon-a-lost-wonder-of-britain-part-1/#:~:text=Arthur's%20O'on%20(%3D'Oven,village%20has%20taken%20its%20name

https://falkirklocalhistory.club/about/about-the-society/faqs/what-was-arthurs-oon/

https://falkirklocalhistory.club/around-the-area/houses-and-estates/dunipace-house/

Alan James added this on the possibility of a bas name at Dunipace:

"Bas doesn’t primarily indicate a ford, it means ‘shallows’. Rather than a crossing here, there might have been a way up and down the riverside usable by a fighting force at low tide. There were routes like that mapped, and used by herdsmen, alongside the estuaries near where I am in recent times."

I would then identify the Bassas River with the bas on the Carron. This battle would then be an intrusion into the list of a battle belonging to a later Arthur.

The Irish Annals place the Dalriadan Arthur’s death in Circenn. [For Arthur’s contest with an opponent at Abernethy on the border of Circenn, see Chapter 4 below.] This has created a major problem, for Circenn is the Pictish province lying to the north of the Firth of Tay and this is quite a distance from the territory of the Miathi. Scholars have tried to account for this confusion over the battle site location in various ways. Bannerman attemtps to offer an explanation (pp. 84-85 Studies in the History of Dalriada) for the two death sites. It would appear several battles had become confused in the Irish annals, with Arthur dying properly in the territory of the Miathi and NOT in Circenn.

However, suppose what we have here is a confused record of battles fought in the North by TWO ARTHURS - one who fought the Miathi at Dunipace/Bassas and another - the Dalriadan one - who was slain while fighting in Circenn?

The Seventh Battle: The Celidon Wood

Caledonia was originally the region of the Great

Glen in Highland Scotland inhabited by the Caledonii.

As such, in Classical usage Caledonia came to mean Scotland north of the Forth-Clyde isthmus. But in Welsh tradition - as is evidenced by the presence of Merlin at 1) Arthuret just north of Carlisle, 2) Drumelzier on the Tweed 3) the region near Glasgow, and 4) a mountain in the central Lowlands [see my The Mysteries of Avalon for a discussion of this last) - the Coed Celidon would appear to be at the heart of the Scottish Lowlands. It is generally accepted by scholars that this is indeed the location of the great wood in the Welsh sources.

We may be able to pinpoint the location of the Coed Celidon battle more precisely.

It is possible that a river-name in the area, believed to be a truly ancient hydronym, may have contributed to the idea in early Welsh tradition that Celidon lay in this part of the Scottish Lowlands.

Caddon Water, a tributary of the Tweed, has a Roman road. The etymology of Caddon (Keledenlee, 1175, Kaledene, 1296) is interesting.

From Alan James of BLITON:

Nicolaisen included Caddon Water among the *cal-eto- river-names. The final syllable is pro ably OE -denu added by Northumbrian English, though a secondary suffix isn't impossible. It is a very common hydronymic formation; *cal-eddoes indeed occur in ethnic names too ("hard men"), including that of the Calidonii.”

When I asked Mr. James whether this name could have contributed to the region thereabouts becoming known as the Celidon Wood, he r sponded:

“Well, yes, a name like *caleden could readily have attracted folk- or learned etymologising and dinnseannachas. I think it would have contracted to something like the modern form Caddon by the 15th ct, so I doubt whether such a thought would have arisen in the early modern period, when renewed interest in Tacitus etc., and even 'Nennius', gave rise to a good deal of fanciful etymologising.

But it's in an area with a good many P-Celtic pns, many of which I consider to be 'late' Cumbric, i.e. 10th-11th ct, when I think there was something of a revival/ reintroduction of the language in the upper Tweed/ Moorfoots/ Lauderdale area, and my hunch is that was the period when Arthurian and other (semi-) legendary associations were being attached to locations in that area, as in the Solway basin.

But I don't think the water-name would have been been given at that time, it's an 'ancient' hydronym that might have come to be associated with Caledonia because of the (accidental) similarity.”

There are remnants of a fort at Caddonlee by

Caddonfoot

(http://canmore.rcahms.gov.uk/en/site/54413

/details/caddonlee/). The famous Eildon Hills fort at the Roman period Trimontium on Dere Street is only a dozen or so kilometers to the east of the mouth of the Caddon. Several other hillforts are in the area and a Roman road went from Trimontium west along the Tweed to the Easter Happrew fort beyond Peebles.

The Eighth Battle: Castle Guinnion

The Castle (‘Castellum’) Guinnion has been identified with the Roman fort of Vinovium at Binchester, although the great Professor Kenneth Jackson thought this unlikely. It has since been noted, however, that Ptolemy’s alternative Vinnovium (B. *Uinnouion) brings us very close to the later name set down by Nennius. Vinnovium should have given in Old Welsh at this stage a form in –wy, but it could be that – ion has been maintained as a so-called ‘learned form’. Thus, the identification should not be rejected.

It is even more possible that Guinnion is a slight error for Guinuion. This idea was proposed by Anscombe, and I have discussed it at length in

https://mistshadows.blogspot.com/2023/03/a-new-identification-of-arthurs.html. A form Guinuion would be from *Vinoviones, the ‘people of Vinovia.’

It is also true that Binchester would have been considered a castellum:

"Yes, it [Binchester] could [be referred to as a castellum]. Modern writers use 'castra' of legionary fortresses, 'castellum' of auxiliary forts, which I would think reflects Roman usage. One of the Vindolanda tablets even refers to 'reditus castelli', the financial accounts of the 'fort'. Binchester was an auxiliary fort, so no different in terms of its classification than Vindolanda. I don't have the figures to hand, but the two forts would have been much the same size. Housing 500 men, as against the 5,000 of a legion.”

- Professor Roger Tomlin

"Castellum or castrum/castra could have been used as terms for Binchester in that period, as earlier.

The latest of the two forts here was approximately 5.6 hectares or 14 acres in size and continued to be occupied in some form or another for about 600 years after Britain ceased to be part of the Roman Empire as evidenced by industrial and animal processing activities and a cemetery."

- Dr. David Mason (primary excavator of the Binchester Roman fort)

Binchester is not far south of Hadrian’s Wall on the Roman Dere Street. The fort stands on a spur of high ground some one and a quarter miles north of Bishop Auckland. It overlooks a loop in the river Wear and is in an excellent defensive position.

The fort was built in 79 CE during the Roman advance into northern England. From the early second century Binchester and the other Dere Street forts became important supply depots for Hadrian’s Wall and developed as military centres controlling the region south of the Wall.

The fort was in continuous military use until the early years of the 5th century. After the final withdrawal of the garrison the fort and the surrounding vicus (civilian settlement) continued to be occupied by the local, native population and it would seem that Binchester remained an important small town. By the beginning of the 6^th century the fort buildings were being torn down and stripped of stone. Part of the site was ut lized by Anglo-Saxons as a cemetery.

I had at one time proposed Carwinning in Dalry parish, Ayrshire, which is from a Caer + Winnian. This looks very good, but if a battle were fought here during Arthur’s time it was certainly not against the English. I discuss this place in further detail in the blog article I mention above.

There are, of course, several “Gwynion” place-names in Wales, but again, none of them work for Arthur.

In passing, I would put forward an additional, though tentative argument in support of Binchester as Guinnion. In the Introduction I alluded to Arthur’s carrying of the image of Mary on his shield during the Guinnion battle. It may not be a coincidence that Binchester is known for having a cult of the Mother Goddesses at its Roman fort. It is possible Mary in the Arthurian battle context is a Christian substitution for the Binchester ‘Mother.’

From http://theses.univ-lyon2.fr/documents/lyon2/2009/beck_n#p=0&a=top (Goddesses in Roman Religion, thesis by Noemie Beck, 2009):

“In Britain, the dedications to the Matres amount to approximately fifty inscriptions, all but a few from military sites, notably along Antonine’s and Hadrian’s Wall, and dedicated by soldiers. One of the few exceptions is the inscription to the Matres Ollototae, which is from the non-military site of Heronbridge, Cheshire. This suggests that the cult of the Matres and Matronae was brought to Britain by auxiliary troops from the Continent, such as by the Germanic legionaries of the Roman army. However, it does not mean that the Celtic peoples from Britain did not have any cultural notions of the Mother Goddesses, only that some particularities in the worship must have come with the army. The cult of the Mothers in Britain is clearly Romanized, for they all bear Roman epithets, such as Transmarinae, Campestres, Domesticae or Fatae, apart from the Matres Ollototae and the Matres Suleviae. The Matres Ollototae are undeniably Celtic, for their name is composed of Celtic ollo-, ‘all’ and teuta, touta, ‘tribe’. They are thus ‘The Mothers of All the Peoples’. They are mentioned in an inscription from Heronbridge (Claverton, Cheshire): Deabus Matribus Ollototis Iul(ius) Secundus et Aelia Augustina, ‘To the Mother Goddesses Ollototae, Julius Secundus and Aelia Augustina (set this up)’, and in three inscriptions from Binchester (Durham): Deab(us) Matrib(us) O[l]lot(otis) T[i]b(erius) Cl(audius) Quintianus b(ene)f(iciarius) co(n)s(ularis) v.s.l.m., ‘To the Mother Goddesses Ollototae Tiberius Claudius Quintianus beneficiaries of the governor, willingly and deservedly fulfilled his vow’ ; [M]atrib(us) O[lloto(tis)] CARTO VAL MARTI Vetto(num) GENIO LOCI LIT . IXT, ‘To the Mother Goddesses Ollototae … Cavalry Regiment of Vettonians….’ ; I(ovi) O(ptimo) M(axiom) et Matribus Ollototis sive Transmarinis, ‘To Jupiter, Best and Greatest, and to the Ollototae or Overseas Mother Goddesses’.”

The Ninth Battle: City of the Legion

The City of the Legion (Urbs Legionis) is, in this context, the Roman legionary fortress at York, the Romano-British Eburacum.

Dere Street began at the fort and ran north to Hadrian’s Wall and beyond. The argument against York is that, according to Welsh sources, the only Roman forts called Cities of the Legion were Chester or Deva and Caerleon or Isca. But to claim the Welsh were ignoarant of the fact that York was a legionary fortress seems very doubtful.

To begin, we have chieftains such as Peredur son of Efrauc (Efrauc = Eburacum/‘York’) and Peredur son of Eliffer (Eleutherius) Gosgordfawr. Peredur is a Welsh rendering of the Roman rank of Praetor. The governor or legate of Britannia Inferior, that is Northern Britain, was in the later period of praetorian rank.

The Roman emperor Caracalla reviewed the administration of Britain and split the province into two: Britannia Superior in the south, which had a consular governor based at London with two legions, the Twentieth at Chester and the Second at Caerleon. Britannia Inferior in the north had a praetorian governor with only one legion, the Sixth at York, where the governor also resided.

The Romans constructed their first fort at Eboracum in 71 CE. The fort’s rectangular construction consisted of a V-shaped ditch and earthen ramparts with a timber palisade, interval towers and four gateways. It covered about 50 acres of a grid-plan of streets between timber barrack blocks, storehouses and workshops. More important buildings included the huge Principia (Headquarters Building), the Commandant's House, a hospital and baths. The fort was designed to house the entire legion and remained a military headquarters almost to the end of Roman rule in Britain.

The fortifications at York were strengthened around 80 CE by a caretaker garrison while the Ninth Legion campaigned with the governor, Julius Agricola, in Wales and Scotland. The original fort was replaced in 108 CE by a massive stone structure with walls that survived long enough to be incorporated into the defenses of Viking and even later medieval York.

The one thing that makes York somewhat suspect as an Arthurian battle site is the presence there during the Roman period of the camp prefect Artorius, from whose name Arthur derives. It is certainly possible the memory of this Artorius influenced the placement of the Dark Age Arthur at the fort.

Matthews (http://www.historiabrittonum.net/wp-content/uploads/2018/09/The-Arthurian-battle-list-of-the-Historia-Brittonum.pdf) settles for the tried and true Chester for Arthur's City of the Legion battle, even though it makes absolutely no sense in the context of battles against the invading English. He says

"A recent attempt to identify it with York (Field 1999; Malcor 1999; Reid 1999, 223; Field 2008, 15; Hunt 2012, 115) is misguided, as there FIGURE 8: GUINNION FIGURE9: URBE LEGIONIS 16 is no evidence to indicate that York was ever known as anything other than Eburacum/Eboracum in British Latin or Cair Ebrauc in Old Welsh (Rivet & Smith 1979, 337; Green 2007, 209)."

Once again, what is important in this statement is what it omits. He does not like the idea that the name Arthur - indisputably from Latin Artorius - could have been preserved in the north because among this or that group there was a folk or ancestral memory of the L. Artorius Castus who was stationed at York in the 2nd century. In addition, he does not take into account the link between Peredur son of Eliffer of of the Great Warband and Peredur son of Efrawc/Ebrauc (this last being an eponym for York). I have suggested that the Great Warband (W. gosgordd) stands for the Sixth Legion, which was stationed at York. The name Eliffer itself is from Eleutherius, was a Greek title known to be applied to Constantine the Great, whose association with York is well known. The other Arthurian battles stretch up and down Dere Street, either on that Roman road or slightly to the east or west of it. Thus York, undeniably the most important British legionary fortress during the late Empire, remains the best candidate for Arthur's urbs legionis.

The Tenth Battle: Shore of the River Tribruit

The location of the shore (W. traeth) of the river Tribruit has remained unresolved. The clue to its actual whereabouts may lie in the two possible meanings assigned to this place-name.

According to Kenneth Jackson (_Once Again A thur's Battles_, MODERN PHILOLOGY, August,

1945), Tribruit, W. tryfrwyd, was used as an a jective, meaning "pierced through", and sometimes as a noun meaning "battle". His rendering of traeth tryfrwyd was "the Strand of the Pierced or Broken (Place)". Basing his statement on the Welsh Traeth Tryfrwyd, Jackson said that "we should not look for a river called Tryfwyd but for a beach." However, Jackson later admitted (in The Arthur of History, ARTHURIAN LITERATURE IN THE MIDDLE AGES: A COLLABORATIVE HISTORY, ed. by Roger Sherman Loomis) that "the name (Traith) Tribruit may mean rather 'The Many-Coloured Strand' (cf. I. Williams in BBCS, xi [1943], 95).

Most recently Patrick Sims-Williams (in The Arthur of the Welsh, THE EARLY WELSH ARTHURIAN POEMS, 1991) has defined traeth tryfrwyd as the "very speckled shore" (try- here being the intensive prefix *tri-, cognate with L. trans). Professor Sims-Williams mentions that 'trywruid' could also mean "bespattered [with blood]." I would only add that Latin litus does usually mean "seashore, beach, coast", but that it can also mean "river bank". Latin ripa, more often used of a river bank, can also have the meaning of "shore".

The complete listing of tryfrwyd from The Dictionary of Wales (information courtesy Andrew Hawke) is as follows:

tryfrwyd

2 [?_try-^2^+brwyd^2^_; dichon fod yma fwy

nag un gair [= "poss. more than one word here"]]

3 _a_. a hefyd fel _e?b_.

6 skilful, fine, adorned; ?bloodstained; battle,

conflict.

7 12g. GCBM i. 328, G\\6aew yg coryf, yn toryf,

yn _tryfrwyd_ - wryaf.

7 id. ii. 121, _Tryfrwyd_ wa\\6d y'm pria\\6d

prydir, / Trefred ua\\6r, treul ga\\6r y gelwir.

7 id. 122, Keinuyged am drefred _dryfrwyd_.

7 13g. A 19. 8, ymplymnwyt yn _tryvrwyt_

peleidyr....

7 Digwydd hefyd fel e. afon [="also occurs as

river name"] (cf.

8 Hist Brit c. 56, in litore fluminis, quod vocatur

_Tribruit_; 14 x CBT

8 C 95. 9-10, Ar traethev _trywruid_).

Tryfrwyd itself, minus the intensive prefix,

comes from:

brwyd

[H. Grn. _bruit_, gl. _varius_, gl. Gwydd. _bre@'t_

`darn']

3 _a_.

6 variegated, pied, chequered, decorated, fine;

bloodstained; broken, shattered, frail, fragile.

7 c. 1240 RWM i. 360, lladaud duyw arnam ny

am dwyn lleydwyt - _urwyt_ / llauurwyt escwyt

ar eescwyd.

7 c. 1400 R 1387. 15-16, Gnawt vot ystwyt

_vrwyt_ vriwdoll arnaw.

7 id. 1394. 5-6, rwyt _vrwyt_ vrwydyrglwyf rwyf

rwyd get.

7 15g. H 54a. 12.

The editors of GCBM (Gwaith Cynddelw

Brydydd) take _tryfrwyd_ to be a fem. noun =

'brwydr'. They refer to Ifor Williams, Canu Aneirin

294, and A.O.H. Jarman, Aneinin: Y

Gododdin (in English) p. 194 who translates

'clash', also Jarman, Ymddiddan Myrddin a Thaliesin,

pp. 36-7. Ifor Williams, Bulletin of the

Board of Celtic Studies xi (1941-4) pp. 94-6 suggests

_try+brwyd_ `variegated, decorated'.

On brwydr, the National Dictionary of Wales has

this:

1 brwydr^1^

2 [dichon ei fod o'r un tarddiad a@^

_brwyd^1^_, ond cf. H. Wydd. _bri@'athar_ `gair']

3 _eb_. ll. -_au_.

6 pitched battle, conflict, attack, campaign,

struggle; bother, dispute, controversy; host, army.

7 13g. HGC 116, y lle a elwir . . . y tir gwaetlyt,

o achaus y _vrwyder_ a vu ena.

7 14g. T 39. 24.

7 14g. WML 126, yn dyd kat a _brwydyr_.

7 14g. WM 166. 32, _brwydreu_ ac ymladeu.

7 14g. YCM 33, llunyaethu _brwydyr_ a oruc

Chyarlymaen, yn eu herbyn.

7 15g. IGE 272, Yr ail gofal, dial dwys, /

_Brwydr_ Addaf o Baradwys.

7 id. 295.

7 1567 LlGG (Sall) 14a, a' chyd codei _brwydyr_

im erbyn, yn hyn yr ymddiriedaf.

7 1621 E. Prys: Ps 32a, Yno drylliodd y bwa a'r

saeth, / a'r _frwydr_ a wnaeth yn ddarnau.

7 1716 T. Evans: DPO 35, Cans _brwydr_ y

Rhufeiniaid a aethai i Si@^r Fo@^n.

7 1740 id. 336, _Brwydrau_ lawer o Filwyr arfog.

Dr. G. R. Isaac of The University of Wales, Abe ystywyth, in discussing brwyd, adds that:

"The correct Latin comparison is frio 'break up', both < Indo-European *bhreiH- 'cut, graze'. These words have many cognates, e.g. Latin fr uolus 'friable, worthless', Sanskrit bhrinanti 'they damage', Old Church Slavonic britva 'razor', and others. The Old British form of brwyd would have been *breitos. It is sometimes claimed that there is a possible Gaulish root cognate in brisare 'press out', but there are difficulties with that identification.

It may be worth stressing that the 'tryfrwyd' which means 'very speckled' and the 'tryfrwyd' which means 'piercing, pierced' are the same word, and that the latter is the historically pri mary meaning. The meaning 'very speckled' comes through 'bloodstained' from 'pierced' ('bloodstained' because 'pierced' in battle). But I do not think this has any bearing on the arguments.

Actually, Tryfrwyd MAY mean 'very speckled', but that is conjecture, not certain knowledge. Plausible conjecture, yes, but no more certain for that."

That "pierced" or "broken" is to be preferred as the meaning of Tribruit is plainly demonstrated by lines 21-22 of the _Pa Gur_ poem:

Neus tuc manauid - "Manawyd(an) brought

Eis tull o trywruid - pierced ribs (or, metaphorically, "timbers", and hence arms of any kind,

probably spears or shields; ) from Tryfrwyd"

Tull, "pierced", here obviously refers to Tribruit as "through-pierced".

Professor Hywel Wyn Owen, Director of the Place-Name Research Centre, University of Wales Bangor, has the following to say on traeth + river names (personal correspondence):

"There are only two examples of traeth + river name that I know of, both in Anglesey (Traeth Dulas, Traeth Llugwy) but there may well be others. The issue is still the same however. Where a river flows into the sea would normally be aber. The traeth would only be combined with the river name if the river name was also used of a wider geographical context, and became, say, the name of the bay. Hence traeth + bay name rather than traeth + river name directly."

In the poem, the shore of Tryfrwyd battle is listed one just prior to Din Eidyn and once just after the same fort (I will have more on the Pa Gur battle sites below). The Gwrgi Garwllwyd or ‘Man-dog Rough-grey’ who is also placed at Tryfrwyd has been associated with the Cynbyn or ‘Dog-heads’ Arthur fought at Din Eidyn.

Manawyd's role at Tryfrwyd may suggest that this river or its shore is to be found in or on the borders of Manau Gododdin, which was the district round the head of the Firth of Forth, whose name remains in Slamannan and Clackmannan.

The Fords of Frew west of Stirling have been proposed as the site of the battle, but Jackson claims W. frut or ffrwd, ‘stream’, cannot have yielded frwyd. Jackson also countered Skene's theory that this was the Forth, on the grounds that the Welsh name for the Forth, Gweryd, which would be *Guerit in OW.

The poem may be even more specific, in that Traeth Tryfrwyd is said to be 'ar eidin cyminauc'

(line 28), ‘at Eidyn on the border’. Now, the ‘bo der’ here could be the Firth of Forth, but it is much more likely to be the line of division between Gododdin proper and Manau Gododdin.

The Cynbyn or ‘Dog-heads’ may partly owe their existence to the Coincenn daughter of Aedan, father of the Dalriadan Arthur, and to the Coinchend in the Irish story The Adventure of Art son of Conn. In this Irish tale, Art battles a monstrous woman named Coincenn or ‘Doghead’ who is a member of a tribe bearing the same name.”

The name of Art son of Conn's mother may be significant in this context. She was called Eithne, which was also the name of the mother of the god Lugh. The place-name Eidyn is of u known etymology. Because Din Eidyn was the capital of Lothian, and Lothian is derived from Middle Welsh Lleudinyawn, Brittonic *Lugudunia:non, land of ‘Lugh's (W. Lleu's) Fortress’, it would be reasonable to suggest that Eidyn as Lugh's fortress represents a British form of Irish Eithne. Din Eidyn would then be the Fort of (the goddess) Eithne.

[NOTE: The following is from Alan James on Lleudinyawn…

On Lothian, Haycock rejects Lugudunum - cf my BLITON note:

Lothian CPNS pp. 101-3 ? + -dīn- + suffix -*jānā- > -jǭn > -iawn: this etymology, yielding neoBrittonic *löw’ðïnjǭn > Middle Welsh Lleud[d]iniawn (as recorded circa 1170) was first proposed anonymously in Royal Commission on the Ancient and Historical Monuments of Scotland (1924) at p xviii (see Wilkinson, 2004, at pp. 83-4 n46), and subsequently by Koch in YGod(K) at p. 131. It would imply an unlocated *Lugudūnum somewhere in the region, possibly the Lugundunum recorded in the Ravenna Cosmography (see above). However, the suffix 'would seem to rule out' such a formation, Haycock 2013, p. 31 n45, which see with ibid. pp. 10, 11, 32 n46, and 34 n59, on the occurrence of this name in the 12th cent. 'Gwalchmai's Boast'.

Haycock, Marged, (2013) 'Early Welsh Poets Look North', 2ndAnderson Memorial Lecture, in Woolf, Alex, ed. Beyond the Gododdin, pp. 9 – 40.]

The Coincenn of the Irish are thought to be a reflection of the Classical Cynacephali.

Ole Munch-Pedersen cites the following note from Cecile Ó Rahilly text of the Irish heroic epic Cath Finntrágha or the “Battle of the White Strand” (Irish traigh is cognate with Welsh traeth):

"The Coinchinn or Coinchennaig are frequently mentioned in Irish literature. From the 8th cen-tury on the name was applied to pirates who ravaged Ireland. Cp. Thurneysen, Zu Ir. Hss., p. 24. In the Adventures of Art mac Cuinn they are represented as living in Tir na nIngnad whose King is called Conchruth (Éiriu III. 168). They are mentioned in a poem in the Book of the Dean of Lismore (Rel. Celt. I. 80) and in a poem is Duanaire Finn (xxxviii) where they are said to have invaded Ireland and been defeated by Finn. In the YBL tale Echtra Clérech Choluim Cille (RC XXVI 160 § 45, 161 § 48) men with dogs' heads are 'of the race of Ham or of Cain'. Similarly in the late romance Síogra Dubh the Caitchean-naigh and Coincheannaigh and Gabharchean-naigh are said to be do chinéal Caim mic Naoi (GJ XIX 99 5-6, cp. LU 122)." (Cath Finntrágha, (1962), lch. 65).

From the English translation of the Battle of Ventry/Cath Finntragha (http://www.maryjones.us/ctexts/f20.html):

“'O soul, O Glas son of Dreman,' said the king of the world, 'not a harbour like this didst thou promise my fleet would find, but shores of white sand where my army might assemble for fairs and gatherings whenever they were not fighting.' 'I know a harbour like that in the west of Erinn,' said Glas, 'namely, Ventry Harbour… They went onward thence to Ventry, and filled the borders of the whole harbour so that the sea was not vis-ible between them, and the great barque of the king of the world was the first to take harbour, so that thenceforward its name was Rinn na Bairci (The Point of the Barque). And they let down their many-coloured linen-white sails, and raised their purple-mouthed speckled tents, and consumed their excellent savoury viands, and their fine intoxicating drinks, and their harps were brought to them for long playing, and their poets to sing their songs and their dark conceits to them...

Now, these hosts and armies came into Ciarraige Luachra and to red-maned Slieve Mis, and thence to Ventry Harbour. 'O Tuatha De Danand,' said Abartach, 'let a high spirit and courage arise within you in the face of the battle of Ventry. For it will last for a day and a year, and the deed of every single man of you will be related to the end of the world, and fulfil now the big words ye have uttered in the drinking- hous-es.' 'Arise,O Glas, son of Dreman,' said Bodb Derg the son of the Dagda ,'to announce combat for me to the king of the world.' Glas went where the king of the world was. 'O soul, O Glas,' said the king of the world, 'are those yonder the fi-anns of Erinn?' 'Not they,' said Glas, 'but anoth-er lot of the men of Erinn, that dare not to be on the surface of the earth, but live in sid-brugs (fairy mansions) under the ground, called the Tuatha De Danand, and to announce battle from them have I come.' 'Who will answer the Tuatha De Danand for me?' said the king of the world. 'We will go against them,' said two of the kings of the world, namely, Comur Cromgenn, the king of the men of the Dogheads, and Caitch-enn, the king of the men of the Catheads, and they had five red-armed battalions in order, and they went on shore forthwith in their great red waves.

'Who is there to match the king of the men of the Dogheads for me?' said Bodb Derg. 'I will go against him,'said Lir of Sid Finnachaid,'though I have heard that there is not in the great world a man of stronger arm than he.’”

It is the Dogheads who would appear to hold the key to unravelling the Traeth Tryfrwyd mystery. Thanks to Lothian native and place-name expert John Wilkinson, who consulted a friend on the matter, I have learned the following:

“Ardchinnechena<n> is a place which the St. Andrews Foundation Account B says was where Hungus son of Forso placed the head of the de-feated Saxon king Athelstan on a pole “within the harbour which is now called Queen’s Ferry” (i.e. North Queensferry?); and which the shorter Account calls Ardchinnechun. Simon Taylor’s Fife Vol 3 offers ‘height/promontory of the head’ for the first and hints at a dindshenchas con-taining con ‘dog’ (in genitive) for the second.”

Ardchinnechena[n] is generally supposed to be the headland used by the Railway Bridge (see “Place-names of Fife”, vol. 1, 381-2, vol.3, 582-3).

This ‘Height of the Dog’s Head’ in North Queensferry Harbor reinforces my view that the Welsh tryfrwyd, ‘through-piered’, is an attempt to translate Latin trajectus, which has the exact literal meaning. However, trajectus also was the word used for a river-crossing, like the one at Queensferry. Frere (in his BRITANNIA, p. 162) discusses TRAJECTUS on a coin of Caracalla for a boat-bridge over the Forth or the Tay or both.

In passing, I would mention the Venicones tribe north of the Firth of Forth in Fife and Tayside. Their tribal name has been rendered either ‘Hunting hounds’ (Koch) or “Kindred hounds’ (Breeze). It is not impossible that these dog warriors may have contributed to the Dogheads being present at Tribruit.

The Eleventh Battle: Mount Agned and/or

Mount Breguoin

Mount Breguoin, found only in late rescensions of the Historia Brittonum has been associated with the ‘cellawr Brewyn’ or cells of Brewyn where Urien of Rheged later fought, a site generally agreed to be the Roman fort of Bremenium at High Rochester on Dere Street. Kenneth Jackson, who thought the name might also be an interpolation, came to this conclusion (“Arthur’s Battle of Breguoin”, ANTIQUITY 23, Jan 1 1949, p. 48). Most scholars now think that the Breguoin battle was taken from the Urien poem and incorporated into the Arthurian battle-list in the HB. As Arthur was linked to the Welsh word arth, ‘bear’, it may not be a coincidence that a bear god named Matunus was worshipped at Bremenium during the Roman period.

Mount Agned has hitherto escaped philological analysis. From Kenneth Jackson's time on, one original form proposed has been Angned. But this is an unknown word and has failed to produce a viable site. Most authorities feel that Agned is a corruption.

The simplest explanation for Agned as a corrupt form has been supplied by Dr. Andrew Breeze of the University of Navarre. Dr. Breeze proposed that the /n/ of Agned should be read as a /u/. This is a brilliant solution to the problem, although his attempt to then identify Agned with Pennango or Penangushope near Hawick is seriously flawed. I would see in this last a Pen, “headland”, plus the Gaelic personal name Angus. To Breeze, this is Pen + angau, the W. word for ‘death.’

As Alan James makes clear,

“We know in that area in the heart of the Southern Uplands, P-, Q-Celtic, Anglian and Scandinavian pers names were being used promiscuously, irrespective of the language or ethnicity of the bearers, and P-Celtic was probably still current late enough for a pen- to be named after an Angus.”

Penangushope would be ‘the narrow/enclosed valley of Angus’s Headland.’

Dr. Graham Isaac has the following to say on the idea that Agned could represent an original Agued:

“The n > u copying error is a common one. [See my treatment above of Guinnion as Guinuion, following Anscombe’s suggestion.] The word agued is a rare one, and is used only three times in the early materials. It means something like ‘dire straits, difficulty, anxiety’.”

The most important use of this word, for the present purpose, is found in Canu Aneirin line 1259, where it occurs in the phrase 'twryf en agwed', ‘a host in dire straits’. We will return to this phrase in a moment.

We have discussed the possibility that the Arthur section in HB represents a Latin retelling of an OW heroic poem. Such a poem could have had a line in it like 'galon in agued', 'the enemy in dire straits, great difficulty', much like the Canu Aneirin’s 'twryf en agwed'. It is conceivable that an author responsible for the Harleian recension of the HB (who may not have been entirely versed in the diction of OW heroic poetry) may have mistaken this 'agued' for a actual place-name, and wrongly placed the battle there: instead of 'the enemy in dire straits', he understood 'the enemy at Agued', easily miscopied.

Under this interpretation, the only location for the battle that was ever correct was Breguoin/Bremenium. This analysis at least solves the problem of 'Where was Agned?' with the answer, ‘There never was such a place, and so no need to look for it.’

What we may have in ‘Mount Agued’, then, is a confused reference to a battle at Mount Breguoin/Bremenium where the enemy found itself ‘in dire straits’. If so, we would have four, and possibly five battles said to have been fought by Arthur on Dere Street: York, Binchester, Devil’s Water, Celidon Wood and High Rochester.

The argument against Bremenium/High Rochester as an Arthurian battle, which relies upon the presence of gellawr brewyn, the ‘cells of Brem nium’, in the Urien battle poem list, ignores the very real possibility that more than one battle could have been fought at Bremenium at differ ent times. Bremenium is situated in a very strategic position, essentially guarding the pass over which Dere Street crosses the Cheviots. It is also true that Urien’s Brewyn could just as easily have been borrowed from the Arthurian battle-list as the other way around.

While it may well be that Agued/Agned is merely an error for Bregouin or a poetic name for the latter, there is a second possible identification for this Arthurian battle site. The ‘Twryf yn aguedd’ phrase mentioned above comes from the ‘Gwarchan Tudfwlch’, a poem appended to The Gododdin.

What is surprising about the ‘Gwarchan Tu fwlch’ example is that the phrase is preceded by two lines that copy part of a line found in Strophe 25 of The Gododdin proper:

“Arf anghynnull, Anghyman ddull, Twryf en agwed…”

“Arf anghynnull, anghyman ddull…”

Now, in the case of The Gododdin line, the poet Aneirin is referring to Graid son of Hoywgi’s prowess at the disastrous battle of Catraeth, Roman Cataractonium, modern-day Catterick on

Dere Street in Yorkshire. The Battle of Catraeth is, of course, the subject of The Gododdin poem.

The hero Tudfwlch hailed from the region of Eifionydd in Gwynedd, but he fought and died at

Catraeth. While he engaged in military actions in his homeland (the ‘Gwarchan’s’ ‘Dal Henban’ is almost certainly modern Talhenbont at Llanystumdwy in Eifionydd), it is probable that the lines borrowed from The Gododdin are meant to indicate that the following ‘Twryf yn aguedd’, ‘a host in distress’, is a reference to the British army at Catraeth.

Dr. Graham agrees with me on this assessment, saying that

“Phrases like twryf yn aguedd are characteristically used in early Welsh poetry to set up a general atmosphere of warrior violence, but, to judge from the final lines of the poem, it would seem to be primarily concerned with the 'Battle of Catraeth'.”

Part of the Roman fort at Catterick was built on the rising ground above the River Swale known as Thornbrough Hill. And Arthur is mentioned in

Line 972 of The Gododdin. Whether this is an interpolation or not, it is generally thought to be one of the earliest occurrences of his name in the written records:

“He fed black ravens on the rampart of a fortress

Though he was no Arthur.

Among the powerful ones in battle,

In the front rank, Gwawrddur was a palisade.”

Are we to see as a coincidence Arthur’s being mentioned in the context of the Battle of Catraeth when it is in this same battle, alone among all battles of the period, that a host finds itself in ‘agued’?

There are two possible ways to read this passage on Arthur in The Gododdin. First, the hero Gwawrddur, while a great warrior, was not nearly as great as Arthur. This is the standard interpretation. But let us suppose that what is really meant is that Arthur had fought at Catterick as well, a generation earlier, only he proved more powerful than Gwawrddur and won a victory over the Saxons on Thornbrough Hill, i.e. Mount Agned.

In this context, the Arthurian Mount Agned of the HB could be an anachronistic reference to the hill at Cataractonium, where the British army of Gwawrddur’s time found itself in ‘distress’ or ‘dire straits’ just prior to its annihilation by the Saxon foe.

So if we assume Agned = Agued = Catterick, where did Arthur fight – at High Rochester or Catraeth?

Well, the simple answer is ‘Either, both or neither.’ If Breguoin is indeed borrowed from the Urien poem, then Arthur did not fight at High Rochester. If Agned is Thornborough Hill at Catterick, then the site may have been chosen merely because his name was mentioned in The

Gododdin. Or Arthur may have fought at either or both places PRIOR to the later battles at those sites.

Almost the entire defensive circuit of the High Rochester/Bremenium fort is preserved, with the remains of the western gateway being particularly fine. There is also evidence of several periods of rebuilding in the western intervaltower of the south side. The ditches are well preserved to the north and east, outside which the line of Dere Street marches north-west. Between the thick stone ramparts the fort measures around 440 ft north-south by about 420 ft east-west, giving an occupation area of about 4.25 acres. There are inner stone buildings.

On the north, the remains of as many as thirteen ditches can be distinguished. On the east and south, four, and six ditches curve around the north-west angle. It is unknown how many ditches were on the west side of the fort.

The Roman fort at Catterick was likely founded during the early 70 CE's to guard the crossing of

Dere Street over the River Swale. At the very latest, the fort must have been in place by 79 CE, in order to guard the northern supply route of Agricola's Scottish campaigns. After an undetermined period of neglect, it would appear that the fort was recommissioned during the administration of Gnaeus Julius Verus in the aftermath of the Brigantian revolt of 155 CE, at which time the Antonine Wall was abandoned and the troops pulled back to Hadrian’s Wall in order to control the Brigantes. No trace of the fort remains, as it was overlain by the town of Catterick. A crop-mark east of Catterick Rac course has been identified as a Roman temporary camp not far from the fort.

The Twelfth Battle: Mount Badon

Badon is a difficult place-name for an unexpected reason. As Kenneth Jackson proclaimed:

"No such British name is known, nor any such stem." [To be briefly mentioned in the context of Badon is the Middle Welsh word bad, 'plague, pestilence, death' (GPC; first attested in the 14th century), from Proto-Celtic *bato-, cf. Old Irish bath. Some have asked me whether this word could be the root of Badon - to which Dr. Graham I. Isaac, of the National University of Ire-land, Galway, responds emphatically, "No, absolutely no. A (modern) W form _bad_ etc. would have been spelt in the W of the ancient period as _bat_ and there can be no connection since _Bad(on)_ is what we find." Other noteworthy Celtic linguists, such as Dr. Simon Rodway of Aberystwyth University, Dr. Richard Coates of the University of the West of England and Professor Ranko Matasovic of the University of Zagreb, agree with Isaac on this point. Matasovic adds: “Professor Isaac is right; since we have references to Badon in Early Welsh sources, the name would have been spelled with –t- (for voiced /d/). The spelling where the letter <d> stands for /d/ and <dd> for the voiced dental fricative was introduced in the late Middle Ages.”]

Graham Isaac has the following to say on the nature of the word Badon, which I take to be au-thoritative.

His explanation of why Gildas's Badon cannot be derived from one of the Badburys (like Liddington Castle, often cited as a prime candidates for Badon) is critical in an eventual identification of this battle site. Although long and rather complicated, his argument is convincing and I have, therefore, opted to present it unedited:

"Remember in all that follows that both the -d - in Badon and the -th- in OE Bathum are pronounced like th in 'bathe' and Modern Welsh - dd-. Remember also that in Old English spelling, the letters thorn and the crossed d are interchangeable in many positions: that is variation in spelling, not in sound, and has no significance for linguistic arguments.

It is curious that a number of commentators have been happy to posit a 'British' or 'Celtic' form Badon. The reason seems to be summed up succinctly by Tolstoy in the 1961 article (p. 145):

'It is obviously impossible that Gildas should have given a Saxon name for a British locality'.

Why? I see no reason at all in the world why he should not do so (begging the question as to what, exactly, is the meaning of 'British locality' here; Gildas is just talking about a hill). This then becomes the chief crutch of the argument, as shown on p. 147 of Tolstoy's article: 'But that there was a Celtic name ‘Badon’ we know from the very passage in Gildas under discussion'.

But that is just circular: ' "Badon" must be "Celtic" because Gildas only uses "Celtic" names'. This is no argument. What would have to be shown is that 'Badon' is a regular reflex of a securely attested 'Celtic' word. This is a matter of empirical detail and is easily tested; we have vast resources to tell us what was and was not a 'Celtic' word. And there is nothing like 'Badon'.

So what do we do? Do we just say that 'Badon' must be Celtic because Gildas uses it? That gets us nowhere.

So what of the relationships between aet Bathum - Badon - Baddanbyrig? The crucial point is just that OE Bathum and the Late British / very early Welsh Badon we are talking about both have the soft -th- sound of 'bathe' and Mod.Welsh 'Baddon'. Baddanbyrig, however, has a long d-sound like -d d- in 'bad day'. Both languages, early OE and Late British, had both the d-sound and the soft th-sound. So:

1) If the English had taken over British (hy-pothetical and actually non-existent) *Badon (*Din Badon or something), they would have made it *Bathanbyrig or the like, and the modern names of these places would be something like *Bathbury.

2) If the British had taken over OE Baddanbyrig, they would have kept the d-sound, and Gildas would have written 'Batonicus mons', and Annales Cambriae would have 'bellum Batonis', etc. (where the -t- is the regular early SPELLING of the sound -d-; always keep your conceptions of spellings and your conceptions of sounds separate; one of the classic errors of the untrained is to fail to distinguish these).

I imagine if that were the case we would have no hesitation is identifying 'Baton' with a Badbury place. But the d-sound and the soft th -sound are not interchangeable. It is either the one or the other, and in fact it is the soft th -sound that is in 'Badon', and that makes it equivalent to Bathum, not Baddanbyrig.

(That applies to the sounds. On the other hand there is nothing strange about the British making Bad-ON out of OE Bath -UM. There was nothing in the Late British/early Welsh language which corresponded to the dative plural ending - UM of OE, so it was natural for the Britons to substitute the common British suffix - ON for the very un-British OE suffix -UM: this is not a substitution of SOUNDS, but of ENDINGS, which is quite a different matter. That Gildas then makes an unproblematic Latin adjective with -icus out of this does not require comment.)

To conclude:

1) There is no reason in the world why a 6thcentury British author should not refer to a place in Britain by its OE name.

2) There was no 'British' or 'Celtic' *Badon.

3) 'Badon' does not correspond linguistically with OE Baddanbyrig.

4) 'Badon' is the predictably regular Late British / early Welsh borrowing of OE Bathum.

Final note: the fact that later OE sources occasionally call Bath 'Badon' is just a symptom of the book-learning of the authors using the form. Gildas was a widely read and highly respected author, and Badon(-is) (from Gildas's adjective Badon -icus) will quickly and unproblematically have become the standard book-form (i.e. pri-marily Latin form) for the name of Bath. Again, all attempts to gain some sort of linguistic mile-age from the apparent, but illusory, OE variation between Bathum and Badon are vacuous."

It is thus safe to say that 'Badon' must derive from a Bath name. However, we must not restrict ourselves to the Southern Bath, which makes no sense in the context of a Northern Arthur.

For as it happens, there is a major Northern ‘Bath’ site that has gone completely unnoticed!

In the the High Peak District of Derbyshire we find Buxton. This town had once been roughly on the southernmost boundary of Brigantian tribal territory (thought to lie along a line roughly from the Mersey in the west to the Humber in the east). It was also just within Britannia Inferior (that part of northern Britain ruled from York), whose boundary was again from the Mersey, but probably more towards The Wash.

In the Roman period, Buxton was the site of Aquae Arnemetiae, ‘the waters in front of (the goddess) Nemetia’. To the best of our knowledge, Bath in Somerset and Buxton in Derbyshire were the only two ‘Aquae’ towns in Britain.

But even better, there is a Bathum name extant at Buxton. The Roman road which leads to Buxton from the northeast, through the Peak hills, is called Bathamgate. Batham is ‘baths’, the ex-act dative plural we need to match the name Bathum/Badon. -gate is ‘road, street’, which comes from ME gate, itself a derivative of OScand gata. Bathamgate is thus ‘Baths Road’.

The recorded forms for Bathamgate are as fol-lows:

Bathinegate (for Bathmegate), 1400, from W.

Dugdale's Monasticon Anghcanum, 6 vols, London

1817-1830

Bathom gate, 1538, from Ancient Deeds in the

Public Record Office

Batham Gate, 1599, from records of the Duchy of Lancaster Special Commissions in the Public

Record Office.

Buxton sits in a bowl about one thousand feet above sea level surrounded by mountains and is itself a mountain spa. The natural mineral water of Buxton emerges from a group of springs at a constant temperature of 82 degrees Fahrenheit and is, thus, a thermal water. There are also cold springs and a supply of chalybeate (iron bearing) water. The evidence of Mesolithic man suggests a settlement dating to about 5000 BCE and archaeological finds in the Peak District around the settlement show habitation through the Neolithic, Bronze and Iron Ages to the time of the Romans.

From the historical evidence we can say that Buxton was a civilian settlement of some importance, situated on the intersection of several roads, and providing bathing facilities in warm mineral waters. In short, it was a Roman spa. Place-names in and around Buxton, and Anglo-Saxon finds in burial mound excavations, suggest a continuing inhabitation of the area and probable use of the mineral waters.

It has long been speculated that we should expect to find a military installation at Buxton. However, subsequent archaeological fieldwork, including excavations, in and around suggested locations at the spa town have singularly failed to establish a military presence. A 'ditch feature' identified initially through resistivity survey and then from aerial photography above Mill Cliff, Buxton, gave rise to the almost confident interpretation of this site as being that of the fort: subsequent evaluation in advance of development, however, has shown that these features were geological rather than man-made, and the absence of Roman finds of any description from a series of evaluation trenches suggests that if

Buxton had a fort it was located elsewhere.

Today, the site of the probable Roman baths is covered by the Georgian Crescent building. In this area during the seventeenth and eighteenth century discoveries of lead lined baths, red plaster and building remains were made at some considerable depth in the sediments which surround the area of St Anne's well. In the eighteenth century, Pilkington investigated a mound overlooking the site of the previous discoveries. Here he found a structure which has been interpreted as a probable classical temple - one of only three known from Britain. In the mid-seventies, following the removal of a 20th century swimming pool, a brick structure was exposed and a deposit containing 232 Roman coins, 3 bronze bracelets and a wire clasp ranging in date from the 1st to the end of the 4th century CE was excavated.

This intriguing series of early discoveries lends tangible support to the interpretation of Buxton as the 'Bath of the North', but the character and extent of civilian settlement - and whether this was in association with a military installation or not, remains obscure. A considerable range of small finds, together with occasional glimpses of apparently Roman contexts, from the backgardens of houses has failed to provide a clear sense of the extent of Roman Buxton, let alone a soundly based understanding of its chronology and development. The dating of coinage in the 'votive' deposit from near the Crescent might be seen to indicate heightened frequencies of offer-ings during the third and fourth centuries. To what extent this might correlate with the development of settlement at Buxton is a matter of some conjecture.

At Poole's cavern, Buxton, excavations between 1981 and 1983 by Peakland Archaeological Society and Buxton Archaeological Society produced a large Romano-British assemblage containing a considerable body of metalwork including coins and brooches, rolls of thin sheet bronze, along with ceramics, a faunal assemblage and burials. The dating of the coins and fibulae point to use between the late 1st and 3rd centuries, with the majority being of 2nd century date. Indeed, reanalysis of the material has suggested that the cave saw its principal period of use between 120 and 220 CE. The excavators appeared to reveal some spatial separation of the coin and fibulae finds from the pottery and faunal remains, although this has been questioned.

Discussing the possible character of the use of the site Bramwell and Dalton draw attention to the comparative absence of spindle whorls, loom weights and bone hairpins which might be expected from a domestic site. Instead, they see the evidence as supporting the interpretation of the site as that of a rural shrine or sanctuary.

This too has subsequently been questioned and rejected. Instead, Branigan and Dawley interpret the site as essentially domestic, but with the additional refuse from a metalworker’s activities. They see a link between Poole's Cavern and the growth of Buxton as a spa centre providing a ready local market for small decorative trinkets.

The general trend of the evidence suggests that the Roman site may have consisted of a temple overlooking a set of Roman baths. At Bath we have a clear idea of the layout of a significant bath/water shrine complex which consisted of two major ranges: a temple and a religious precinct, within which lay the sacred spring; along-side this range were a line of three baths within a major building, at one end of which lay a typical Roman bathhouse or sauna. The Bath buildings were lavishly built in a classical style and the whole complex attracted visitors from outside the province.

In essence the Buxton layout mirrors that a Bath: parallel to the spring line is a temple and alongside the springs is a range of possibly Roman baths. As the Buxton temple is two-thirds the size of that at Bath we could assume the Buxton complex was somewhat smaller.

If the grove of the goddess Nemetia continued as an important shrine well into Arthur’s time (and the presence of St. Anne’s Well at the site of the town’s ancient baths shows that the efficacy of the sacred waters was appropriated by Christians), there is the possibility the Saxons targeted Buxton for exactly this reason. Taking the Britons’ shrine would have struck them a demoralizing blow. If the goddess or saint or goddess-become-saint is herself not safe from the depredations of the barbarians, who is?

A threat to such a shrine may well have galvanized British resistence. Arthur himself may have been called upon to lead the British in the defense of Nemetia's waters and her templegrove.

There may be a very good reason why Gildas (or his source, or a later interpolator) may have opted for English Bathum (rendered Badon in the British language of the day). The two famous 'baths' towns were anciently known as Aquae Sulis and Aquae Arnemetiae for the two goddesses presiding over the hot springs. As Arthur is made out to be the preeminent Christian hero, who in the Welsh Annals has a shield bearing the Cross of Christ that he carries during the Battle of Badon, it would not do for the ancient Romano-British name to be used in this context. To have done so would inevitably have referred directly to a pagan deity. Hence the generic and less “connotation-loaded” Germanic name for the place was substituted. This explanation might do much to placate those who insist on seeing Badon as a Celtic name.

And where is the most likely location for the monte/montis of the Baths/Batham/Badon, where the actual battle was fought?

I make this out to be what is now referred to as The Slopes, at the foot of which is the modern St. Ann’s Well, and the Crescent, under which the original Roman bath was built. The Slopes were once called St. Ann’s Cliff because it was a prominent limestone outcrop. The Tithe map of 1848 shows that the upper half of the Cliff was still largely covered in trees. I suspect the spring was anciently thought to arise from inside the Cliff, and that the trees covering it marked the precincts of the nemeton or sacred grove of Arnemetia.

The three days and three nights Arthur bore the cross (or, rather, a shield bearing an image of a cross) at Badon in the Welsh Annals are markedly similar to the three days and three nights Urien is said to have blockaded the Saxons in the island of Lindsfarne (British Metcaud) in Chapter 63 of the HB. In Gildas, immediately before mention of Badon, we have the following phrase: "From then on victory went now to our countrymen, now to their enemies…" Similarly, just prior to mention of Urien at Lindisfarne, we have this: "During that time, sometimes the enemy, sometimes the Cymry were victorious…" It would seem, therefore, that either the motif of the three days and three nights was taken from the Urien story and inserted into that of Arthur or vice-versa.

What is fascinating about this parallel is that Lindisfarne or ‘Holy Island’, as it came to be known, was an important spiritual centre of Northern Britain. The inclusion of the three days and three nights (an echo of the period Christ spent in the tomb) in the Badon story suggests that we can no longer accept the view that Arthur's portage of Christian symbols at Badon was borrowed solely from the Castle Guinnion battle account in the HB. Aquae Arnemetiae, like Lindisfarne, was a holy place. Arthur's fighting there may have been construed as a holy act.

Supposedly, 960 Saxons were slain by Arthur at Badon. In the past, most authorities have seen in the number 960 no more than a fanciful embellishment on the Annals' entry, i.e. more evidence of Arthur as a ‘legend in the making’. But 960 could be a very significant number, militarily speaking. The first cohort of a Roman legion was composed of six doubled centuries or 960 men. As the most important unit, the first cohort guarded the Roman Imperial eagle standard.

Now, while the Roman army in the late period no longer possessed a first cohort composed of this number of soldiers, it is possible Nennius's 960 betrays an antiquarian knowledge of earlier Roman military structure. However, why the Saxons are said to have lost such a number cannot be explained in terms of such an anachronistic description of a Roman unit.

The simplest explanation for Nennius's 960 is that it represents 8 Saxon long hundreds, each long hundred being composed of 120 warriors. To quote from Tacitus on the Germanic long hundred:

"On general survey, their [the German's] strength is seen to lie rather in their infantry, and that is why they combine the two arms in battle. The men who they select from the whole force and station in the van are fleet of foot and fit admirably into cavalry action. The number of these chosen men is exactly fixed. A hundred are drawn from each district, and 'the hundred' is the name they bear at home. What began as a mere number ends as a title of distinction." [Germania 6]

Curiously, in the Norse poem Grimnismal, 8 hundreds of warriors (probably 960) pass through each of the doors of Valhall, the Hall of the Slain, at the time of Ragnarok or the Doom of the Powers.

Osla or Ossa Big-Knife and Caer Faddon

It has often been said that the Welsh Caer Faddon is always a designation for Bath in Avon. However, the only medieval Welsh tale to localize Arthur's Badon places it at Buxton in Derbyshire.

I am speaking, of course, of the early Arthurian romance ‘The Dream of Rhonabwy’, sometimes considered to be a part of the Mabinogion collection of tales. Rhonabwy is transported back in time via the vehicle of a dream to the eve of the battle of Caer Faddon. Arthur has apparently come from Cornwall (as he is said to return thither after a truce is made; this is almost certainly in this context a folk memory for the Cornovii kingdom, the later Powys) to mid-Wales and thence to Caer Faddon to meet with Osla or Ossa, a true historical contemporary of Arthur who lies at the head of the royal Bernician pedigree.

Here is the entry on Osla from P.C. Bartram's A CLASSICAL WELSH DICTIONARY:

"OSLA GYLLELLFAWR. (Legendary). ‘O. of the Long Knife’. According to the tale of ‘Rhonabwy's Dream’ it was against Osla Gyllellfawr that Arthur fought the battle of Badon (RM 150). He is represented as sending forty-eight horsemen to Arthur to ask for a truce till the end of a fortnight and a month (RM 159, 160). In RM 159 the name is spelt Ossa. The ‘Long Knife’ identifies him as Saxon, and as such he is foisted into the pedigree of Oswald, king of Northumbria, in a late version of Bonedd y Saint (§70+71 in EWGT p.64), where he roughly occupies the place of Ossa, grandfather of Ida king of Bernicia. In this context he is called ‘Offa (or Ossa) Cyllellfawr, king of Lloegr, the man who fought against Arthur at the battle of Badon’, and father of Mwng Mawr Drefydd. It is curious therefore to find Osla Gyllellfawr mentioned as one of the warriors of Arthur's Court in the tale of ‘Culhwch and Olwen’. Here it is said that he carried Bronllafn Uerllydan, (‘Shortbroad’). When Arthur and his hosts came to a torrent's edge, a narrow place on the water would be sought, and his knife in its sheath laid across the torrent. That would be a bridge sufficient for the hosts of the Island of Britain and its three adjacent islands and its spoil (WM 465, RM 109-10). Later in the story Osla took part in the hunting of the boar Trwyth. He and others caught him and plunged him into the Severn. But as Osla Big-knife was pursuing the boar, his knife fell out of its sheath and he lost it; and his sheath thereafter became full of water, so that as he was being pulled out of the river, it dragged him back into the depths (RM 140-1). A WELSH CLASSICAL DICTIONARY 587 It may be inferred that after the battle of Badon, Osla Gyllellfawr, being defeated, was supposed to have become subject to Arthur and to have served him until he was drowned in the Severn (PCB)."

Arthur is said to progress from Rhyd-y-Groes to Long Mountain and Cefn Digoll (Beacon Ring hillfort). As far as the text is concerned, Faddon is hard by this location. Yet we find nothing whatsoever in the region that could possibly be identified with Faddon.

However, if Rhyd-y-Groes were to be rendered directly into the English, it would be Crossford. And we find Crossford at a Roman road crossing of the Mersey. If one continues north from Crossford to Manchester, a Roman road then led straight to the SSE to Buxton. To me, therefore, it seems obvious that Rhyd-y-Groes is a standard relocation of the original site.

"Stretford proper lies in the south, taking its name from an ancient ford over the Mersey, also called Crosford. Leland about 1535 crossed the Mersey 'by a great bridge of timber called Crossford Bridge.'" (from https://www.british-history.ac.uk/vch/lancs/vol4/pp329-335)

While the romance is entirely fanciful, the chronological accuracy in the context of choosing Osla/Ossa is rather uncanny. Ossa is known in English sources for being the first of the Bernicians to come to England from the Continent. Under his descendants, Bernicia became a great kingdom, stretching eventually from the Forth to the Tees. In the 7th century, Deira – which controlled roughly the area between the Tees and the Humber - was joined with Bernicia to form the Kingdom of Northumbria.

In its heyday, Northumbria shared a border with its neighbor to the south – Mercia – at the River Mersey of ‘Boundary River’. The Severn is another major boundary river and perhaps that is why Crossford was wrongly placed there.

A tradition records Rhyd-y-Groes at Welshpool in Powys:

http://www.ancientwalesstudies.org/id211.html

https://coflein.gov.uk/en/site/403405/

This possible siting of the ford is also very far to the North, and if correct, points again towards Buxton and not to Bath.

The Mersey flows east to Stockport, where it essentially starts at the confluence of the River Tame and Goyt. The Goyt has its headwaters on Axe Edge, only a half a dozen kilometers from Buxton in the High Peak.

If we allow for the story’s author to have properly chosen Ossa as Arthur’s true contemporary, but to have viewed Northumbria in an anachronistic fashion – i.e. as extending to the River Mersey during Arthur's time – then Ossa coming from Bernicia in the extreme north of England, and Arthur coming from Powys of the Cornovii to the southwest, coming together for a battle at Buxton makes a great deal of sense.

Ossa would have been viewed as engaging in a battle just across the established boundary.

If I am right about this, the Welsh knew of the ‘Bathum’ or Badon that was Buxton and placed Arthur here as the victor in the great battle.

A Note on Caitlin Green’s Baumber

According to Matthews in http://www.historiabrittonum.net/wp-content/uploads/2018/09/The-Arthurian-battle-list-of-the-Historia-Brittonum.pdf,

“Caitlin Green (2007, 213) has suggested a site at Baumber (>Badeburg in Domesday Book) near

Horncastle, Lincolnshire. Her reasoning is based around the archaeological evidence for the early

(and apparently mass) settlement of Lindsey in the fifth century by people identified as Anglo Saxon (Leahy 1993, 36) and the presence of possibly two other sites in the battle list (glein and

dubglas) in the region of the *lindenses.”

Two major problems with this idea. Firstly, Badeburg or ‘Bada’s fortified place’ (Victor Watts, The Cambridge Dictionary of English Place-Names) does not accord with the form Badon-/Bathum. And unless we go with later spellings like Baenburch and Baumbir - themselves quite possible influenced by the nearby River Bain – we are again talking about something that should accord with Baddan-.

Secondly, there is no hill or mountain at Baumber. Furthermore, there is no hill-fort.

We can easily, then, dispense with this site as a candidate for Badon.

The Thirteenth Battle: Camlann

After these many victories, Arthur is said to have perished with Medraut at a place called Camlann.

Camlann has long been linked with Camboglanna, the ‘Crooked Bank’, a Roman fort towards the western end of Hadrian’s Wall. The only other candidates for Camlann are in NW Wales (the Afon Gamlan and two other Camlanns near Dolgellau), and these do not have anything to do with the Northern Arthur. For what looks to be a relocation of Arthur to a Camlan in NW Wales, see Appendix III.

Crawford pointed out that the best etymology for

Camlann would actually be B. *Cambolanda, ‘Crooked Enclosure’, an utterly unknown name, but Jackson had no problem with the derivation from Camboglanna.

Those who point to Camelon on the Antonine

Wall are ignorant of the fact that this place was originally called Carmuirs. It was renamed Camelodunum in 1526 by the antiquarian Hector Boece. He did this because the Camelon fort has been identified with the Colania of Ptolemy and the Ravenna Cosmography. Colania was confused with Colonia or Colchester, itself called

Camulodunum.

The Castlesteads fort sits on a high bluff overlooking the Cambeck valley and the break on the mosses to the north-west which carries the modern road from Brampton to Longtown. The site was drastically leveled in 1791, when the gardens of Castlesteads House were laid out and today nothing is visible of the fort aprt from the southern edge of the fort platform, while the view described above is obscured by trees. The Cam Beck has so far eroded the north-west front of the fort that the side gates now lie only 50 ft from the edge of the gorge. From east to west the fort measures 394 ft and it is thought to have been originally about 400 ft square, covering some 3.75 acres, though it is not impossible that the fort faced south rather than north and was therefore somewhat larger.

Excavations in 1934 revealed the east, west and south walls of the fort, the east and west doubleportal gates and south-west angle tower. The gate towers were built one course deeper than the fort wall, whose foundations were the normal

6 ft wide. All walls had been heavily robbed, but roof-tiles occurred in a number of the towers at ground-floor level, suggesting the possibility of oven-bases, as at Birdoswald, rather than collapsed roofs. Space allowed only for the identification of one ditch, 16 ft wide. No contact

has been made with any internal building, but an external bath-house was located and partly dug in 1741.

Castlesteads is unique along the whole Wall for sitting between the Wall and Vallum but not being attached to the former; presumably either its pre-existence or the lie of the land dictated its location.

A carved stone dated roughly 466-599 CE was found at Castlesteads. Because in the past the inscription has been read wrongly, i.e. upside down as ‘BEDALTOEDBOS’, this has been considered a corrupt attempt at the divine name BELATUCADROS, altars to whom were found here in a Roman context.

However, I have parsed the inscription as actually reading ‘SUB DEO LAUDIB[US]’, which according to Professor David Howlett of Oxford can be translated as ‘with the accompaniment of praises of God’. Therefore, this stone clearly denotes a Christian presence at Castlesteads during the time of Arthur.

In fact, this area may have been a Christian center during the generation preceding that of Arthur (see the note on the home of St. Patrick in the chapter below).

The Name Medraut/Modred (= Mordred) and His Possible Lineage

On February 26, 1996, I received a letter from Professor Oliver Padel of Cambridge. This was in response to a query I had sent him some time earlier in which I proposed that the name Medrawt – born by the personage who died with Arthur at Camlann – may represent the Roman name Moderatus. What Padel had to say on this possibility is important enough for Arthurian studies to be reprinted in full below:

“Not much has been done on the name of Medrawt or Mordred… In an article on various words in Welsh with the root med, Medr-, Ifor Williams suggested that the name might be connected with the Welsh verb medru ‘to be able, to hit’; but he did not develop the idea, only mentioned it in passing.

Middle Welsh Medrawt cannot formally be identical with Old Cornish Modred, Old Breton Modrot (both of which are recorded, indicating an original Old Co.Br. *Modrod), since the Welsh e in the first syllable should not be equivalent to a Co.Br. o there.

What people do not seem to have asked is what this discrepancy means: we can hardly say that Welsh Medrawt is a different name, since it clearly belongs to the same character as

Geoffrey’s [Geoffrey of Monmouth] Modredus < Co.Br. Modrod.

Which is ‘right’? I would suggest that the Co.Br. form is the ancient one, and that the Welsh form has been altered, perhaps indeed by association with the verb medru.

That was already my conclusion, but I did not have a derivation for Modrod. However, Modrod would be the exact derivative of Latin Moderatus, as you suggest. Your suggestion is most attractive, and neither I nor (so far as I know) anyone else has previously thought of it.

Like you, I should be relucatant to say that Modrod couldn’t have a Celtic derivation; but it fits so well with Moderatus that I personally don’t feel the need to look further.”

If Medrawt or, rather, Modrod, is Moderatus, this may be significant for a Medraut at Cambloglanna on Hadrian’s Wall, for we know of a Trajanic period prefect named C. Rufius Moderatus, whose unit left inscriptions at Greatchesters on the Wall and Brough-under-Stainmore in Cumbria (CIL iii. 5202, RIB 1737, 166-9, 2411, 147-51). The name of this prefect could have become popular in the region and might even have still been in use among Northern British noble families in the 6th century CE.

I once thought that the name of Medraut's father (the Loth of Geoffrey of Monmouth, properly rendered as Lleu in the Welsh sources; Lothian is derived from Middle Welsh Lleudinyawn, Brittonic *Lugudunianon, land of ‘Lugh's/Lleu's Fortress’) should perhaps be attached to Carlisle, the Romano-British Luguvalium, This was principally because the latter, whether interpreted as a place-name or a description of the fort, meant 'Lleu-strong.'

However, in going back over the Triads I realized that I'd missed something: Triad 70 lists as a son of Cynfarch, and thus brother of Urien and Efrddyl, a certain Lleu. As it happens, the Welsh version of Geoffrey of Monmouth's tale makes this Lleu son of Cynfarch the father of Medraut.

Lleu son of Cynfarch's mother was Nefyn - a name universally held to be cognate with the Irish goddess name Nemhain. Nemhain, in turn, often appears as the trio of battle goddesses which includes the Morrigan. In my THE MYSTERIES OF AVALON I made my case for the Welsh Morgan being a substitute for the Morrigan in Arthurian story. If I'm right, then Nemhain wife of Cynfarch could also be seen as the Morrigan/Morgan, grandmother of Medraut.

In Geoffrey's story, Medraut's mother is Anna, Arthur's sister. This may point to the Annan River (from a British form of the Irish goddess name Anu, or at least from the same root). According to the "Gorhoffedd" of Hywel ab Owain Gwynedd, Caerliwelydd, i.e. Carlisle in Cumbria, was in Rheged (see John Koch's CELTIC CULTURE: A CULTURAL ENCYCLOPEDIA).

I've recently shown that the heartland of Rheged was, in fact, Annandale, just across the Solway Firth from Carlisle. But it is not impossible that at some point Rheged did hold Carlisle, and that it was Cynfarch's son Lleu whose name may serve as a sort of partial eponym for that city. Medraut, then, would be from Luguvalium.

The Welsh name Gwyar as Medraut's mother, as she was the mother of Medraut's supposed brother Gawain. But I've shown in my treatment of the Northern kingdoms that Gwyar's people belonged at the pre-Saxon Bamburgh.

While it is true that Medraut son of Lleu son of Cynfarch and Nemhain-Morrigan/Morgan look very promising, we are still stuck with the c. 537 date for the former’s passing with Arthur at Camboglanna. According to Bartram, 585 or 586 are the most likely years to have seen the death of Urien, Medraut's uncle.

The chronology, in other words, doesn’t add up.

Conclusion: Arthur’s Military Role in the North

While some of the Arthurian battle sites as I have identified them must be considered problematic or even doubtful, there is no denying that when they are plotted out on a map (see p. 12 above) they stretch from south to north in a fairly well-defined line. Many center on the Roman Dere Street, which must be considered a sort of boundary or frontier zone between the Britons and their enemy, the Germanic invaders.

A battle at Camboglanna does indeed look like an internal conflict, and the tradition which records Medrawt/Moderatus as Arthur’s opponent may, in fact, be correct.

To summarize, I include here Alan James’ opinion of my Arthurian battle map, found at the beginning of this book:

“If you're assuming late 5th century, the archaelogical and (earliest OE) p-n evidence suggests the main concentration of Germanic-speakers would have been around the Humber, with control of York and extending west to the Magnesian Limestone/ Dere Street - i.e. the beginnngs of Deira and Lindsey; smaller but significant settlements along the Tees, and in the Yorkshire Gap, with control of Catterick; likewise along the Tyne and eastern part of Hadrian's Wall. Further north probably still P-Celtic, but there were of course strategic sites on both sides of the Forth; likewise to the west, strategic sites along the Wall and either side of the Solway Firth.
Whether or not Arthur was involved, I can well believe there were battles at all the places you've marked!”

CHAPTER 4

ARTHUR’S OTHER BATTLES: MYTHOLOGICAL OR MISTAKEN

The Pa Gur Battle Sites

The Arthur presented to us in the early Welsh poem Pa Gur is a very different personage from the one we find in the battle list of Nennius' HB. In Pa Gur, Arthur numbers among his men the mythological Manawyd(an) son of Llyr. He and his men fight monsters and witches. We have clearly departed from history and have embraced the realm of the fantastic.

While the Pa Gur is, alas, a fragmentary poem, the following battles or locations are listed in the order in which they occur.

Elei

Tryfrwyd

Din Eidyn

Celli

Afarnach's hall

Dwellings of Dissethach

Din Eidyn

Shore of Tryfrwyd

Upland of Ystawingun

Mon

Elei is generally identified as the Ely (Elai) River in southern Wales. However, the context of the PA GUR strongly suggests this site is in the North, and it must be one that can be associated with Mabon. In a note to my blog essay https://mistshadows.blogspot.com/2020/08/eil-mehyn-of-urien-rheged-and-eildon.html, I show that the hillfort Mabonlaw is situated very near the Catrail dyke in the parish of Roberton. The Catrail’s second element, a word that means “fence, palisade”, is likely found in the Elei river-name (see Schrijver, Peter, EMANIA 20, JUNE 2006, Early Irish Ailenn: An Etymology). Thus I proposed that Elei was a substitution of a more familiar name for the ail/eil in Catrail.

I have proposed above that Traeth Tryfrwyd is the shore of the trajectus at Queensferry west of Edinburgh.

Din Eidyn, as is well known, is Edinburgh. Arthur’s opponents in this battle are the Cynbyn or ‘Dog-heads’, whom I believe to be an echo of the Venicones tribe.

Afarnach’s hall may be a reference to the Pictish capital of Abernethy. Watson discussed the etymology of Abernethy as follows:

"Thus Abur-nethige of the Pictish Chronicle, now Abernethy near Perth, has as its second part the Genitive of a nominative Nethech or Neitheach (fem.), which is Gaelicized either from Neithon directly, or from a British river name from the same root."

Witches Hole is a small cave in a rocky face on the north side of the Castle Law fort at Abernethy. It is supposed to have been the residence of some of the Witches of Abernethy (http://canmore.rcahms.gov.uk/en/site/27921/details/witches+hole+castle+law/).

I would add that Neithon comes from an original Nechtan or Neachtan, which appears to be cognate with L. Neptune.

Abernethy is on the border region between the Pictish kingdoms of Fortriu and Circenn. We have seen above that the Dalriadan Arthur is said to have fought in Circenn, and the Abernethy/Afarnach battle may well be a traditional memory of the Circenn conflict.

For quite awhile now I've remained unsatisfied with the identification of Celli ("Grove") in the Welsh poem PA GUR. We are told in that work that Cai fought bravely at the site, and that the grove was lost (although who exactly lost it is impossible to tell from the context).

The best that top Arthurian scholars have been able to do so far is best summarized in see Nerys Ann Jones' ARTHUR IN EARLY WELSH POETRY, p. 40, Note 33, where it is suggested Celli may be Arthur's Cornish Kelliwic.

In the past, given that the majority of the PA GUR battle sites are in the North, I've sought to situate Celli in that region of Britain. The Ravenna Cosmography's Medionemeton or Middle Sacred Grove was a logical candidate. This nemeton may be Cairnpapple in West Lothian, although the Roman stone temple of Victory called Arthur's O'on on the Antonine Wall has also been proposed.

Alas, I overlooked two things. One, Cai of the PA GUR is otherwise placed in contests in Derbyshire and in Gwynedd. He is referred to as a Lord of Emrys (the Ambrosius who supposedly ruled Gwynedd from Dinas Emrys). And, two, we have Cai in a saint's VITA fighting - and winning! a battle at a place that is literally overlooking the Roman fort of Gelligaer in Glamorgan, Wales. Gelligaer is, transparently, the 'fort of the grove.' I'm talking, of course, about Fochriw Carn, where Arthur, Cai and Bedwyr are found sitting in the Life of St. Cadoc:

[http://ancienttexts.org/library/celtic/ctexts/cadog.html]

After a long interval of time the aforesaid king Gwynllyw, depending now on his kingdom, desired with ardent affection on account of the excessive sweetness of her fame that a certain girl should be joined to him in lawful wedlock, born of most noble lineage, of elegant appearance, very beautiful moreover in form, and clad in silk raiment, whose name was Gwladus, the daughter of a certain regulus, who was called Brychan. Accordingly, he sent very many messengers to the virgin’s father to the end that they might more resolutely demand that she might be given to him as wife. But the father of the girl, having received the message, was indignant, and, full of anger, refused to bestow his daughter on him, and slighted the messengers, and dismissed them without honour. They, taking this very badly, returned, and told their lord what had been done to them. When he had heard, the king, raving with excessive fury, armed with all possible speed three hundred of his young men to take the aforesaid girl by force. Then starting at once on their journey, when they reached the court of the aforementioned regulus, which is called Talgarth, they found the said virgin sitting with her sisters before the door of her chamber and at leisure in modest conversation, whom they immediately took by force, and beat a hasty retreat. When this was known, her father, Brychan, moved by grief of heart, sorrowing inwardly at the loss of his beloved daughter, called to his aid all his friends and his subjects to recover his daughter. When all his helpers had assembled together, with rapid steps he follows up the enemy and his confederates. Gwynllyw, when he had seen them, ordered that the oft-mentioned girl should be brought up to him, and he made her ride with him. He, carrying the girl cautiously with him on horseback, preceded the army not indeed for flight, but to await his soldiers and to exhort them manfully to war. But Brychan with his men, boldly attacking the savage king and his satellites, slew two hundred of them and followed them up as far as the hill, which is on the confines of either country, which in the Britannic tongue takes the name Boch Rhiw Cam, which means the cheek of the stony way. But when Gwynllyw had arrived at the borders of his land, safe in body with the aforesaid virgin, although sorrowful at the very great slaughter in the fight with his adversaries, lo, three vigorous champions, Arthur with his two knights, to wit, Cai and Bedwyr, were sitting on the top of the aforesaid hill playing with dice. And these seeing the king with a girl approaching them, Arthur immediately very inflamed with lust in desire for the maiden, and filled with evil thoughts, said to his companions, ‘Know that I am vehemently inflamed with concupiscence for this girl, whom that soldier is carrying away on horseback.’ But they forbidding him said, ‘Far be it that so great a crime should be perpetrated by thee, for we are wont to aid the needy and distressed. Wherefore let us run together with all speed and assist this struggling contest that it may cease.’ But he, ‘Since you both prefer to succour him rather than snatch the girl violently from him for me, go to meet them, and diligently inquire which of them is the owner of this land.’ They immediately departed and in accordance with the king’s command inquired. Gwynllyw replies, ‘God being witness, also all who best know of the Britons, I avow that I am the owner of this land.’ And when the messengers had returned to their lord, they reported what they had heard from him. Then Arthur and his companions being armed they rushed against the enemies of Gwynllyw and made them turn their backs and flee in great confusion to their native soil. Then Gwynllyw in triumph through Arthur’s protection together with the aforesaid virgin Gwladus, reached his own residence, which was situated on that hill, which thenceforward took from his name the British appellation Alit Wynllyw, that is, Gwynllyw’s Hill. For from Gwynllyw is named Gwynlliog, and Brycheiniog from Brychan.

Fochriw Carn may be one of the cairns on Mynydd Fochriw, but others prefer to identify it with the more significant cairn of Bugail.

https://coflein.gov.uk/en/site/301283/details/carn-y-bugail-gelligaer-common

https://coflein.gov.uk/en/site/535709/details/mynydd-fochriw-cairn-ii

https://coflein.gov.uk/en/site/225353/details/mynydd-fochriw-ring-cairn

The folks at CADW, for instance, opt for Carn y Bugail:

[https://cadw.gov.wales/sites/default/files/2019-05/South%20Wales%20Blaenau%20Gwent_EN.pdf]

When Arthur came to Gelligaer

The cairns at Fochriw captured the imagination of our medieval ancestors and were drawn into

Arthurian legend. Sometime in the 1070s or 1080s, a monk from Llancarfan in Glamorgan

by the name of Lifris wrote a Latin ‘biography’ of St Cadog, the Vita Cadoci, and among the tales he tells about the saint is an account of his birth. A local Gwentian king called Gwynllyw — after whom Gwynll[g (the western part of Gwent) was named — eloped with Gwladys, a daughter of Brychan, king of Brycheiniog. Brychan, naturally being somewhat put out at this unruly behaviour, gave chase with his warriors. When he had almost caught the couple, Gwynllyw and Gwladys came to a hill named Boch Rhiw Carn where they met Arthur and his boon companions, Cei and Bedwyr, playing dice. In later French and English tales these red-blooded Celtic heroes were transformed into the rather surly Sir Kay and wishy-washy Sir Bedivere. The ‘three vigorous heroes’ (tres heroes strenui) promptly defeated Brychan and his men in a bloody battle, but not before Arthur has considered kidnapping Gwladys for himself! The happy couple went on their way, and the result of their

newly-wed passion was the holy St Cadog himself. Boch Rhiw Carn (‘the cairn of Fochriw’) is clearly a reference to Carn Bugail. So here we have Arthur, Cei and Bedwyr fighting a battle on the bleak moors above Rhymney — at least in

the fevered imagination of an eleventh-century cleric. A further reminder of the story can be found at Capel Gwladys, about 3 miles (4.5km)

along the Roman road and about 1 mile (1.5km) from Bargoed (ST 125993). Here, within an impressive boundary dyke, you can see the restored foundations of a small, rectangular chapel with a modern carved cross marking the site of the altar. Although tradition has it that the chapel was founded by Gwladys in the sixth century, these remains are medieval in date. A carved grave slab found here and dated to the eighth or ninth century can be seen in the porch of Gelligaer church. On the open moorland of

Fforest Gwladys, about 550 yards (0.5km) to the southeast of Capel Gwladys, is one of the best-preserved and most accessible of the Roman practice camps (ST 131991).

The following excellent source tells us more about Fochriw Carn and its relationship to nearby Gelligaer:

https://www.caerphilly.gov.uk/CaerphillyDocs/Equalities/Place-Names-in-Caerphilly-County-Borough-2014.aspx

Fochriw (originally boch+rhiw+garn)

(phonetic: voch-riw)

OS Grid Reference - ST 103 054

The more recent form of the name should correctly be Y Fochriw and the literal translation of this village name is often given as “slope of the pigs” as the assumption is that the word “moch” (pigs) has mutated to “foch”. The original word however, was “boch” as the full name for the settlement is Bochriw’r Garn. This changes the meaning, as “boch”, though usually meaning “a cheek” as on a face, can also mean a bulge in the ground or a hill, possibly referring to a rounded piece of rock on the slope (“rhiw”) below the “carn”, the Roman stone found above the village on Gelligaer Common. Examples of the name can be found as far back as c1170 with Bohrukarn, later y voyghryw garn c1700 and Y Fochriw in 1867.

Gelligaer (gelli+caer)

(phonetic: gare-ll-ee-guy-rr)

OS Grid Reference - ST 135 969

Literally meaning “grove by the fort”, the village gets its name from its history as a Roman auxiliary fort and even further back in history from when there was an Iron Age fort on the adjoining hill, Buarth-y-gaer that is immediately to the east of the village. One of the greatest Welsh saints of the 6th century, Cadog, was born in Gelligaer (the local ward name is Saint Cattwg) and legend has it that he was a monk, had magical powers, was a kind and generous host and was a very successful dairy farmer - in fact the name carries on in Llangadog in West Wales, famous for the now closed creamery that produced fantastic custard and rice pudding, and you can still purchase the Welsh Cadog cheese in local supermarkets. The spelling of Gelligaer has altered over the years in reflection of the way the name has been pronounced e.g. Gelligâr from 1750. Gelligaer Church Hall, erected in 1911 has a plaque with the spelling Neuadd Kell Y Gaer 1911 which is still there. Early map spellings also have the name beginning with the letter K, such as Kil-gaer 1281, Kylthy-gaer in 1307, Kilthi-gaer in 1349 and Kethlygajer on Pieter van den Keere's map of Monmouthshire in 1605.

Gelligaer Common / Comin Gelligaer (gelli+caer)

(phonetic: com-in-gare-ll-ee-guy-rr)

OS Grid Reference - ST 125 985

Not so much a settlement as a scattering of dwellings on this open upland, the main population being Welsh Mountain ponies and other horses. Running roughly northwards across the Common is the Roman road from Cardiff to Y Gaer, near Brecon and this is still clearly visible above Fochriw.

I now have no reason to believe that Cai's 'Celli' is anything other than Gelligaer, and the 'grove' was lost to Brychan.

Dissethach, where Arthur’s opponent is Pen Palach, looks like Tig Scathach, ‘House of Scathach’, and Beinn na Caillich (allowing for the difference between P- and Q- Celtic), ‘Hill of the Witch’. Dunsgiath or Dun Scathach, the ‘Fort of Scathach’, and Beinn na Caillich, are both in the southeast of the Isle of Skye. From Beatrix Faerber, CELT project manager, we learn that there is a reference in Tochmarc Emire, which incorporates the story of Cu Chulainn’s training at arms with Scathach. In this case, Scathach’s house is tig Scathgi (= Schathaigi).

The upland of (Y)stawingun, where nine witches are slain by Cei, is quite possibly Stanton Moor in Derbyshire, where we find the stone circle called the Nine Ladies. The ‘lord of Emrys’ mentioned in the poem just prior to (Y)stawingun is a known periphrasis for Gwynedd, as Ambrosius/Emrys was the traditional lord of that land. Emrys in this context may actually be a reference to the Amber river, which lies just east of Stanton Moor.

The –gun, if from an earlier –cun, could have come about by mistaking in MS. an original t for c. The middle –w- may represent a u, such as is found in Staunton, a known variant of Stanton.

Much later story substitutes the hero Peredur and transplants the witches to Gloucester, presumably because of the presence in Gloucestershire of towns named Stanton and Staunton.

There is no mystery regarding Mon, as this is the common Welsh name for the Isle of Anglesey in northwest Wales. Welsh tradition insists that Cath Palug or Cath Palug, which Cai battles on Mon, is the cat of a person called Palug. Modern scholars prefer to view palug as perhaps meaning ‘scratching’ or ‘clawing’, hence Cath Palug as the Clawing Cat.

Cath Palug is linked in line 82 of the poem to ‘lleuon’, i.e. lions. The association of lions with Arfon (where the cat is born) and Mon may have to do with the simple confusion of llew, ‘lion’, for lleu, the god who is the Lord of Gwynedd in Welsh tradition. The letters u and w readily substitute for each other.

Two Additional Poetic References

Much has been made of early references to Arthur in three important poems: The Gododdin, Marwnad Cynddylan and Geraint son of Erbin. As I have discussed The Gododdin reference already above (Chapter 3) in the context of Arthur’s battle at Mount Agned, here I will restrict myself to a brief treatment of the other two poems.

MARWNAD CYNDDYLAN

Scholar Jenny Rowland has done a very nice job of disposing of the difficulty posed by Line 46 of Marwnad Cynddylan. The line in question reads:

Canawon artir wras dinas degyn

This has in the past been amended to read:

Canawon Arthur wras dinas degyn: "whelps of Arthur, a resolute protection"

Jenny Rowland, wisely, opts instead for:

Canawon artir[n]wras dinas degyn, i.e.: Canawon arddyrnfras dinas degyn: "strong-handed whelps…"

This nicely eliminates our having to associate Arthur with the Powys kingdom in east-central Wales.

GERAINT SON OF ERBIN

A harder thing to dispose of is the presence of Arthur’s name in the poem Geraint son of Erbin. While different versions of the poem exist, all are in agreement in including the name Arthur in one of their stanzas. This would not be a problem, were it not for the fact that, in Jenny Rowland’s words, "Despite the Arthurian link in Geoffrey of Monmouth’s work there can be no question that ‘Geraint fab Erbin’ is older than the Historia Regum Britanniae." In other words, someone, for some reason, seems to have placed Arthur in Dumnonia (Devon and Cornwall) prior to Geoffrey of Monmouth’s doing so.

If, as is genuinely agreed, Geraint son of Erbin is to be dated between the ninth and eleventh centuries, how do we account for Arthur being in Dumnonia? This is a critical question, for Geraint son of Erbin would seem to be our earliest source seeking to situate Arthur in extreme southwest England.

Using Rowland’s composite text, I can make the following observations: Geraint’s name occurs in 18 out of 27 stanzas. To these we may add a 19th stanza containing ‘the son of Erbin’. Other than the names of Geraint and Erbin, and the single occurrence of the name of Arthur, there are no other personal names in the poem.

Also, it is suspicious that Arthur’s name is used in exactly the same way as is that of Geraint. The variants of the ‘Arthurian’ line are as follows:

“En Llogporth y gueleise Arthur… En llogporth y gueleise y Arthur… Yn llongborth llas y Arthur…”

Professor Patrick Sims-Williams has suggested that to solve the problem posed by the ‘syntactically and semantically ambiguous’ y before Arthur’s name that this line be considered ‘a poetic inversion’ for ‘men to (i.e. vassals of) Arthur’, the ‘men’ in question being the warriors of the following line:

Gwyr dewr kymynynt a/o dur; "brave men, they hewed with steel"

The Red Book of Hergest has instead: "In Llongborth Arthur lost brave men, they hewed with steel"

Of course, the y is in front of Arthur’s name even in the Red Book version. The odd thing about the poem is that Geraint’s name is used in exactly the same context. We have the Black Book of Carmarthen’s:

En Llogporth y llas y Gereint…

Which is, however, rendered in the Red Book of Hergest as

En Llogporth y llas Gereint…

The cumulative effect of the panegyric, with its formulaic repetition of Geraint’s name, and the sudden intrusion of Arthur’s within the same cymeriad, is designed to enable us to see Arthur in this context not as a separate individual, but as an honorific being applied to Geraint.

In other words, just as we find a warrior in The Gododdin compared unfavorably to Arthur, who is there decidedly a famous figure of the past, in the Geraint fab Erbin elegy the heroic nature of Geraint is so great during the Llongporth battle that he symbolically is Arthur, the ‘emperor’ and ‘ruler of battle’.

Those who attempt to account for Arthur’s presence in the poem have in the past resorted to two explanations. First, that Arthur really was there, which would put this particular Geraint back in Arthur’s time, or that a warrior troop whose predecessors had served under Arthur was still, in Geraint’s day, referred to as ‘Arthur’s men’.

There are two problems with these explanations. In the first case, it seems fairly certain that the Llongporth battle is to be identified with the battle fought at Langport by the Wessex chieftains Ine and Nunna against a Dumnonian Geraint in c. 710. This event is memorialized in the ASC, where it is described as a Saxon victory. Needless to say, the 8th century is well outside the time period of Arthur.

That the men fighting with Geraint are composed of a troop whose members originally flocked to Arthur’s standard makes little sense, given that the same ‘brave men’ (gwyr dewr) are ascribed to Geraint:

“In Llongborth Geraint lost [or ‘I saw to’] Brave men from the region of Dyfnaint. And before they were killed, they killed.”

In following Geraint, these warriors were fighting for a chieftain who in the praise language of the poem was an incarnation of Arthur. While it could be argued that Geraint’s fighting alongside Arthur or the latter’s men might be considered praise enough, from the perspective of the panegyrist, whose sole goal was to glorify Geraint, to use Arthur or his men in this fashion would actually have diminished Geraint’s stature. Why would a poet seeking to praise Geraint distract his audience by calling attention to the presence of another, greater hero?

We need only ask this final question: who is greater, a Geraint who by virtue of his martial prowess is literally an Arthur, or a Geraint who needs the help of Arthur and/or Arthur’s men in battle?

The Three Prisons of Arthur

Triad 52 of the Triads of the Island of Britain concerns itself with the ‘Three Exalted Prisoners of the Island of Britain’. After listing the three prisoners, the Triad continues as follows:

"And one [prisoner], who was more exalted than the three of them, was three nights in prison in CAER OETH AND ANOETH, and three nights imprisoned by GWEN PENDRAGON, and three nights in an enchanted prison under the STONE OF ECHYMEINT [Llech Echemeint]. This exalted prisoner was Arthur."

Can we identify these prisons and Gwen Pendragon with known places or personages? Might they have had something to do with the Arthurian battle sites?

Gwen Pendragon has not been identified in the past. Gwen is the feminine form of Gwyn and means ‘the white or fair one’ (later, the ‘blessed one’). It is possible that here Gwen is being used as an eponym for the Guinnion of Castellum Guinnion, an Arthurian battle site in Nennius. Guinnion derives from the same word meaning 'white' plus a locative suffix akin to Latin -ium. As the "holy" Mary is carried on Arthur's shield during this battle, it is not inconceivable that 'Gwen' is a reference to Mary, and that Guinnion itself is being thought of as being named for her.

However, given that one of the famous dragons of Dinas Emrys was white, we should perhaps interpret Gwen as the genius of the Saxons. This white dragon was found by Emrys (or Ambrosius, later identified wrongly by Geoffrey of Monmouth with Myrddin/Merlin) in a subterranean context. This monster’s companion in the ‘Otherworld’ below Dinas Emrys was the red dragon, the genius of the British people. It is my guess that here Arthur is being identified with the red dragon, buried in the prison of the white ‘chief dragon’, a comparable leader of the Saxons.

According to Geoffrey of Monmouth, both Ambrosius, uncle of Arthur, and Uther, Arthur’s father, were buried at Stonehenge next to Amesbury, ancient Ambresbyrig, a site confused in the tradition with Dinas Emrys. Arthur placed in a typical ‘death-prison’ at Stonehenge with his father and uncle may preserve an otherwise lost Welsh tradition which runs counter to the more popular one situating him at Avalon. In passing, I should mention that the white and red dragons are also placed in a pit at Oxford in the MABINOGION tale "Lludd and Llevelys." Oxford is there claimed as the center of Britain, and the Rollright Stones may be the local substitute for Stonehenge.

The Llech or Stone of Echemeint would appear to be a reference to Bath, which the Welsh identified with Arthur’s Mount Badon. According to the ASC (year entry 973 CE), Bath was also known by the name Acemannes-ceaster. This alternate name for Bath appears to be a development from the ancient Romano-British names for the town, Aquae Sulis and Aquae Calidae.

And what about Caer Oeth and Anoeth? Oeth means ‘something difficult to obtain or achieve, a difficulty, a wonder; something strange or wonderful’. Anoeth has essentially the same meaning, as the prefix an- is merely an intensifier: ‘a wonder, something difficult to acquire; something strange or difficult’.

The Caer Oeth and Anoeth placename is also mentioned in the Mabinogion tale Culhwch and Olwen, where it is one of the castles Arthur boasts of gaining entrance to. Once again, in the Stanzas of the Graves we are told that the burial ground of the host of Caer Oeth and Anoeth can be found in Gwanas near Cadair Idris in Ceredigion.

Gwanas is very near the Camlans of Merionethshire. In fact it is exactly between the Afon Gamlan to the northwest and the other two Camlans to the east and southeast. This can hardly be coincidence and probably indicates the “wonderful” grave for Arthur near the fatal battle site, which Welsh tradition (see Appendix III) relocates to NW Wales.

In the Welsh 'Stanzas of the Graves', we are told 'anoeth bin u bedd arthur'. This has been translated in various ways. But some (myself included) have noticed that anoeth in this line may be an oblique reference to both the teulu (house-hold warriors) of oeth and anoeth and Cair ('fort') Oeth and Anoeth. In Triad 52 we are told Arthur was a prisoner in Caer Oeth and Anoeth.

We know where a military force from Caer Oeth and Anoeth ended up: Gwanas, a mountainous region situated exactly between the Welsh Camlanns. They had gone there in order to rob the rich graves of, we must presume, their “anoetheu” or “wonders.”

A detailed investigation of the Gwanas region reveals an interesting candidate for the so-called 'beddau hir' or long graves of the place. The best account of this candidate is found on the COFLEIN site:

"A small square earthwork set upon a ridge summit has been identified as a possible Roman military tower. It is set on the crest of a south-facing ridge, commanding extensive views across the upland basin below Pen-y-Brynnfforchog and the course of the Roman road between Caer Gai and Brithdir.

A range of alternative interpretations can be ad-vanced, notably that this is a Roman or early Medieval square ditched barrow, such as are found at Druid beyond Bala (NPRN 404711), and Croes Faen near Tywyn (NPRN 310263). As such it would, with Tomen-y-Mur (NPRN 89420), be a rare surviving earthwork example, most sites being known only from cropmarks. This monument might be compared to the small practice work at Llyn Hiraethllyn (NPRN 89703), otherwise the smallest example of its type known in Wales.

It is a square platform about 5.0m across with a shallow ditch up to 2.8m across on the south-east, 1.1m wide on the north-east and south-west and not discernable on the north-west. The platform has low banks on the north-east and south-west sides. As a Roman work the earth-work has been associated with a road or track passing below the ridge to the south-east (NPRN 91903), suggested as part of the Roman road be-tween Caer Gai and Brithdir (Rigg & Toller 1983, 165; Britannia XXVIII (1997), 399), although this has been disputed as it is a modern feature (Browne 1986) and is depicted on the 1st edition OS 1" map of 1837 (sheet 59 north-east). A tower on this site would command extensive views of the tributary valley to the south-east, but not of the main Wnion valley on the north-west and the Brithdir military settlement (NPRN 95480) may be out of sight. The earthwork is intervisible with the 'Rhyd Sarn' works 11.5km to the north-east towards Bala Lake (NPRN 303162-3)."

The 'low banks' of this monument (if that is what it really is!) nicely answer for the 'long graves' of Gwanas.

There are no other candidates for the beddau hir. Of course, time and the combined ravages of Man and Nature may long since have destroyed any other such monuments in the region.

But what of Caer Oeth and Anoeth itself? The site has not been identified.

I think, though, that the clue is in the name. Rachel Bronwich in her TRIADS suggests that the fortress was “of difficult access”, which is one possible meaning of the word oeth. Anoeth would in this context merely have an amplified sense. But the important thing to remember here is that anoeth could be a wonderful, strange treasure or Otherworld object. Or the word can refer to the process of obtaining such an object, this being a difficult task to accomplish. Such is made plain in the MABINOGION story “Culhwch and Olwen.” Fiona Dehghani discusses anoetheu in her article “The Anoetheu Dialogue in Culhwch ac Olwen” (Proceedings of the Harvard Celtic Colloquium, Vol. 26/27 (2006/2007), pp. 291-305). Other scholars have discussed the word at some length, including Rachel Bromwich in the notes to her TRIADS.

Arthur is released from the prison of Caer Oeth and Anoeth by one Goreu son of Custennin, who plays an important role in “Culhwch and Olwen.” I believe this point has been overlooked. There is even a point near the end of the tale in which Culhwch and Goreu return to the castle of Ysbaddaden the Giant with the anoetheu. Goreu slays the giant and Culhwch takes the fort, marring the giant’s daughter, Olwen. There is no mention of the anoetheu (here the objects of the difficult tasks the giant had laid upon Culhwch) leaving the place.

It seems fairly obvious to me, therefore, that Ysbaddaden’s castle is Caer Oeth and Anoeth. This is the only fort of anoetheu that we know of, and the only one directly involving Arthur and his men. It is true that Arthur and his warriors raid various Otherworld castles in the poem “The Spoils of Annwm”, but these actions are not defined as anoetheu and we are not told of a castle where stolen Otherworld objects are stored.

Can we determine where Caer Oeth and Anoeth is located? I believe we can. If the fort is, in fact, that of Ysbaddaden, I successfully identified the site years ago (see my book THE MYSTERIES OF AVALON). It is none other than The Wrekin hillfort in Shropshire.

Wrekin is a form of the Romano-British name Viroconium (see Rivet and Smith’s THE PLACE-NAMES OF ROMAN BRITAIN for an extensive discussion of the various forms). The Romans built the city of Viroconium after they had destroyed the hillfort of the Cornovii.

CHAPTER 5

THE NORTHERN KINGDOMS

To give some idea of the political landscape of Arthur’s Britain, it might be helpful to examine some of the “Men of the North” and the kingdoms they controlled.

The most northern of these kingdoms was, of course, the ancient territory of the Votadini or Gododdin, which in the Roman period is believed to have stretched from the Wear or the Tyne through Northumberland and the Lothians to the Forth.

The term ‘Lothian’ appears to have been of Dark Age origin, which as we have seen stands for an original Lleudiniawn, ‘Place of the Fort of [the god] Lugus’. There is an eponymous king recorded in the Life of St. Kentigern called Leudonus, i.e. Lleuddun, and his kingdom in Welsh was known as Lleuddunion. He was supposed to have ruled from Traprain Law, which was earlier called Dunpelder, the ‘Fort of the Spear (shaft)’.

In the late 6th century, the king of the Votadini was, apparently, one Mynyddog Mwynfawr. He is said to have ruled from Din Eidyn or Edinburgh and was the son of a certain Ysgyran, and probably succeeded Clydno Eidyn. The Gododdin poem implies that the Britons who fought the English at Cattraeth assembled at Mynyddog’s court at Edinburgh. Clydno Eidyn, in turn, was the son of Cynfelyn son of Dyfnwal Hen. Myynyddog is also given the epithet ‘Eidyn’ meaning, undoubtedly, ‘of Eithne’. Once again, Eidyn is likely the British form of Eithne, mother of the god Lugh in Irish tradition.

Pabo Post Prydain, the ‘Pillar of Britain’, is the son of Ceneu son of Coel Hen, both famous chieftains of the North. Pabo is spelled Pappo in the genealogies appended to the HB. Coel Hen’s name is believed to be preserved in Kyle in Ayreshire.

A son of Pabo is Dunod Fwr, who is probably the chieftain who fought against the Rheged princes in Erechwydd, which itself is usually placed somewhere in Cumbria. We may relate this Dunod to Dent in Northwest Yorkshire, his lands here being termed the ‘regio Dunotinga’, kingdom of the descendents of Dunod. From John Morris’s The Age of Arthur:

“DENT: regio Dunotinga is one of four districts of north-western Yorkshire overrun by the English in or before the 670s, Eddius 17 [Life of Wilfrid]. The passage is overlooked in EPNS WRY 6, 252, where the early spellings Denet(h) are rightly related to a British Dinned or the like, and Ekwall’s derivation from a non-existent British equivalent of the Old Irish dind, hill, is properly dismissed. EPNS does not observe that Dent was, and still is, the name of a considerable region, and tha thte village is still locally known as Dent Town, in contrast with the surrounding district of Dent…. Regio Dunotinga plainly takes its name from a person named Dunawt, Latin Donatus, as does the district of Dunoding in Merioneth, named from another Dunawt, son of Cunedda.”

The regio Dunotinga was associated with the Ribble and other places in the north of the West Riding. As the Dent River is a tributary of the upper Lune in Lonsdale, and Upper Lonsdale seems to have been within the canton of the ancient Carvetii tribe, it is likely that Dunot was himself descended from the ‘People of the Stag’. The Carvetii (see Cerwyd/Cerwydd below) ruled over what we now think of as Cumbria and adjacent areas.

Bran son of Ymellyr[n] is associated with both Dunawt of Dent and Cynwyd of Kent (see below for the Cynwydion). The patronymic here is transparently from Old Norse a, river, plus melr, sandbank, identifying his region with Ambleside in Cumbria just to the west of the River Kent.

The 'Drws Llech' or 'Door-stone' of Sawyl's Brother, Dunod

Dunod Fwr (Bwr), brother of Sawyl, my candidate for the father of the famous Arthur, is associated with five places. I have dicussed the regio Dunutinga above. I have also treated of Erechwydd (= Eamont in Cumbria; see Chapter 5), a place also associated with Urien of Rheged. Bran son of Ymellyr[n] is associated with both Dunawt of Dent and Cynwyd of Kent. Ymellyr is transparently from Old Norse a, river, plus melr, sandbank, identifying his region with Ambleside in Cumbria just to the west of the River Kent. [1] And, yes, the use of English or Norse names in the context of figures belonging to the sub-Roman or early medieval period is horribly anachronistic. But this does happen in the heroic poetry, demonstrating pretty obviously that some of the traditions were recorded quite late.

Not generally mentioned in the context of Dunod's homeland is the fact that he supposedly married the daughter of Lleenog of the Kingdom of Elmet.

But one place has remained a mystery: Drws Llech. From P.C. Bartrum's A CLASSICAL WELSH DICTIONARY:

"DUNOD FWR ap PABO POST PRYDYN. (500, d.595). He is mentioned in a triad (TYP no.5) as one of the ‘Three Pillars of Battle’ of Ynys Prydain, and in another triad (TYP no.44) he is mentioned as one of the seven who rode the horse Corfan to view the battle-fog of [the host of] Gwenddoleu at Arderydd. This was in the year 573 (AC). His pedigree is given in HG 11, ByS §12, BGG §4 in EWGT pp.11, 56, 73. The cognomen Fwr, ‘fat’, occurs in TYP no.44 and ByS §12, but becomes Fawr in one late version of BGG §4 and Achau'r Saint §§21, 51 in EWGT pp.70, 71. He was the father of St. Deiniol by Dwywai ferch Lleenog (ByS §12). In the ‘Hoianau’ in the Black Book of Carmarthen he is called Dunaud deinwin, ‘Dunod white teeth’, the father of Deiniol (BBC 56 l.1). In Annales Cambriae s.a.595 we find: Dunaut rex moritur, but MS.B reads: Dunauut filius Pabo obiit. It seems probable that the identification is correct, but it must be supposed that he died at a good age, especially as his son, Deiniol, is recorded as dying in 584. Dunod is mentioned in a poem on the death of Urien in the ‘Llywarch Hen’ Poetry (CLlH III.3). Llywarch Hen is represented as saying: Let savage Unhwch guide me; It was said in Drws Llech, ‘Dunod ap Pabo does not retreat.’ Further on in the same poem it is probably the same person who seems to be described as making war on Owain and Pasgen, sons of Urien: A WELSH CLASSICAL DICTIONARY 236 III.37 Dunod, horseman of the chariot, planned to make a corpse in Yrechwydd against the attack of Owain. III.38 Dunod, lord of the land, planned to make battle in Yrechwydd against the attack of Pasgen. Another poem tells how Llywarch Hen, after the death of Urien, was living in a state of poverty and was advised by a friend to migrate to Powys. The friend says: V.5 Trust not Brân, trust not Dunod. The location of Dunod's family may be represented by the regio Dunutinga, which was presented to the church of Ripon in about 675 (Eddius's Life of Wilfrid, Ch.17). It is associated with the Ribble and other places in the north of the West Riding (H.M.Chadwick, Early Scotland, p.143 and n.3). The place is represented by modern Dent which is the name of a considerable region surrounding a village known as Dent Town [11 miles West by South of Hawes] (John Morris, The Age of Arthur, 1973, p.573). Dunod's bard was perhaps Cywryd (q.v.). FICTIONS Geoffrey of Monmouth included Dunod ap Pabo among the princes who were present at Arthur's coronation (HRB IX.12), similarly Brut y Brenhinedd. Two other sons of Dunod Fwr are mentioned in the Iolo MSS. p.126, namely Cynwyl and Gwarthan. For the origin of these fictions, see s.n. Gwarthan."

Ifor Williams, the editor of the Llywarch Hen poetry, has this on Drws Llech:

3b drws llech. Enw rhyw fwlch yn y mynyddoedd Ile bu brwydro ? [Translation: "The name of a gap in the mountains where there was a battle?"] Ar llech gw. P.K.M. 303 ; drws, cf. Drws y

Coed, y bwlch rhwng Rhyd Ddu a Nantlle : Drws y Nant,

ger Dolgellau : Bwlch Oerddrws, rhwng Dolgellau a Mallwyd ; (Llywelyn ap Gruffudd), M.A. 240a. Nyt oet hawt y dreissyaw ger drws deuvynyt ; E. Ll. 24.

Thus, he believed that the place was a bwlch/fwlch, a pass or defile. And this is despite the fact that the primary meaning of drws is 'door.'

Llech presents us with a bit of a problem, for it is not a generic term for stone or rock. Here is the word's definition in the GPC:

slate (in geol.); roofing-slate; writing-slate; bakestone, griddle; (rectangular) slab of stone, flooring- or paving-stone, flag-(stone); gravestone; rock, boulder, cliff.

Brittonic place-name expert Alan James had some interesting information to impart on the geology of the Yorkshire and Cumbria mountainous terrain, and also kindly commented on some my ideas for Drws Llech:

"Drws llech is an anomalous formation - is it a loose compound, 'door-slab'? Or a phrasal one, 'door/pass with a slab'? We'd expect *drws y lech for the latter. Llech is ‘A slab, a slate, a flat stone’, in place-names also 'a shelf of rock' - see BLITON.

Lech is indeed used for grave-slabs - horizontal in medieval times, upright gravestones later, and by metonymy for 'a grave'. I think your idea that the name might have referred to some prehistoric structure is a good one, though I don't think Cumbric speakers would necessarily have recognised that some of them were remains of 'graves'. The fact that llech doesn't have a plural inflection may not rule out a 'door' with a pair of slabs, the modern plural forms, llechau, llechi, are analogous formations, probably not used in Cumbric.

Your point about limestone pavements is a good one too. Such 'door-like' passes as I can think of in the Pennines, Yorkshire Dales and Peak District are almost all on limestone - not necessarily with pavements, but characteristically bedded with large slabs. In the Lakeland fells, more typically igneous rocks, with slate slabs.

There aren't many 'door-like' passes through the Pennines, or the Cheviots. 'Doors' with 'slabs' are more typical of the Lakeland fells.

I think drws implies a fairly narrow gap between high, rocky sides. The distribution of llech reflects geology, much more likely in places where there are bedded layers that weather or break easily into flattish slabs."

https://getoutside.ordnancesurvey.co.uk/guides/the-welsh-origins-of-place-names-in-britain/ has for drws:

gap, narrow pass

James at https://historicplacenames.rcahmw.gov.uk/ adds that drws in the Welsh landscape means

"Technically, drws means the opening itself, the slab of wood on hinges with which you close the opening is a dôr, so the term also applies to physical features."

In https://spns.org.uk/wp-content/uploads/2019/03/Alan_James_Brittonic_Language_in_the_Old_North_BLITON_Volume_II_Dictionary_2019_Edition.pdf, a Truss Gap in Cumbria is listed:

drus (m)

IE *dhwōr- (ō-grade of *dhwer- 'pierce') + -est- > eCelt *durestu- > Br *drustu- > OW drus > MMnW drws, cf. MCorn darat > Corn daras; O-MnIr, G dorus, Mx dorrys; ?cf. Lat foras 'out of

doors', fores 'double doors'.

The precise history of the Celtic forms is ‘thoroughly obscure’ according to P. Schrijver (quoted

in EGOW at p. 51). They exist alongside the more regular development eCelt *durā- > OW dor >

W dôr, MCorn dor, Bret dor, OIr dor, cogn. Lat foris ‘outside’, OE dor > ‘door’ (also OE duru >

northern ME/ early Scots dure), Gk thúrā-, Skt dvarau, and ‘in all major Indo-European groups’,

OIPrIE §72 at p. 108, and see also DCCPN p. 18.

‘A door, doorway, gate, gateway’. It occurs in later Welsh place-names and in early Modern

Welsh literature in the sense of ‘a narrow gap or pass’, but its presence in earlier Welsh

toponymy is not certain. For Irish and Scottish Gaelic examples, see DUPN p. 59 and PNFif5 p.

356.

Whaley (2001), pp. 77-96, and in DLDPN pp. 348-9, argues for this element in the following, but

see also *trǭs:

a1) Truss Gap Wml (Shap) PNWml2 p. 178, DLDPN p. 349 and plate 2.

a2) Trusmadoor Cmb (Ireby) DLDPN pp. 348-9 and plate 1 (not in PNCmb) + -μa [+ OE –dor

‘door’].

As it happens, this Truss Gap is located pretty much exactly between Dentdale and Eamont (Erechwydd).

But even better, there is a Trussgap Brow in this place. Trussgap is, of course, a tautology (with English gap reproducing the meaning of Cumbric truss/drws). Brow here is for the crag, which itself sports cliffs. I take this for Drws Llech, given that llech could mean a cliff.

Perhaps significantly, the place-name Swindale (according to Victor Watts in his THE CAMBRIDGE DICTIONARY OF ENGLISH PLACE-NAMES) means 'valley where swine are raised.' This finds it echo in the name of the man associated with Drws Llech in the Llyward Hen poetry - Unhwch. According to Dr. Simon Rodway of The University of Wales, Unhwch means 'unique pig.'

Alan James' comment on the site?

"GPC gives clogwyn as the last of the senses for llech, though even that word can mean a steep crag. I rather doubt whether llech would have been used in Cumbric for cliffs in general, though it's obvious how the sense could extend by metonymy. And the crags and screes on either side have plenty of slabs still attached or fallen off - see the photo of Gouther Crag for example: there's a huge 'slab' on that edge, and the gap between Gouther and Outlaw Crags might well be a drws."

The last word on the subject is courtesy Diana Whaley, from her article on Cumbris pass place-names in NOMINA:

https://www.snsbi.org.uk/Nomina_articles/Nomina_24_Whaley.pdf

"The element drws/trus `door, pass' is extremely rare outside Wales, to judge from the standard authorities,9 though it is noted by Coates, who suggests an antecedent, in the plural, of the Welsh oerddrws `wind-gap', literally `cold-door' as the etymon for the first part of Overmoigne, Dorset (SY7685).10 It is, however, the key that, as well as opening Trusmadoor, explains another problematic Lakeland name, Truss Gap. The farm of this name is situated in the former Westmorland, at a notable narrowing in Swindale, three miles south of Bampton (NY5113). Truss Gap is listed, as Trussgap, in the English Place-Name Survey for Westmorland, where its appearance on the First Edition One Inch OS map of 1859 is noted, but no explanation of the name given.11 I have not found significant pre-datings, though the name is also recorded from 1728. 12 Nevertheless, the coincidence of topography and the collocation of the drws-like Trus- with an element meaning `gap, pass' puts it beyond reasonable doubt that this is another Cumbric `door-way, pass'."

[1]

From P.C. Bartrum's A CLASSICAL WELSH DICTIONARY:

BRÂN ab YMELLYRN. (Legendary). He is mentioned in a poem on Urien Rheged put into the mouth of Llywarch Hen (CLlH. III.40, p.17 and notes): Brân ab Ymellyrn planned to exile me, and burn my houses: A wolf howling at the door(?)! Another poem tells how, after the death of Urien, Llywarch Hen was living in a state of poverty and was advised by a friend to migrate to Powys. The friend says (CLlH. V.5, p.22 and notes): Trust not Brân, trust not Dunawd; Consort(?) not with them in hardship. Herdsman of calves, go to Llanfawr. Gruffudd Hiraethog found that Brân ab Ymellyrn was identified with Brân Galed (q.v.). In Peniarth MS.176 p.185 he wrote: Kynan ap Bran Galed ap Emellyr ap Kynwyd Kynwydion, a hwnnw oedd Bran Galed yn gynnar ac a elwid wedi hyny Bran Ewerydd. Hen Llyfr Bodeo[n]. 'Bran Ewerydd’ seems to be an attempt to identify the same Brân with Bran mab Ywerit [Brân ab Iwerydd] of a poem in the Black Book of Carmarthen. See s.n. Iwerydd Glyn E.Jones, in BBCS 25 pp.105-112, discusses the possible identification of Brân ab Ymellyrn, Brân [Hen] ap Dyfnwal [Moelmud] (see Dyfnwal Moelmud (2)), and Brân ab Iwerydd, but A WELSH CLASSICAL DICTIONARY -59- comes to no definite conclusion. He points out that a certain Brân was ‘at Cynwyd’ (Bran yg Kynwyt) according to ‘Gwarchan Tudfwlch’ in Canu Aneirin, l.1291. Also in CLlH VII.17 a battle of Cynwyd is mentioned in connection with Pelis, a soldier of Urien Rheged. Brân ab Ymellyrn may be referred to here.

Another son of Pabo’s is Cerwyd or Cerwydd, who is otherwise completely unknown. This name is transparently an eponym for the Carvetii tribe. We have just seen that Dunod’s Dent seems to have been a part of the territory once covered by this ancient tribal kingdom.

The form Cerwydd as a direct eponym for the Carvetii is not possible; we would need Cerwyd for an exact linguistic correspondence. However, as Cerwydd means ‘stag-like one’, we can say with a fair degree of certainty that he does represent the People of the Stag.

As for Pabo, father of Dunod, we may situate him at Papcastle (Pabecastr in 1260), the Derventio Roman fort in Cumbria. Pap- is thought to be from ON papa, papi, for ‘hermit’, but this seems an unlikely name for a ‘ceaster’. Instead we should look to early W. pab, ‘pope’, i.e. papa, pl. pabeu, and Llanbabo church of St. Pabo in Anglesey. Pabo's Chester would seem to do quite nicely. We could then locate Pabo within the Carvetii kingdom of his sons Cerwyd/Cerwydd and Dunod.

I would add that Pabo’s epithet ‘Post’ or ‘Pillar’ is possibly a reference to the Solway, which is believed to be from OScand. sul, ‘pillar or post’, and vath, ‘ford’. It has been proposed, quite reasonably I think, that the pillar or post of the Solway is the Lochmaben Stone at Gretna Green. A ‘papa’ or ‘father’ of the post/pillar named for the Divine Son Mabon makes for an interesting combination of place-name elements!

However, it is true that the Papcastle fort is not on the Solway. The name of the Roman period fort here – Derventio – was named for the river Derwent, the ‘oak-river’ or ‘river in an oakwood’. As the oak was a very sacred tree to the early Celts, it is possible the ‘post’ or ‘pillar’ that gave its name to Pabo was an oaken one and thus an indirect reference to the place-name.

Sawyl Benisel ("Low-head"), yet another son of Pabo and my candidate for Arthur’s father, is dated c. 480 CE. On the Ribble, not far south of ‘regio Dunotinga’, is a town called Samlesbury. The place-name expert Eilert Ekwall has Samlesbury as ‘Etymology obscure’, but then proposes OE sceamol, ‘bench’, as its first element, possibly in the topographical sense of ‘ledge’. A.D. Mills follows Ekwall by saying that this place-name is probably derived from scamol plus burh (dative byrig). However, sceamol/scamol is not found in other place-names where a ‘ledge’ is being designated. Instead, the word scelf/scielf/scylfe, ‘shelf of level or gently sloping ground, ledge’ is used.

I would suggest as a better etymology for Samlesbury ‘Sawyl’s fort’ (see Chapter III above). There are, for example, Sawyl place-names in Wales (Llansawel, Pistyll Sawyl, now Ffynnon Sawyl). Sawyl is the Welsh form of the name Samuel.

Dr. Andrew Breeze of Pamplona, a noted expert on British place-names, agrees with this proposed etymology:

“I feel sure you are right. The form surely contains the Cumbric equivalent of Welsh _Sawyl_<_Samuel_. Your explanation of this toponym in north Lancashire is thus new evidence for Celtic survival in Anglo-Saxon times.”

Now that we have placed Pabo and his descendents on the map, we need to investigate what has been explained as an intrusion on their pedigree.

An Arthwys and his father Mar are both inserted into the Pabo genealogy. Instead of Pabo son of Ceneu son of Coel Hen, we have Pabo son of Arthwys son of Mar son of Ceneu, etc. This same Arthwys is made the grandfather of a Cynwyd of the tribal group known as the Cynwydion (of the Kent river in Cumbria - Kent being from Kennet, which in Welsh is Cynwyd), of Gwenddolau of Carwinley (Caer Gwenddolau just a little north of Carlisle) and father of Eliffer (Eleutherius) of York. Eliffer in another pedigree is the son of Gwrgwst Ledlum (Fergus Mor of Dalriada) son of Ceneu son of Coel Hen.

Mar is made the father of Lleenog, father of Gwallog of the kingdom of Elmet (a small kingdom centreed about Leeds, probably from Welsh elfydd, ‘world, land’), but in another pedigree it is Maeswig Gloff, i.e. Maeswig ‘the Lame’, who is father of Lleenog.

Mar is an attempted eponym for the Mor/Mer-ingas of Westmorland.

Maeswig (Masguic Clop in the Harleian genealogies) as a name appears to be from *Magos-vicos, ‘Fighter of the Plain’. He appears to be a sort of personification for the broad plain that is the Vale of York.

At least one Northern chieftain was placed at a Roman fort. Pabo Post Prydain belongs at Papcastle, west of Burrow Walls on the River Derwent.

The name Arthwys has frequently been brought into connection with that of Arthur/Artorius. This name is from Arth-, ‘Bear’, + -wys. Dr. Simon Rodway of The University of Wales tells me that

“There is an element –wys found in a number of words of obscure meaning and derivation which could be present in Arthwys, cf. doublets like mam ~ mamwys, neuadd ~ neuaddwys (Ifor Williams, The Poems of Taliesin, trans. J. E. C. Williams (Dublin, 1968), p. 51).”

Professor Paul Russell of The University of Cambridge states that:

“Most of the other forms in -wys derive from Latin -ensis (thus also Powys) and that is what it is in mamwys, etc.” [For the territorial suffix –wys, see below.]

To which Professor Richard Coates of The University of West England adds:

"The accepted etymology valid in other names, or words derived from names, viz. < Latin -enses, is beyond dispute."

The best example of such a name is Glywys, the Welsh equivalent of Glevensis, ‘people of Glevum’, i.e. Gloucester. Glywys is thus merely an eponym for people who traced their origin to Gloucester. In this, sense, then, Arthwys would be for *Artenses, ‘people of the Arth/the Bear.’

Dr. Andrew Breeze has made a case for the river Irthing containing the word Arth, ‘bear’. From his article “Celts, Bears and the River Irthing” (Archaeologia Aeliana, 5^th series, volume XXXII):

Irthing, which has early forms Irthin, Erthina, and Erthing, would also make sense as ‘little bear’, with a Cumbric diminutive suffix corresponding to Middle and Modern Welsh –yn (Old Welsh –inn), as in defynyn ‘droplet’ from dafn ‘drop’ or mebyn ‘young boy’ from mab ‘boy’. As the th of Arth is pronounced like that of English bath, but that of Irthing like that of brother, the process of voicing here would take place after borrowing by English, not before.”

The claim has been made that Arthwys should be Athrwys, as this spelling is found in later sources. The argument would seem to have some support as the name Athrwys is found in Wales. If it was Athrwys, the first element would be W athro 'teacher' (< PIE *pH2tro:w- ‘uncle’). However, as Professor Ranko Matasovic has pointed out to me via private correspondence, while we have plenty of examples of Arth- or bear names, other than the presumed Athrwys, we have absolutely no other extant names containing athro.

A discussion of the -wys suffix can be found in John T. Koch's Celtic Culture, among other sources. Regedwis, for example, is 'people of Rheged' - or maybe better, 'inhabitants of Rheged'. The entry for -wys (1) in the University of Wales Dictionary confirms it as a Latin borrowing and as a nominal plural ending, giving the examples of Gwennwys, Lloegrwys and Monwys.

Dr. Delyth Prys of the place-name experts at The University of Wales, Bangor told me via personal correspondence that: “I've no independent evidence for this, but river names are sometimes used as the name for a more general area and by extension it could be the people of the Arth (area)." Now, if the Irthing is not from ir-t, but from erth/arth +inga (belonging to, not descendents of), it would be the 'tun belonging to Arth' or belonging to the bear. But if the river itself were originally the Arth/Erth, then the tun itself would belong to the river.

Alan James of BLITON states that river-names can sometimes be also the names of adjacent regions, or - probably more correctly - some river-names may have originally have been regional names (or vice versa). This may have been the case with Llwyfenydd/ Lyvennet of Urien. The kind of river-names that seem to double as district names tend to be ones that refer to local terrain, etc., but that may just because such topographical names are more obviously linked to the area. Again, rivers were sometimes boundaries, but they're as likely to flow through a territory perceived as one as to divide such a territory into two. A hypothetical Arth/’Bear’ region could have included both the Irt and the Irth of Irthington, not necessarily been bounded by them.

Arthuret near Carwinley was the scene of a great battle featuring Myrddin, among others. In Welsh this place was called ‘Weapon-fierce’ (https://digitalcommons.brockport.edu/jlo/vol2/iss1/1/), Armterid/Arfderydd, the British name for the river which now bears the English designation Liddel. The actual site of the battle was Liddel Strength between Carwinley and Liddel at the confluence of the Liddel and Esk. We know this because in “Lailoken and Kentigern” Carwinley is called Carwannock and the battle is said to take place between the latter and the Lidel on a plain. I had proposed that -wannock was either derived from Cumbric gwaun, “high and wet level ground, moorland, heath; low-lying marshy ground, meadow” or might be a hypocoristic form of Gwenddolau. Gwenddolau itself looks to be a place-name, as it means, literally, “White dales” (dol being “meadow, dale, field, pasture, valley”). Brythonic place-name expert Alan James confirmed both possibilities for me:

“The meaning of derivatives of *wāgnā in the Brittonic languages is primarily ‘level, marshy ground’, whether upland or lowland; developments include gwaun ‘a meadow’ in Welsh, goon ‘downland, unenclosed pasture’ in Cornish. Br -āco-/ā-> -ǭg is an adjectival and nominal suffix, indicating ‘being of the kind of’, ‘association with’, ‘abounding in’, the stem-word. It occurs very widely in river-names, hill-names and other topographic names. It 's not diminutive, though in hypocoristic personal names like Gwennock it might be affectionate.”

Modern Arthuret somewhat to the south appears to be a relocation of the name.

Dreon son of Nudd, another famous hero at the Arthuret battle, is likely a son of the Nudd mentioned on an early 6^th century tombstone at Yarrow Kirk.

Not far west of the Carwinley of Gwenddolau on the coast of Galloway is the fort of Caerlaverock. The name of this fort is referred to in Welsh tradition as the ‘Lark’s Nest’ and it is said to have been the cause of the Battle of Arfderydd. But ‘lark’ is itself either a mistake or pun for the personal name Llywarch, in this case Llywarch Hen son of Elidir Lydanwyn. Llywarch was first cousin to Urien Rheged. Caerlaverock is, therefore, Caer Llywarch.

There is another interesting reference to a place in Cumbria that I might mention. In the ‘Cambridge’ group of Historia Brittonum MSS., an interpolation tells us that Vortigern is said to have built “Guasmoric near Carlisle, a city which in English is called Palme castre.” Palme castre has long been erroneously identified with the Old Carlisle Roman fort one mile south of Wigton in the parish of Westward. There is a double-error in the Historia Brittonum, for Guasmoric itself is not the same place as the Palme castre fort.

Guasmoric must be Gwas Meurig, the “Abode of Meurig or Mauricius.” This is clearly an attempt at rendering the Gabrosentum Roman fort in Cumbria at Moresby. According to both Ekwall and Mills, Moresby (Moriceby, Moresceby) is Maurice’s By, Maurice being a Norman name and -by being Old Scandinavian for “farmstead, village, settlement”. Whether we can propose an original Welsh Meurig underlying Maurice is questionable. In all likelihood, the interpolation is late and Guasmoric represents Maurice’s By. If originally a Meurig place-name, this may commemorate the 6th century Meurig son of Idno son of Meirchion, who married a daughter of Gwallog of Elmet.

As for Palme castre, this is a place now called Plumpton (Plumton, ‘tun where plum trees grow’; see Ekwall) in Cumbria. Directly between Plumpton and Plumpton Foot is the Voreda Roman fort. Rivet and Smith (The Place-Names of Roman Britain) list the fort as being “at Old Penrith, Plumpton Wall, Cumberland, beside the river Petteril”. Voreda means ‘horse’ in British.

As archaeology has shown us, there were two main centres for the Carvetii kingdom. One was the ancient tribal centre near Brougham, the Roman Brocavum, with its triple sacred henges at Eamont. One of these henges is actually called King Arthur’s Round Table and another the Little Round Table. There is evidence in the form of a concentration of inscriptions at Brougham that the primary Carvetii deity worshipped at these henges was a horned god (doubtless a stag, given that Carvetii means ‘People of the Stag’) named Belatucadros.

But there was also an important region called variously Erechwydd or Yr Echwydd, mentioned in connection with Urien, his sons, Gwallog son of Lleenog of Elmet and with Dunod Fwr. No wholly satisfactory identification of Erechwydd has yet been made, but it would seem to be somewhere in or close to Cumbria.

What we do know about Erechwydd is that the Er- prefix is not the definite article yr, even though the name is sometimes wrongly written ‘yr echwyd’ in the poetry, but a form of Ar-, as found in other place-names, e.g. Arfon. Ar- as a prefix originally meant ‘in front of’. But it came to have the senses of ‘upon, on, over, at, in, across from’.

The National Dictionary of Wales defines echwydd as ‘fresh (of water, as opp. to salt); fresh water’. However, although this meaning has been extrapolated from the contexts in which the word is used, no good etymology had yet been proposed.

I asked Graham Isaac if the word could come from ech, ‘out of, from’, plus a form of the Indo-European root *ued, ‘wet’. His response was:

“The etymology echwydd < *exs-wed-yo-, or *exs-ud-yo- (either would probably do it) seems plausible enough.”

The literal meaning would then be the ‘out-water’, but the sense of the word would be simply ‘flowing, fresh water’. Again, the Welsh texts which use this word leave no doubt that we are talking about fresh water emerging from springs or lakes.

So where was Erechwydd/Yr Echewydd, the ‘Place by the flowing, fresh water’? Our clue lies not only in the name of the region, but in the battles fought there between Dunod Fwr of the Dent region and Gwallog of Elmet against Urien’s sons. These engagements are recounted in the Llywarch Hen poetry. Given that Urien Rheged seems to have had his origin in the tribal territory of the Novantae, and to have been lord of Annandale; see my book THE MYSTERIES OF AVALON), and both Dunod and Gwallog had kingdoms in southeastern Cumbria and just southeast of Cumbria, respectively, the most logical place to seek Erechwydd would be the twin valleys of the Eden and Petteril.

A Roman road led from the south up through the valley of the river Lune right past Dunod’s Dentdale. This road continued north to the Eden Valley. Another Roman road led west from Leeds and joined with the Lonsdale road. Gwallog could have taken this route to the Eden or he could have gone north up Dere Street and then cut over through the Pennines at Stainmore.

The Eden and Petteril Valleys were the heartland of the ancient Carvetii kingdom. Not only did the twin valleys provide the obvious natural route from Carlisle towards Lancaster and York, the area has been shown to have supported a widespread and occasionally dense pattern of rural settlement in the Roman period.

It is even possible that Erechwydd as a regional designation can be more precisely localized within the Eden and Petteril Valleys. The headwaters of the Petteril lie just west-northwest of Eamont. We have already discussed the importance of Eamont with its sacred henges. The river Eamont (a back-formation from the name Eamont itself, from AS ea-gemot, ‘river-meet’, i.e. confluence) and Lowther join at Eamont Bridge and continue for a short distance eastward to the Eden. There was also, of course, a nexus of Roman roads at Eamont.

In my opinion, the Anglo-Saxon place-name ea-gemot/Eamont may overlie an original British Echwydd. Ekwall thought Eamont refers to the confluence of the Eamont and the stream from Dacre, although given the location of the Brougham/Brocavum Roman fort at the juncture of the Eamont and Lowther, it makes much more sense to see this ea-gemot as the confluence of the latter two rivers. If I am right, then Arechwydd was the Eamont area, specifically the land at and around the Brougham fort and the three Carvetii henges. The ‘out-water’ would be a reference specifically to the Eamont, which is formed by the outflow from the Ullswater, the second largest lake in Cumbria.

Just a few miles south-southest of Eamont is the Lyvennet Beck, a tributary of the Eden. This has been identified with the Llwyfenyd over which Urien is said to have been ‘ruler’ (Welsh teithiawc).

'Eil Mehyn' of Urien Rheged

In THE BOOK OF TALIESIN (VII, 61), a number of battles are listed for Urien Rheged. While some remain unidentified, we do know the locations of others. In this blog post I would like to offer a tentative identification for 'yn eil mehyn'.

This place is listed between that of the Lyvennet Beck in Cumbria, the ford of Alclud in Strathclyde (although Alclud makes more sense as Auckland in Durham; see https://spns.org.uk/wp-content/uploads/2020/05/Alan_James_Brittonic_Language_in_the_Old_North_BLITON_Volume_II_Dictionary_2020_Edition.pdf) , the Manor Water in Scottish Borders and Bremenium/High Rochester in Northumberland. We might, then, reasonably expect for 'eil mehyn' to be somewhere in Lowland Scotland, and preferably between the other battles - assuming, of course, we can allow for such a geographical ordering of the various engagements.

The other battles of Urien can be pretty firmly identified:

Pencaitland

South Low (his death site)

Catlow near Yeadon

River Leven

Catterick

Lyvennet

Arkle Beck/Arkengarthdale (see below under Gwen Ystrad)

On the place-name designation (or description) 'eil mehyn', here is Ifor Williams' note:

It is possible 'eil mehyn' as the fenced/palisaded place may not be a real place-name; it may instead be merely a descriptor. Some weight is given to this possibility, given that the precededing location is 'eidoed kyhoed', the 'conspicuous defenses.' Or both eil mehyn and eidoed kyhoed may be one and the same site!

Let us, though, assume for the sake of argument that eil mehyn is an actual fortified site.

Now, opting to take this angle is interesting, for one of the seriously proposed etymologies for the Eildon Hills, with its Eildon Hill North fort (the largest in all of Scotland), is an English dun preceded by a Cumbric word meaning palisade or fence. The early forms for Eildon are as follows (see http://spns.org.uk/wp-content/uploads/2017/07/MayWilliamsonComplete.pdf):

Ældona, 1119-24 (c 1320) Kelso; Eldune, 1143 LSMM; 1166-70 BM; -doun, c 1153 (16th) Dryb; -dunum (acc) 12th SD; -dun, c 1208 BM; Eladune, c 1150 C de M.

This discussion of the name Eildon is from Alan James, noted expert of place-names in Northern Britain (see http://www.clanntuirc.co.uk/JSNS/V7/JSNS7.pdf, as well as https://spns.org.uk/wp-content/uploads/2019/03/Alan_James_Brittonic_Language_in_the_Old_North_BLITON_Volume_II_Dictionary_2019_Edition.pdf):

"It might include a plural form *eil of *al, which Watson 1926, 32‒33, saw as a cognate of Old Irish ail ‘rock’), alternatively *eil ‘wattle fence or structure’, but Old English ǣled ‘fire’ or ǣlǣte ‘desert, empty place’ (+ OE -dūn ‘hill’) are also good possibilities."

I would propose, therefore, that 'the place of the fence or palisade' is a Welsh rendering of Eildon, and that this is where Urien won a major victory, presumably against the English. [1] Because the Eildons is not only the most prominent landmark in the Scottish Lowlands, but also sports Scotland's largest hillfort on Eildon North Hill, it seems unlikely that the Welsh panegyricists would have been ignorant of the site.

NOTE: The fascinating thing about the Urien map is that it shows a war leader a generation after Arthur fighting the English along pretty much exactly that same North-South line.

[1] Another possibility for the 'fence place' would be the Catrail dyke in the Scottish Lowlands. The name (see again Alan James) has been etymologized with ail, 'palisade', as its second component. From James' BLITON:

*eil (m) eCelt *al-jo- > Br *aljo- > M-MnW ail, eil; OIr aile > (in compounds) Ir, G –aile, Mx –ayl. The Celtic root *al- is associated with weaving, and with the construction of fences, buildings, etc using woven wattles. So Welsh eil is ‘a shelter, a shed’, Old Irish aile ‘a fence, a palisade’, Irish/Gaelic buaile, Manx boayl, ‘a cattle-fold’. Williams, PT pp. 85-6, saw this element in Alclüd, suggesting that it referred to wattle-built defences both here and at the unlocated Eil Mehyn BT61(VII), but see also *al and alt. a1) Eildon Hills Rox PNRox pp. 7 and 40 [+OE –dūn ‘a hill’], but see discussion under *al. b1) The Catrail Slk CPNS p. 181 ? + cad- ["battle"] + analogical –r- (for ‘erroneous’ -ï[r]-, cf. CPNE p. 7 and, for similar cases in Gaelic toponymy, SPN² p. 161). A discontinuous series of earthworks crossing upper Tweed, Yarrow, Ettrick and Teviot dales; both its archaeology and its etymology are obscure. c2) Potrail or Powtrail Water Lnk (a headwater of the Clyde) CPNS p. 181n2 ? + *polter-.

If this great ditch served a military purpose, it may be where Urien fought.

https://canmore.org.uk/search/site?SIMPLE_KKEYWOR=Catrail

Urien's Gwen Ystrad and Llech Wen Battle Site

Several attemtps have been made to identify the site of Urien of Rheged's great battle of Gwen Ystrad, the 'White Strath'. None have been particularly successful. Why? Because the candidates proposed by various scholars have etymological origins which do not allow them to be equated with Welsh Gwen Ystrad. It is necessary to instead assume that W. gwen, 'white', is being employed to replace a similarly spelled place-name that derives from another language and which has an entirely different meaning.

Wensleydale, for example, is derived from an OE personal name *Waendel (Mills, Ekwall, Watts). Winster may be from Welsh Gwensteri ('white stream'), but according to J. Lloyd-Jones (cited in the notes to THE POEMS OF TALIESIN by Sir Ifor Williams), "that this is a reference to Gwenystrad is very improbable." In addition, steri is found only in Breton. Gwensteri is mentioned in a poem concerning Gwallawc of Elmet. Williams observes that "Across the mountain passes the river [Winster] was very near the old Elmet." Other guesses as to the location of Gwen Ystrad lack etymological and/or geographical plausibility. I do not find Andrew Breeze's treatment of the place-name in his "British Battles 493-937: Mount Badon to Brunanburh" at all convincing. His argument for Gwen Ystrad = Gwensteri is poorly constructed and, in my opinion, wholly invalid.

Gwen Ystrad may, in fact, be related to the Llech Wen battle site found in the same Taliesin poem (see "The Triumph Tree: Scotland's Earliest Poetry, AD 550–1350," ed. by Thomas Owen Clancy, Edinburgh: Canongate, 1998, pp. 79-80). Llech Wen or Llech [G]wen is simply 'White Stone.'

Of great importance to our finding Gwen Ystrad is the mention at the beginning of the poem of the men of Catterick, as well as the reference to the Cumbrian River Eden. An action involving warriors from these places is interesting, as the headwaters of the Eden and the Swale are only a few miles away from each other. We now know, thanks to LIDAR, that there was a Roman road running from Bowes in Durham to Bainbridge. Thus access to Swaledale, where the Roman road passed through Feetham and Crackpot at the river, could have been gained by warriors coming up the Eden and thence east to Bowes, as well as by warriors from Catterick coming north from Bainbridge. Alternately, there may have been routes directly into upper Swaledale from the upper Eden, and the men of Catterick could, of course, have simply gone up Swaledale from their own city. It is also true that forces from Catterick and the Eden could have followed the Roman roads to another location; we need not limit our candidates for Gwen Ystrad to Swaledale.

For a nice, recent treatment of the Roman roads in Yorkshire (and adjacent Cumbria), see the following links:

http://roadsofromanbritain.org/yorkshire.html

http://roadsofromanbritain.org/gazetteer/yorkshire/rr732.html

http://roadsofromanbritain.org/gazetteer/yorkshire/rr82.html

http://www.romanroads.org/gazetteer/cumbria/cumbriapages.html

The difficulty in pinpointing Gwen Ystrad and Llech Wen is that these 'white' names almost certainly refer to geographical and geological features displaying the exposed white carboniferous limestone of the Yorkshire Dales. This is a very large area.

Does this mean that we simply can't find Gwen Ystrad and Llech Wen? Well, I think we can at least narrow down the possibilities.

First, we must glean what we can from the Urien poem to try and tease from it clues as to the location of the battle site.[1] We are told that the action occurs at a 'low rampart' (see notes from Ifor Williams). As the W. prefix go- of gofur or go-mur means the same as Latin sub, this word may indicate a wall that is under or below another wall, rather than a wall that is inferior in height. The battle also takes place at a ford, in gravel or shingles/pebbles of a river-bed, where horses tails are in the waves of the water. Obviously, then, this is a battle at/in a river. We are talking, it would appear, about a dyke, of the sort that was constructed to block passage up a river-valley. The dyke in question was, perhaps, considered to be lower down the valley than one positioned above it.

In all of Yorkshire, there is only one place that fits this description: the Grinton-Fremington Dykes. These are discussed by local expert Will Swales at the following excellent Web pages:

https://willswales1.wordpress.com/finding-hodic-and-stainburghanes/

https://willswales1.wordpress.com/grinton-fremington-dykes-defensive-system-or-land-boundaries/

Now, the dykes to the south of the river do not abutt upon any exposed limestone. It is a different story for the two northern dykes. On one end they terminate near the Arkle Burn, but on the other they are flush with Fremington Edge. That escarpment or 'high scar' is a classic carboniferous limestone scar (https://archive.yorkshiredales.org.uk/about-the-dales/landscape/landscapecharacterassessment/lca_swaledale-arkengarthdale.pdf, https://myyorkshiredales.co.uk/hills/fremington-edge/).

I would argue that Gwen Ystrad is the valley of the Arkle Beck, or Arkengarthdale, named the White Strath for the white limestone cliffs flanking the river.

And the Llech Wen/White Stone?

Probably just a name for the scar itself, as opposed to the valley name. According to the GPC, llech did have the meaning of "cliff." Cf. the Welsh place-name Harlech, 'Fine Crag'.

On the ford in question, I have the following from Will Swale (personal communication):

"Only one of the north-bank dykes reaches anywhere near the River Swale – it’s the one that passes through Fremington village and then crosses the flood plain of Arkle Beck towards its confluence with the Swale.

The other one north of the Swale is higher up Arkle Beck (both are on its east side).

There would certainly have been a ford between them, crossing Arkle Beck, where now stands Reeth Bridge, carrying the main road connecting Fremington and Reeth.

The fords over the Swale in this general area would have been at what is now Grinton Bridge, and higher upstream at Scabba Wath, which isn’t at what the OS map wrongly calls Scabba Wath Bridge (known to indigenous locals always as Whita Bridge).

The ford of Scabba Wath is still visible about 300 metres downstream of Whita Bridge, where Browna Gill Beck enters the Swale, and where the OS map marks a cattle grid in the road.

These are the only crossing points near the dykes that are obvious today."

[1]

Catraeth's men set out at daybreak

Round a battle-winning lord, cattle-raiser.

Urien he, renowned chieftain,

Constrains rulers and cuts them down,

Eager for war, true leader of Christendom.

Prydain's men, they came in war-bands:

Gwen Ystrad your base, battle-honer.

Neither field nor forest shielded,

Land's protector, your foe when he came.

Like waves roaring harsh over land

I saw savage men in war-bands.

And after morning's fray, torn flesh.

I saw hordes of invaders dead;

Joyous, wrathful, the shout one heard.

Defending Gwen Ystrad one saw

A thin rampart and lone weary men.

At the ford I saw men stained with blood

Down arms before a grey-haired lord.

They wish peace, for they found the way barred,

Hands crossed, on the strand, cheeks pallid.

Their lords marvel at Idon's lavish wine;

Waves wash the tails of their horses.

I saw pillaging men disheartened,

And blood spattered on garments,

And quick groupings, ranks closed, for battle.

Battle's cloak, he'd no mind to flee,

Rheged's lord, I marvel, when challenged.

I saw splendid men around Urien

When he fought his foes at Llech Wen.

Routing does in fury delights him.

Carry, warriors, shields at the ready;

Battle's the lot of those who serve Urien.

And until I die, old,

By death's strict demand,

I shall not be joyful

Unless I praise Urien.

A Note of Godeu of the North and Urbs Giudi

A very important region in the North of Britain was called Godeu. This place is mentioned in two of the Taliesin praise-poems of Urien. In both cases, Godeu is paired with Reget, i.e. Rheged. Yet Godeu has remained unidentified.

Locating Godeu is complicated by its use in an ancient battle poem called Kat Godeu, the ‘Battle of Godeu’. Because this battle poem tells of the god Gwydion’s magical activation of an army of trees, it has in the past been assumed that Godeu meant ‘forest’, cf. Welsh coed/goed. However, the word godeu or goddeu/goddau actually existed in early Welsh. The National Dictionary of Wales has as the meaning of this word ‘intention, design, purpose, object or aim, end in view.’

There are some clues about where we might find the Godeu of Kat Godeu. Firstly, we know Gwydion was most firmly associated with Gwynedd. One other character mentioned in the poem – a certain Peblig, can be put in Gwynedd. The only Peblig known to Welsh tradition was the saint of Llanbeblig, the parish church of Carnarvon. This Peblig is involved in the actual battle in Godeu, at a fort called Caer Nefenhir.

In the Mabinogion tale Math Son of Mathonwy, Gwydion fights Pryderi of Dyfed in Gwynedd. The battle was fought over some magical swine Gwydion had stolen from Pryderi. Pryderi had gotten these swine from Arawn, king of Annwm, the Welsh Otherworld.

A 17th century account of the Battle of Godeu tells us that Amaethon son of Don, Gwydion’s brother, had stolen a white roebuck and a whelp from Annwm. The battle was between Arawn and Amaethon. On one side was Bran, a god regularly associated with Gwynedd. In another Taliesin poem, we are told that Lleu also took part in the battle. He, too, was a figure frequently placed in northwestern Wales.

All the clues seem, therefore, to point to Gwynedd as the location of Godeu and Caer Nefenhir.

The fort in question looks to me to be Caer Nefyn Hir, the Fort of Nefyn the Tall (cf. Cai Hir, ‘Caius the Tall’). This points strongly to Nefyn on the Lleyn Peninsula, not far from Peblig’s Carnarvon. There are two forts at Nefyn.

The first is the hill-fort of Garn Boduan or Bodfuan, the ‘Cairn of the Dwelling of Buan’. Buan was a saint in the area. The second fort at Nefyn is the promontory fort of Dinllaen, the ‘Fort of the Laigin’ or Leinsterman.

The goddess Modron appears in the Kat Godeu poem and it is noteworthy that the Carn Madryn fort can be found just a little southwest of Nefyn.

But if Nefyn is the location of Caer Nefyn Hir, where is Godeu? Well, any Godeu on the Lleyn Peninsula was probably relocated there from the North. Cynfarch, father of Urien of Rheged, is given a wife Nefyn. While we could make here name out to be a nice cognate with the Irish goddess name Nemhain, there is a better possibility.

From Alan James (https://spns.org.uk/wp-content/uploads/2019/03/Alan_James_Brittonic_Language_in_the_Old_North_BLITON_Volume_II_Dictionary_2019_Edition.pdf):

"On the possibility that the eponym of Kaer Nefenhir in a poem to Llywelyn the Great by

Prydydd y Moch (c. 1215) is *Novantorix, 'ruler of the Novantae', see Haycock 2013 p.14."

If so, Godeu would be a name for the old territory of the Novantae bordering on Rheged, which started in Annandale but ran down through Cumbria.

I have discussed the most probable etymology for Godeu in an article that may be found at

https://mistshadows.blogspot.com/2021/05/godeugoddeu-identification-of-plants.html. For some sites associated with the Northern Godeu, see https://mistshadows.blogspot.com/2021/05/the-battle-of-argoet-llwyfain.html.

The old idea that Godeu is related to the Gododdin tribal name is incorrect. But this erroneous identification does bring up a related and important point: the true etymology and location of the Venerable Bede’s urbs Giudi on the Firth of Forth. Giudi is found in the epic Welsh poem The Gododdin as Iodeo. The 9th century HB spells the place-name Iudeu. Finally, the Middle Irish Mothers of the Saints mentions muir n-Giudan, where muir is ‘sea’, a reference to the Firth of Forth.

It has become customary, for no really good reason, to identify Giudi or Iodeo with Stirling Rock.

[Professor Thomas Owen Clancy is, at present, usually considered the man who has produced the best etymology for the obscure place-name Stirling. For his study on this topic, see http://www.clanntuirc.co.uk/JSNS/V11/JSNS11%20Clancy.pdf.

The idea is that Stirling or "Strivelin" is from Gaelic srib-linn, 'stream-pool.' This derivation has gained traction in recent years, despite the originator's own confession that srib "is not a common place-name element in either Ireland or Scotland". In addition, Clancy must account for why the name does not apply to Stirling Rock:

"The original srib-linn, then, would refer to this point on the river (perhaps then meaning ‘highest navigable point on an estuary’, or referring to a river-pool allowing harbourage), rather than any land feature such as the castle rock which overlooks it."

Northern British place-name expert Alan James, author of BLITON, commented thusly (via personal communication) on Clancy's proposed form:

"On the whole, I think it's quite doubtful, but in phonetic terms, it works."

Clancy's Gaelic etymology for Stirling allows him to suggest that the original British name for the place was Iudeu, a name I have recently been able to associate fairly strongly with Edinburgh (https://mistshadows.blogspot.com/2021/04/bedes-urbs-giudi-and-udd-urfai-of-eidyn.html). In Clancy's words:

"I noted at the outset that this has consequences of a sort for a separate set of debates. These relate to the question of whether Bede’s urbs Giudi (and the forms in other texts related to it, such as Iudeu, Iuddew) was Castle Rock at Stirling (see most fully Fraser 2008). Whilst the present note does not directly address that question, it does clear away one problematic sub-argument, which is the suggestion that that location could not be Giudi because it already had a different, Brittonic name, the name lying behind Stirling. This note demonstrates that, whatever the location of urbs Giudi, the name Stirling was Gaelic, referred to the river Forth, and was not originally the name of Castle Rock."

Well, I think we can rather easily account for Stirling or Strivelin being a British name - and that we can, in fact, supply a known precedent from Wales in support of such a contention.

I had independently found some interesting place-names in Merionethshire: Streflyn, Strevlin, Ystreflyn, Ystrevlyn, etc. Alan James knew of these:

"I notice in AMR there is an Ystreflin in Llanycil, and Llysystreflin in Llanuwchllyn, both in Merioneth."

I went ahead and asked my friend James at Historic Place-Names of Wales (https://historicplacenames.rcahmw.gov.uk/) what the etymology of these places might be. He responded:

"According to our Merioneth Inventory, the correct form of Streflyn is Istreflyn, 'the lower holding by the lake'. Since two of the Commissioners were John Rhys and Edward Annwyl, I think we can be reasonably confident in that analysis."

I double-checked the spelling in the Melville Richards Archives (http://www.e-gymraeg.co.uk/enwaulleoedd/amr/cronfa_en.aspx):

ISTREFLYN LLANYCIL Meironeth IAMW INVENTORIES OF THE COMMISSION OF ANCIENT MONUMENTS IN WALES AND MONMOUTHSHIRE vi.145

The sense of Is Tref Llyn as applied to Stirling would doubtless be the town below Castle Rock next to the pool in the river. Such a derivation thus allows us to identify both the original settlement atop the Rock, as well as the town that grew up below it next to the said pool.

Best of all, Istreflyn for Stirling adheres to the phonological requirements for the place-name as defined by Clancy. There is a problem with the constant [i] in Strivelin, rather than the [e] we would expect from tref. Twenty-two spellings with Striu- versus only two with Streu-. But I think it is reasonable to assume that a [I]stref would have been assimilated to srib by Gaelic speakers. Ironically, leading Clancy to deduce that srib must be the original first element of the place-name.]

Andrew Breeze likens the root of Giudi to Old Welsh iudd, Middle Welsh udd, ‘lord’, and thus interprets the name as meaning ‘lord’s place, place possessed by a lord’. As a purely formal etymology, this is quite acceptable.

Now Bede says that urbs Giudi is ‘in medio’ of the Firth of Forth. This does not mean, of course, that the city is in the middle of the Firth, but rather that is it situated in the middle of the shore of the Firth in the Gododdin region. This geographical fix immediately eliminates the traditional Stirling from consideration.

However, Din Eidyn, the Dark Age capital of the Gododdin, is itself in the middle portion of the shore of the Firth. I suspect the ‘city of the Gododdin’ is, in fact, Din Eidyn. Evidence of an udd or lord of Eidyn is found in the Gododdin poem itself:

Line 954:

Gnawd gwayw rhudd rhag udd Eidyn, Urfai

Usual was a blood-stained spear before the lord of Eidyn, Urfai

A Gorthyn Hir of the same poem is called udd of Gwynedd. Interestingly, Gorthyn is said to be the son of an Urfai. If the same Urfai is meant, then udd may be a hereditary title.

When I asked place-name expert Alan James about this possibility, he replied:

“Yes, that raises all sorts of intriguing possibilities. I'd agree that, given we really don't know what -eu signifies, Iudeu could possibly indicate the chief place of a *Iudd, and that title might have been associated with some dynasty (inheritance, as we know, was a complicated business).”

The Home of St. Patrick

A great deal of controversy still exists over the whereabouts of the home of the famous Saint Patrick. I will not here go over the various candidates, none of which have convinced the scholarly community. Instead, I will make my case for just one of these candidates, as I think new evidence can be provided in support of it.

We are told that the saint was born at ‘uico [vico – “town”] bannauem taburniae (variants taberniae, thaburinde), ‘where three roads meet’ and that this place is ‘near the western sea’. This town is otherwise known as Uentre (variants Nentriae, Nemthur).

It has long been recognized that the form ‘bannauem taberniae’, i.e. bannaven taberniae, shows an incorrect division of this place-name. Instead, it should read

Banna Venta Berniae

Venta is best defined thusly (from “Brittonic Language in the Old North”, The Scottish Place-Name Society):

“In all the cases mentioned, a sense ‘a market, a trading-place’ is quite plausible, but the apparent similarity to Latin vendere, ‘to sell’ and its Vernacular Latin and Romance derivatives is probably misleading. Both *Bannaventa and *Glannoventa, as topographical names, might incorporate the suffix seen in the river-names above, or be based on lost river-names with that suffix. Nevertheless, Sims-Williams in APN p119 includes *Bannaventa and *Glannoventa along with the Venta group, under the sense ‘market’.”

Banna’s etymology is as follows (also from “Brittonic Language in the Old North”):

“Non-IE *ban-, *ben- > Early Celtic *banno/ā- > Brittonic, Gaulish banno/ā-, also Gaulish benno- (in place-names) > Old Welsh bann- > (in place-name Banngolau AC s.s. 874) > middle - modern Welsh ban > middle Cornish ba[d]n > Cornish ban (see CPNE p. 16), Old Breton bann > modern Breton ban; Irish, Gaelic benn > (and Gaelic, manx beinn).

Primarily, 'a horn, prong, antler-time', so also 'a drinking-horn, a sounding-horn'. In Celtic place-names generally 'a point, promontary, spur', and in Brittonic and Pritenic place-names 'summit, top', a use which shaped the Gaelic and manx development of the dative (locative) singular beinn to an independent noun, especially in hill-names.: see G. Barrow in Uses, p. 56 (however, given the rarity of ban[n] in surviving hill-names, the influence of unrelated pen[n] may also have been a factor).

To me, it is fairly obvious that 'Uentre', first found in the Life by Muirchu, is merely a duplication in slightly corrupt form of Venta. Venta as 'market town' is a sort of Celtic substitution for Latin vicus, which has come down to us through Anglo-Saxon as wic or wick, 'market town'.

The Roman fort of Banna on the western end of Hadrian's Wall has often been pointed to as this particular Bannaventa (since the one in central England is not near the sea). The vicus or civilian settlement that surrounded the fort was quite large, so there is no problem with the vicus/venta portion of the name.

The problem is the 'Berniae', which no one has been able to make anything sensible out of. This is plainly a reference to the Tyne Gap, a narrow but distinct corridor running east-west through a lowlying gap between the uplands of the Pennines visible to north and south. The Gap spans the distance from the Tyne in the east to the Irthing in the west, and Banna/Birdoswald is right there at the western end of the Gap.

From the Electronic Dictionary of the Irish Language on the Irish word for 'gap':

bern

berna beirn berne bearn Bernai bernad bearnaidh bearna

Keywords: Gap; breach; pass; defile; position; defence; attack; refuge; breach; occupied; warrior; undefended; weak; position; defence; attack; position; danger; strait; fight; gap; break; flaw; drinking-horn; ungapped

Letter: B

COLUMN: 83

Line: 014

bern

ā, f. also berna iā, d and n. as. B. Chon Culaind, RC viii 54.20 ( LL 13397 ). ds. b.¤, Sitzb. , v 93 § 30 , beirn, LL 17995 , gs. berne, FM v 1636.9 . bearn f., IGT Decl. § 39. as. Bernai, Corp. Gen. 206.17

gs. cacha bernad, LL 8030. berna f., IGT Decl. § 4. as. bearnaidh, Éigse xiv 98 § 29. gs. bearna, § 28 .

gap, breach; pass, defile; hence weak position in defence or attack: dot luid i mbernai (ar berna, v.l. ) ar-mo chenn-sa thou camest into the breach against me , Sc. M² 13. is b.¤ ina coṅgaib catha a breach in their battle position , CRR 57.

do coimét na mbernd cumung robui ag techt aran slia[bh], ZCP vi 56.10 . annsa mbeirn = in the gap , Ez. xxii 30. aon- fhear faire re seasamh gach beárnan, Keat. Poems 463. nī fhūicēb-sa an b.¤ (bern-, v.l. ) sin dom c[h]onāch gan caithem, ML² 1412. bearna as mo ré `a part of my life-span ', Dán Dé xvii 8 . ar bearna an bháis, DDána 30.6 . ? berna a eric, Laws ii 98.6 .i. ar in fechiumh nos gaibh, uair do rochuir ní di, 13 Comm. b.¤ na ngrás gur daingen duid refuge, IGT Decl. ex. 411.

In phrr. b.¤, berna churad, ¤míled etc. breach made or occupied by a warrior, etc.: ruc beirnd curad . . . dar cath na nAnmarcach, Cog. 188.23 . ra briss beirn míled i cath naṅGréc, LL 32300 ( TTr. 1488 ). do bris b.¤ céit isin cath i n-urc[h]omair a aigthi `made a breach of a hundred ', Fianaig. 90.30 . berna cēt, TBC² 3672. Phr. b.¤ báegail undefended or weak position in defence or attack: aḟágbáil ar bernadaib bǽgail nó ar doirrsib aideda, Mer. Uil.² 99. ni b.¤ bægail in læch fuil and `no easy victim ', Aen. 750. See G 7.27 . Hence b.¤ position of danger, strait; fight: iarṁbrath na mb.¤, Rawl. 69 a 27 . suan ón bheirn `from fight ' (Vocab.), O'Hara 2609. re ndul san mbeirn, Dán Dé xxv 21 . gap; break, flaw in general: (expl. Bernán Brigte, name of saint's bell) foceirt forru co mmebaid ass a bernn `its gap broke out of it ' (i.e. a piece broke off), Trip. 114.14 . (of a drinking-horn) sēt blāith cen beirn `ungapped', Measgra Uí Chl. 150 § 19. an bhearn do-cháidh san chloinnse `the gap thus broken in her family ' (by a death), Aithd. D. 13.10 . dar bernadaib in inair sin, Acall. 5808 n . tar beirn na luirige, BB 435 b 46 . Compds. trias na beilgibh bernbriste dorónadh las an ordanas broken into gaps , FM vi 2300.2 . See berrbróc. beilge berncairrgidhe na banBhoirne, Hugh Roe 242.13 .

Just as importantly, the early name of Patrick in Tirechan is Magonus. The god Mogons (and variants) is found on Hadrian's Wall, especially the western half/end - where the Banna Roman fort is located. According to Celticist John Koch, the alternative name of Patrick, Magonus, might be related to this god name.

Finally, thanks to the paper by Dr. Andrew Breeze of Pamplona ("St.Patrick's Birthplace", Wlsh Journal of Religious History, 3, 2008, pp. 58-67), I have learned of the 3rd century (?) inscription, apparently from Corbridge but now at Hexham Abbey, by a Q. Calpurnius Concessinius. Martin Charlesworth of Cambridge noticed that this Roman-period name contained both the family names of St. Patrick, whose father was Calpurnius and mother Conchessa. Q. was a prefect of an unnamed cavalry unit celebrating the slaughter of a tribal group called the Corionototae. This stone thus places both of the names of Patrick's parents near the Wall, where Banna/Birdoswald is located.

Morgan Bwlch and Bernicia

According to the Historia Brittonum, Urien of Rheged was murdered at the instigation of Morgan Bwlch, the whereabouts of whose kingdom is unknown. Several unsupportable guesses have been made.

Bwlch means ‘gap, pass, breach’. It is possible, therefore, the Morgan was ‘of the breach’ in a heroic, military sense. However, I’ve demonstrated that the Tyne Gap of St. Patrick’s Banna Venta (the Birdoswald fort on Hadrian’s Wall) was called ‘Bernia’. Kenneth Jackson long ago proposed tht Bernicia is from Celtic *Bern-acci- (cf. Irish bern, berna, ‘gap, pass’). Archaeology has shown us that Bernician settlement began in the Tyne Valley. Thus in all likelihood Morgan Bwlch was a British ruler of the Tyne Gap, whose kingdom was superceded by that of the Bernician English.

Din Guayrdi/Din Guoaroy

According to the Historia Brittonum, the British name for Bamburgh was either Din (“Fort”) Guayrdi or Din Guoaroy. The name has remained a problem for philologists and no satisfactory etymology has been proposed.

I would suggest the Welsh word gwyar, ‘blood’, plus an ethnonymic suffix. In this case, Gwyar is a proper name, possibly the mother of the famous Arthurian hero Gwalchmai. Alan James has informed me that the medial syllable would have been syncopated, so we could expect a form such as *Gwyardi. This fits Guoaroy better than, say, Welsh gwaered, ‘declivity, downward slope.” In the case of Guoaroy, the 'o' could be a miscopying of 'ꝺ', 'insular d'.

Din Gwyardi, the ‘Fort of the People of Gwyar.’

William of Malmesbury said that Gwalchmai had been buried at Ros (Rhos) in Wales. This may be a relocation for Ross Low at Bamburgh.

The Welsh Triads place Gwalchmai’s grave on the Parret in Somerset, but this is doubtless because Gualganus, a form of his name, was wrongly linked to the Anglo-Saxon Chronicle’s Cenwalh, who fought the British at that river.

CHAPTER 7

THE GRAVE OF ARTHUR

It is not my purpose in this chapter to deal with what I consider to be the misidentification of

Glastonbury with Avalon. Others have presented a detailed case against the fraudulent claim of

Glastonbury as the final resting place of King Arthur, and I added some of my own arguments in my previous book, The Mysteries of Avalon.

Here I wish to restrict my attention to the only known place in Britain to actually have born the name Avalon prior to the time of Arthur as well as to this place’s proximity to both Arthur’s Camlann at Castlesteads and his possible ruling center towards the west end of Hadrian’s Wall.

Obviously, the possible location of his grave at

Avalon is of great interest to anyone seeking to demonstrate the reality of a historical Arthur. Geoffrey of Monmouth’s ‘Insula Avallonis’ or ‘Isle of Avalon’ is held by most Arthurian scholars to be a purely mythological designation - no matter where one chooses to localize it.

From a philological standpoint, the –on terminal of Avallon or Avalon demands an original terminal fronted by a broad vowel. Thus there is a problem trying to equate the word with Welsh afallen, ‘apple tree’, or Cornish avallen. This problem can be overcome in two ways: 1) by evoking an attested Continental place-name, e.g. Aballone, modern Avallon, in France or by 2) allowing for the possibility that the plural form of Welsh afal, afalau, cf. Cornish avalow and Breton avalou, at some point underwent a fairly common miscopying of u/w as n.

As it happens, the only known site in all of Roman Britain to bear an ‘Avalon’ name is the Aballava fort at Burgh-By-Sands, 5 miles west of Stanwix on Hadrian’s Wall. This fort is under

14 miles west of Castlesteads. The name Aballava is found listed in the various early sources in the following forms:

Aballava – Rudge Cup and Amiens patera

Aballavensium – RIB inscription No. 883

Avalana, Avalava – Ravenna Cosmography

Aballaba – Notitia Dignitatum

It is the one spelling in the Ravenna Cosmography that stands out here. The v of Aballava/Avalava has been rendered as an n, yielding the spelling Avalana. This is exactly the type of spelling we would need to end up with Geoffrey of Monmouth’s Latinized Avallonis.

The Celtic derivational suffix –ava of Aballava,

British *-aua, is now found in the –au of Welsh, giving as a meaning for Insula Avallonis ‘Island

of the Apple-trees’.

An Arthur who fell at Camlann/Camboglanna at

Castlesteads could easily have been carted along the Roman road or brought down the river system in this region to Burgh-By-Sands.

Camboglanna is on the Irthing, a tributary of the

Eden River. The Eden empties into the Solway

Firth very near Aballava/Avalana.

Two dedications to a goddess Latis were made at the Birdoswald Roman fort, 7 miles east of Castlesteads, and at Aballava. The first (RIB 1897) is addressed to DIA LATI and the second to DEAE

LATI. Latis comes from a British root similar to Proto-Celtic *lati-, ‘liquid, fluid’, and Proto-Indo- European *lat-, ‘wet’. Some authorities have seen in her a goddess of beer (cf. Old Irish laith, ‘ale, liquor’), but here she is manifestly a goddess of open bodies of fresh water, i.e. she is a literal ‘Lady of the Lake’. Burgh-By-Sands had vast marshland to its north. Although these lands have long since been drained, the area is still called ‘Burgh Marsh’. We can be fairly certain, then, that the Avalon fort was considered to be an island of sorts, the true ‘Insula’ of Geoffrey of Monmouth’s apple-tree Otherworld.

Topography dictated the position of the Aballava fort. There was an important crossing of the Solway at Burgh and the existence of this crossing may have influenced the siting of the Roman fort here. The fort sits atop a low hill on the highest ground at the east end of the village. The church sits within the south-east corner of the fort and is partly built of Roman stones. The modern road lies on the line of the Wall. Burgh is one of the least explored and understood of all the forts on the Wall. Although earlier visitors presumed a fort here, no remains were visible.

Excavations north of the church in 1922, when a new burial ground was formed, resulted in the location of the east wall, 6-7 ft thick, with an earth backing, and the east gate of the fort, with a road leading out. Within the fort, stone buildings running north-south were interpreted as barracks-blocks. The Roman levels and buildings were all badly preserved.

The sketch plan of the site prepared on the basis of these discoveries suggests a fort measuring 520 ft north-south by 410 ft east-west, giving an area of nearly 5 acres. Excavations on several occasions between 1978 and 2002 south and east of the fort has led to the discovery of buildings, presumably of the civil settlement. The bath-house, south of the fort, was destroyed in making the canal, itself replaced by the railway line, now also abandoned. Further south, the tombstone of a Dacian tribesman may indicate the location of the cemetery. Recent excavations have failed to clarify the location, size and date of the Wall fort at Burgh. We do know the stone fort lay astride the Wall, but the Wall ditch was infilled and re-cut before it was constructed. It is possible that the fort to the south of the Wall at

Moorhouse was retained for some time before being succeeded by a replacement astride the Wall.

As stated above, the actual Roman period cemetary at Burgh-By-Sands/Aballava is said have been to the south of the fort. When I enquired about the tombstone of the Dacian tribesman found in this cemetery, Tim Padley at the Tullie

House Museum in Carlisle informed me of the discovery of two other fragments. All three are listed in the Roman Inscriptions of Britain as follows:

2046 (tombstone)

...

IVL PII... TINVS CIVES DACVS

2047 (tombstone) D M S

...

2048 (tombstone) VII

Alas, according to Mr. Padley, the placement of the cemetary to the ‘south of the fort’ puts it, in his words, ‘near the vallum, possibly destroyed by the canal and railway.’

The tombstone fragments were in the care of Tullie House when they disappeared.

While it is impossible to know whether Arthur was buried in the Roman period cemetary of the

Aballava fort, this cemetary must remain a primary candidate for the location of his grave.

AFTERWORD

THE KING WHO WILL BE AGAIN

In this book, I have revealed a chieftain named

Arthur, son of a chieftain of Ribchester, who fought a dozen or so Dark Age battles against the Saxons before being taken to Avalon/Burgh-By-Sands to be with the Goddess of the Lake in death.

But for many – and this is as true for those who lived shortly after Arthur’s time as it is for us of the modern era – Arthur does not lie in a grave at Avalon. Instead, he continues to live in a curious limbo, a place composed of equal parts continued literary and artistic invention, entertainment and gaming industry exploitation, academic specialization and Celtic/Neo-Pagan

Reconstructionism.

Under the latter guise, the shift away from an Arthur who is an attestable personage has, ironically, paralleled the renewed academic insistence on the non-historicity of this northern chieftain. For while academics now, almost without exception, view Arthur as a purely legendary figure derived from folklore and developed through the medium of medieval romance, the Celtic Reconstructionists reinterpret this greatest of British heroes in a multitude of ways.

Some still hold to the age-old Messianic view that Arthur is merely being healed of his wounds by Morgan le Fay in a spiritual versus a physical

Avalon. They believe that he will, in the time of

Britain’s most dire need (or, indeed, in the time of Mankind’s most dire need), come forth to defeat some monstrous evil. Others seek to trace their bloodlines to Arthur, to his knights, to Avalon priestesses or to Grail kings in order to inherit the immense spiritual heritage that resides in the Arthurian story.

There are even individuals who claim to be Arthur or, perhaps, a reincarnation of him. I have personally met a man who makes a very decent living ‘channeling’ the spirit of Merlin, a spirit who with profound and pithy pronouncements advises clients on how to go about conducting their daily lives and business affairs.

This ‘New Aging’ of Arthur would seem to be a harmless phenomenon, serving the positive function of bringing many new members into the

Arthurian fold and contributing to a heightened level of spiritual awareness, as well as fostering a sense of ‘connectedness’ with ancestors and nature in an uniquely Celtic fashion. But today’s pagans need to be careful not to create alarming amounts of contrived information masquerading as inspired truth or subjective revelation that might feed naive, inwardly-focused belief systems. These last distract us from objectively obtained realities.

As a person who himself is not immune to mystical experience, let me hasten to add that I am not advocating spiritual matters be excluded from the Arthurian orbit. Cutting off this aspect of our humanity is not only undesirable, but thoroughly impractical. The human psyche simply does not work this way. What I am pleading for is a separation of what is acknowledged as fact or reasonable conjecture from what is intuited as having religious significance. Or, to be more precise, what we choose to adhere to as tenets of belief should be based upon or extrapolated from what we know on a rational level, rather than the reverse. Be spiritual about things/concepts that actually exist or once existed. Do not give in to the temptation to readily accept as the basis for belief anything that contradicts or ignores a body of evidence assembled by decades or centuries of intense scientific effort.

St. Augustine said “I believe so that I may know”. This is a dangerous credo. Instead, “We should know so that we can believe.”

Furthemore, any belief system, no matter how self-satisfying, should be eschewed by anyone who truly cares about Arthur and things Arthurian if it intentionally seeks to hide potential truths from us or block us from paths that may lead to genuine understanding of deeper matters.

Belief systems of this sort are usually promulgated by cult leaders who created them, i.e. those with their own often sinister agendas and need for control and profit. Any cultic use of the Arthurian tradition would be, in essence, antithetical to that tradition. Still, there always remains the danger that unusually susceptible minds could be programmed to make use of the Arthurian tradition in an unacceptable fashion.

Those who wish to be latter-day Knights of the Round Table must guard against such a misuse of a code of conduct that, as it has been refined since the Middle Ages, promises compassionate treatment of all persons, places and things.

Most marvelous of all would be the development of a joint spiritual, aesthetic and scientific mindset whose sole unifying purpose was to increase and enhance opportunities available for those daring individuals questing after a real Arthur. If there were a core group of people of this ilk or predisposition, the whole thrust of exploration into the possibility of an historical Arthur would not only be forever altered, but in my opinion strengthened a hundred-fold. A solitary vision forged from divergent approaches and applied with discipline and conviction to the problem of

Arthur is what is required for us to be able to bring Arthur back from the Otherworld of disbelief in which current scholarly opinion has consigned him. Like the sword Excalibur, made in Avalon and rising up through the water and mist of the lake, a new paradigm in the field of Arthurian Studies must be inaugurated, implemented and sustained.

Of course, our chances of finding Arthur – of ultimately proving his historicity – depend almost entirely on finding an intact, inscribed tombstone somewhere in the vicinity of the Burgh-By-

Sands fort. Natural processes such as erosion, combined with man’s radical alteration of the environment and his accompanying willful destruction and occasional subsequent reuse of ancient monuments, makes the likelihood of our finding a memorial stone set up for Arthur of the

6th century an extremely doubtful proposition. Still, foolish as it may sound, it is just such an artifact that we need to be searching for. The discovery of the Cynric/Cunorix Stone at Wroxeter gives us hope that a similar stone for Arthur may someday be found at Burgh-By-Sands.

New labour saving methods of investigation now at our disposal, e.g. ground penetrating radar, allow for less costly, less invasive, less timeconsuming, less bureaucratic means of finally locating and determining the nature and extent of the Burgh-By-Sand’s Roman cemetery. Whole memorial stones or fragments of such stones may have been incorporated into walls or buildings.

A careful visual inspection of these kinds of structures may yield significant findings. Members of the community of Burgh-By-Sands could be canvassed regarding any stones in their possession that bear what appears to be ancient writing. A correspondent who lives in the town quipped that someone could be using an inscribed stone for a door-stop! Then again, Arthur’s Stone may be awaiting discovery in the foundation of the church or in the stall of a neighboring farmer’s barn.

It is also vital to attempt to ascertain exactly where the missing Burgh-By-Sand tombstone fragments were found. This would entail determining who now possesses the fragments – if the parties concerned or their immediate descendents are still alive. Interviews with such people might shed important light on the cemetary’s location.

If private collectors of Roman artifacts can be made aware of the importance of these stones, perhaps they would be forthcoming with valuable information, especially if they were allowed to do so anonymously or were granted immunity from prosecution in the event the artifacts in question had been obtained illegally.

Archaeological excavation work is always the final resort. All too often it is undertaken in advance of a major building project as an imposed afterthought. Time, money and personnel constraints add up to produce what is all too often a hurried, haphazard and incomplete dig.

Only rarely is a full-scale archaeological project proposed and executed because a site is chosen in advance for its own intrinsic merits.

I would put forward the Burgh-By-Sands Roman period cemetery as an archaeological site of astounding potential. Just imagine what it would be like to discover Arthur’s memorial stone! Granted, we cannot know if a tombstone was ever made for Arthur. Nor can we know whether such a stone, if made, has survived the intervening centuries. And even if such a stone were found, nay-sayers would insist that just because a Arthur of the right period could now be placed at Burgh-By-Sands, it does not follow that such an individual was THE Arthur.

However, if there is any truth to the Avalon story, and Arthur was brought to Burgh-By-Sands to be buried in sacred ground, then there is a remote chance that some trace of his presence at this fort has been preserved. And it seems to me that there must be someone out there whose destiny it is to find that trace.

Someone who wishes to prove, once and for all, that Arthur did exist.

APPENDIX I

A Northern Prototype for the Arthurian Grail Castle*

A half dozen kilometers west of the Aballava/Avalana/"Avalon" fort on the west end of Hadrian's Wall is Congabata at Drumburgh. Congabata might mean ‘dish-like’, perhaps a reference to the bold knoll on which the fort sits, which might have been seen as an upturned dish (see http://discovery.ucl.ac.uk/35459/1/4128635.pdf; Breeze, David J., J. Collingwood Bruce’s Handbook to the Roman Wall, 14th Edition, Society of Antiquaries of Newcastle Upon Tyne, 2006). A gabata was a kind of dish or platter (http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0059%3Aentry%3Dgabata).

However, con- as a Latin prefix (com- becomes con- before /c/) means 'with, together.' For example, concavus means, literally, 'with a hollow.' Thus we could interpret Congabata as 'with a hollow platter.'

According to Du Cange’s medieval dictionary , the kind of plate called a grasal or greil – the word preserved in the Holy Grail of Arthurian tradition – was the same as the gabata (http://ducange.enc.sorbonne.fr/GRASALA; https://arbredor.com/ebooks/Arthur2.pdf).

There thus appears to be sufficient reason to suggest that Drumburgh or Congabata may be the prototype for the later Grail Castles - no matter where these happened to be geographically situated. A 'Fort with a Dish' would have immediately drawn to itself mythological motifs concerning ancient Celtic sacred vessels, themselves to be eventually supplanted by similar cultic items in the Christian religion.

NOTE: Congabata was garrisoned in the late Roman period by the Second Cohort of the Lingones, a Celtic people from Gaul. Their capital city was called Andematunnum, a place-name meaning 'Great Divine Bear' [see Melrose, R., The Druids and King Arthur: A New View of Early Britain, McFarland, 2014, p. 81; Graham Isaac's Place-Names in Ptolemy's Geography, 2004; Ross, A., Pagan celtic britain, London, 1967, p. 375, for Matunus in Britain as 'Divine Bear', and the tombstone for Matuna at Caerleon (HTTPS.ROMAININSCRIPTIONSOFBRITAIN.ORG/INSCRIPTIONS/SEARCH?QV=MATUNA&SUBMIT=). The Math and Mathonwy of the MABINOGION belong to the same Celtic root, and Arthur is associated with the fort of Math.

While it is true that Alexander Falileyev would derive Andematunnum differently (Dictionary of Continental Celtic Place-Names, 2010), neither he nor Isaac were aware of the fact that the First Cohort of the Lingones had garrisoned Bremenium/High Rochester, where the bear-god Matunus was worshipped. This is ample evidence in support of an argument for interpreting Andematunnum as either the 'great Bear Place' or, literally, 'Great Divine Bear.'

* For my treatment of the earliest Arthurian Grail story, see https://mistshadows.blogspot.com/2021/04/diwrnach-irishman-new-interpretation-of.html. And for a general overview of the Grail legend, see https://mistshadows.blogspot.com/2018/03/the-holy-grail-of-arthurian-tradition.html. Lastly, for a new identification of the Grail king Alain de Gros, see https://mistshadows.blogspot.com/2020/05/very-brief-post-on-alain-de-gros-grail.html.

APPENDIX II:

SEGANTII, NOT SETANTII?: DECIDING ON A TRIBAL NAME IN NORTHWEST ENGLAND

The lot of difficulty has attended scholarly attempts to etymologize Seteia, the Roman period name for the River Mersey. To date, we do not possess a satisfactory solution to the riddle of this British hydronym.

What follows is a brief discussion of the current literature on the subject.

"The suggestion that a deity-name *Sentanā- ‘traveller, wanderer’ might underlie the ethnic name Setantii and the river-name Seteia (PNRB pp. 456-7) requires an improbable grafting of an early Goidelic form *Sēt- onto Brittonic suffixes (Cúchulainn’s given name Sétanta raises similarproblems, see CPNS p. 25, DCM p. 102). An ancient river-name unconnected with the root*sent- seems more likely to underlie these (and any connection between the Setantii and Cúchulainn remains doubtful); but see Breeze (2006b). Seteia was probably the River Mersey,and Portus Setantiorum a site (Meols?) on the Mersey estuary (D. J. Breeze 2017, 5)."

https://spns.org.uk/wp-content/uploads/2019/03/Alan_James_Brittonic_Language_in_the_Old_North_BLITON_Volume_II_Dictionary_2019_Edition.pdf

* A.L.F Rivet & Colin Smith : The Place-names of Roman Britain, p 456-457 :

“Setantorum Portus / Fleetwood, Lancashire)

DERIVATION. This ethnic name is mysterious; there seem to be no British roots visible, and very few analogues anywhere of names in Set-. It is tempting, in view of Ptolemy's variants which show Seg- (Seg-) both for the port-name and the river-name, to suspect some confusion with the Seg- of Segontium, a possibility that occurred to Rhys (1904) 315 with regard to the river, though eventually he seems.to wish to main tain Setantii as a proper form. The strongest argument for so doing is provided by Watson CPNS 25, who points out that the first name of the Irish hero Cuchulainn was Setanta (from an earlier *Setant(os) : 'the Setantii were an ancient British tribe near Liverpool. . . the inference is that Setanta means "a Setantian" and that Cuchulainn was of British origin'. But the relation between these two names has been questioned. There is a full exposition of the problem by Guyonvarc'h in Ogam, XIII, (1961), 587-98, with discussion of views of Mac Neill, Osborne, and others, including Brittonic-Goidelic transferences in both historical and phonetic aspects. The essence of the matter is that it is tempting to see in this name Irish sét ('path'; = British *sento-, for which see CLAUSENTUM), but *-ant- suffix (as in DECANTAE) is Brittonic only, for -nt- does not exist in Goidelic. The name might be based on a divine name *Setantios, not otherwise known, and he in turn might be related etymologically and by sense to the goddess Sentona, perhaps 'wayfarer' (see further TRISANTONA1). Clearly there is an additional problem in reconciling the a/e vowels in these forms (Trisantona, Gaulish Santones) if they are indeed connected. There, for the présent, the matter rests; but it is as well to reiterate that one cannot base too much speculation on forms recorded by Ptolemy alone, particularly when, in numbers, the MSS of his work record attractive variants.

IDENTIFICATION. Presumably a minor tribe, but since they appear only as part of a 'descriptive' name in the coastal list (next entry) and not in their own right in the full list of tribes, they probably formed part of the Brigantian confederacy. If the river name seteia is directly connected with them, they should have stretched along the Lancashire coast from the Mersey to Fleetwood.

Setantiorum Portus

From its position on Ptolemey’s map probably a harbour at or near Fleetwood, Lancashire, at themouth of the river Wyre.

Seteia

Evidently the river-name is the basic one of the trio; it might, as often, enshrine a divine name, but this possibility does not help towards an etymology or a meaning.”

From NORTHERN HISTORY, XLIII: 1, March 2006, by Andrew Breeze:

“If we seek a Celtic root that figures in toponyms and might be corrupted to Set-, there seems a better possibility. This is Celtic *met- ‘cut, harvest’, as in Welsh medaf ‘I reap’, Medi ‘September’ (when corn is cut), and MiddleIrish methel ‘reaping party, band of people’.”

Graham Isaac's *sego- [1] was not hard to come by, as we have the following variant spellings for Setantiorum Portus and Seteia:

Segantiorum

Gesantiorum

Segeia

Segeiais

Segeiatis

According to THE DICTIONARY OF CONTINENTAL CELTIC PLACE-NAMES (ed. by Alexander Falileyev), which quotes from G. R. Isaac, Place-Names in Ptolemy’s Geography. CD-ROM. Aberystwyth 2004.:

"Due to the very easy and frequent confusion of tau and gamma in Greek transmission, it is possible that some, or even all, of the apparent instances of set(i)o- are for sego- (as very frequently in names in ms. {Set-} in Hispania, where there are ample other sources to confirm the correctness of the reading {Seg-})."

The same source allows for a possible derivation from seit-:

"But there may also be a genuine base set-, or {se:t-} <*seit-, involved. But if so, I cannot suggest an analysis of it at this time.’"

The only thing I could find on seit- in relation to these Set- place-names is a brief discussion in http://www.asciatopo.altervista.org/narbonensis.html:

Extension: *seit-

Reconstructed from Lithuanian sietuva 'a deep place in the river, pool'

Suffixed full-grade form *seit-i-a in Setia (Latium)

Suffixed full-grade form *seit-i-o in Setius m. (Narbonensis)

Setia

Place: Sezze, province Latina, region Lazio, Italy

Name: Setia (Ptol., Plin., Liv., Dion.)

Etymology: A stem *set- is widely diffused in toponymy. The name has exact counterparts in Setia (Baetica), Setia (Tarraconensis). With different suffixes we have Setovia (Germania), Seterrae (Tarraconensis), etc. An Illyrian Setovia has been explained by [Duridanov] from a *seit-oua, thus from an IE root *seit-. This is not included in Pokorny's dictionary (some Baltic cognate appellatives meaning 'a deep place in the river, pool' are under the root *sei-t- 'to let fall'), but probably is an extension of the huge family *sei-/si- somehow related to waters. This probably is the Pokorny's *sei- 'to be damp, to drip'. The feature *ei>e found in Setia is typically Eastern Italic (Volscan).

Setius m.

Place: Mont Saint Clair (Sète), department Hérault, region Languedoc-Roussillon, France

Name: Setius m. (Ptol., Avien.) Sigius m. (Strab.)

Etymology: A cognate of Setia (Latium), the name derives from the extension *seit- of the IE root *sei- 'to be damp, to drip'. The reason of such a name is that the hill dominates a marshy area (étang de Thau).

Alas, we cannot prove the existence of seit- in Celtic. We do, however, have plenty of examples of sego- place and personal names.

But if Segeia is the right form of the river-name, in what sense was it applied to the Mersey? Breeze, whose met- idea for set- (see above) is unacceptable (there is simply no justification for assuming that a word spelled with either a Greek tau or a Greek gamma would have as its original form one starting with mu), complains that the Mersey was a "sluggish river".

Breeze is not entirely correct in his assertion. The Mersey has the second largest tidal bore in all of Britain (see http://www.merseyestuary.org/the-tidal-bore.html). Furthermore, "The Narrows further downstream of the Inner Estuary are characterised by changes in geology ond the Estuary becomes a straight narrow channel with depths of up to 30m even at low water, and fierce tides of up to six knots (http://www.environmentdata.org/archive/ealit:4093/OBJ/20002960.pdf)." "The River Mersey is an extremely dangerous river. The Mersey has the third fastest tidal run in Europe, with the speed of the water reaching 10 knots in places (https://www.bbc.co.uk/liverpool/localhistory/mersey_times/issue_07/about.shtml#:~:text=The%20River%20Mersey%20is%20an,rapid%20death%2C%20often%20within%20minutes)."

Breeze also contradicts himself, saying right after mentioning the River Seint that sego- is unknown in hydronyms! Seint itself is from British *Segonti, and it is believed the fort of Segontium was named for the river. The root of *Segonti is sego-. He fails to cite a single instance of a river-name derived from the root *met-.

Xavier Delamarre (in Dictionnaire de la langue gauloise: Une approche linguistique du vieux-celtique continental, 2003) proposes to derive the Belisama (= the River Ribble) name from the Gaulish root belo- ('strong, powerful'), rendering Belesama as 'the Very Strong' (cf. Sanskrit baliṣṭhaḥ 'the strongest'). According to him, cognates stemming from *bʰelH- do not seem to connote 'shining', but rather 'white, grey, pale'. If he is right, we could have two major rivers in the same tribal region that meant 'strong'.

In response to my discussion of seg- rather than set- for the river-name, noted Brittonic place-name expert Alan James remarked only that "I think both those ideas [Segeia for the Strong One and Belisama for the Very Strong] are quite, well, 'strong' possibilities."

In brief, we might easily refer to the Mersey as 'the Strong or Forceful One', a goddess to pair with Belisama of the Ribble. She would have given her name to the tribe, who became the people of the Strong One.

[1]

victory *sego(s)- (?), SEMANTIC CLASS: action, Celtiberian Sego-bris (?); seko ‘victory-fort (?)’, Gaulish Sego- (-maros, dūnon, -briga, etc.) ‘victory’, Early Irish seg (MIr.) ‘strength’, Scottish Gaelic seagh ‘sense, esteem’, Welsh hy ‘bold, brave, undaunted, intrepid, valiant, steadfast, confident, daring; audacious, presumptuous, impudent’,

[English–Proto-Celtic Word-list with attested comparanda, University of Wales]

*sego- 'force' [Noun]

GOlD: Mlr. seg [0 m] (DIL sed, seg) 'strength, heed, interest, an equal'

W: MW hy 'bold, brave' (GPC hy, hyf)

GAUL: Sego-maros [PN], perhaps Segestica [Toponym]

LEP: sexe()u (?) 'Lepontic coin'

CELTIB: Segouia (?) [Toponym], Sekobirikez [Abl. s, Toponym] (A8)

PIE: *segh_ 'hold (by force)' (IEW: 888f.)

COGN: Skt. sahate 'be able, support', Gr. ekhi5 'hold, have', Go. sigis

'victory'

ETYM: Mlr. and Early Molr. seg is sometimes spelled sed (Gen. sg. seda,

seadha). Celtib. asekati (Botorrita I) might reflect *ad-seg- (Eska 1989). W

hoel [m] 'nail, peg, stake' has been derived from the o-grade of the root (?

PCelt. *sogHi), but this is not wholly convincing for semantic reasons. MW

hoen [m and f] 'joy, 'gladness, vigour' could be from *sogno-, but again the

difference in meaning is conspicuous. Finally, Olr. sar [0 m] 'outrage', sar-

'exceeding, excellent' may be related, if we start from PIE *saxsro- < PIE

*sogh-sro- and accept the lengthening of vowels before *xsL (cf. *taxslo-

'axe' < *tok-slo-).

REF: LEIA S-68, GPC II: 1945, 1884f., LIV 467, EIEC 123, Delamarre

269f., Jordan Colera 1998: 31, Sims-Williams 2006: 107f., MLH V.l: 329f.

[Etymological Dictionary of Proto-Celtic By Ranko Matasovic]

APPENDIX III:

LUCIUS ARTORIUS CASTUS AND THE WAR AGAINST ARMORICA

It was once thought that ARMORICANOS should fit on the L. Artorius Castus memorial stone. When this proved to be way too long to accommodate the fragmentary ARM[...]S, ARMORICOS was proposed instead. But there was even loud oppposition about this word, which itself - it was claimed - simply wouldn't fit on the stone. Until, of course, I showed that, in fact, it could:

https://mistshadows.blogspot.com/2021/06/the-lucius-artorius-castus-stone-with_14.html

Over the past few months, I've been going back and forth with scholars on the possibility that ARMORICOS was the location where Castus took his legionary vexillations. My main conversation has been with Professor Roger Tomlin, who still thinks ARMENIOS is a better candidate. However, he is the first to admit that ARM[...]S may be something else entirely. As an example of how wrong we can be with this kind of reconstructive guesswork, he presented an example from his own researches. I supply that here in full:

"We can only play with probabilities! It's good to suggest many possibilities, but ...

One of the Bath tablets was missing its top right-hand corner, but I could see its heading was centred:

DOCI[

BRV[

DEAE SAN[

So I read:

DOCI[LIS

BRV[CETI

DEAE SAN[CTAE

since Brucetus was a locally attested name, and deities are often addressed as sanctus.

Then I found the missing piece in the box of another tablet, and it now read:

DOCI[LIANVS

BRV[CERI

DEAE SAN[CTISSIME

The engraver had not centred Docilianus, he had mis-read the t of Brucetus in his draft, and mis-spelt the ending of sanctissimae.

You can't win!"

Tomlin only really has one objection to a reading that favors ARMORICOS as a reference to a purely regional outgrowth of the Deserters' War:

"This is a possible scenario, of course, but it involves assumptions that are not backed by the text – that Castus' opponents were nationalists, not 'deserters', and that they did not ravage (the whole of) Gaul.

If Castus had campaigned only in Armorica against a much wider-ranging opponent, than he might have said 'in Armorica', but he would have been perverse to call his opponent 'the Armoricans'."

Now, while this is logically sound, we might suppose that the deserters in Gaul (specifically Gallia Lugdunensis, which included Armorica) were the match that lit the fire, so to speak. In other words, it could be that the deserter uprising brought about a full rebellion in Armorica - a true nationalist rebellion that was restricted to that sector of Gaul. Certainly, things were bad enough in Armorica to require a new governor, Pescennius Niger, to be appointed over Gallia Lugdunensis to deal with the problem:

“Now Pescennius was on very friendly terms with Severus at the time that the latter was governor of the province of Lugdunensis. For he was sent to apprehend a body of deserters who were then ravaging Gaul in great numbers, and because he conducted himself in this task with credit, he gained the esteem of Severus, so much so, in fact, that the latter wrote to Commodus about him, and averred that he was a man indispensable to the state.”

https://penelope.uchicago.edu/Thayer/E/Roman/Texts/Historia_Augusta/Pescennius_Niger*.html

We know that Armorica gave Rome fits in the later period. The following is from John Koch's CELTIC CULTURE: A HISTORICAL ENCYCLOPEDIA:

"In the later Roman period, Armorica was more than once controlled by regional emperors, backed by the Romano-British garrison, whose authority was not recognized in Italy or the East—for example, Carausius and Allectus (287–96), Maximus (Macsen Wledig; 383–8), and Constantine III (407–9). Intermittently from the late 3rd century, Armorica slipped out of Roman control altogether, as a result of a series of

uprisings by bacaudae (rebel bands made up of peasants and disaffected soldiers). The Byzantine historian Zosimus (6.5.2) relates concerning the events of ad 409: . . . the barbarians from beyond the Rhine overran everything at will and reduced the inhabitants of the British Island and some of the peoples in Gaul to the necessity of rebelling from the Roman Empire and of living by themselves, no longer obeying the Romans’ laws. The Britons, therefore, taking up arms and fighting on their own behalf, freed the cities from the barbarians who were pressing upon them; and the whole of Armorica and other provinces

of Gaul, imitating the Britons, freed themselves

in the same way, expelling Roman officials and establishing a sovereign constitution on their own authority . . . (Trans. Thompson, Britannia 8.306). A shaky Roman rule was re-established in 417 and lapsed more than once before the mid-5th century. By the 460s, we find a ‘king of the Britons’ with the Brythonic name or title Rigotamus ‘supreme king’ and 12,000 men on the Loire, and Gallo-Roman Armorica belongs to the past."

Might we not accept the possibility that under Commodus, when everything seemed to be going wrong, the Armoricans were encouraged by the deserter uprising to attempt to throw off the Roman yoke entirely? And that help was sent for from Britain, with Castus answering the call?

This does not seem unreasonable to me. In fact, as our only other possible candidate (according to a broad consensus in the academic world) is Armenia (with Castus having gone with the British governor Statius Priscus to that country in the early 160s), Armorica right across the Channel from Britain still looks quite attractive. Even Tomlin agrees that Armenia was very far away.

I had researched British vexillations on the Continent in Saxer's work and found that the two instances we know about only went about halfway between Britain and Armenia. Yes, we did have an entire legion taken from Bonn on the Rhine to Armenia, but all in all, speaking purely from a geographical perspective, Armorica is preferable to Armenia.

Of course, if we allow for ARMORICOS being the right rendering of the fragmentary ARM[...]S, that opens up another can of worms: the mission of the 1500 British spearmen to Rome to rid the Empire of the Pratorian Prefect Perennis. There is continuing debate as to whether this mission was headed up by Castus or by the legate Priscus whose British troops tried to raise to the purple.

The speculative reconstruction of Priscus' career, pieced together from badly damaged inscriptions, was undertaken by Tomlin. His conclusion was that Priscus would have been immediately removed (perhaps with the other legates under Perennis' directive) and shipped to the Continent to head up an eastern legion. He then took on a special mission (as praepositus) with either British detachments or German detachments (we can't be sure of which). But in my judgment - and Tomlin agrees - it seems highly unlikely that a man just removed from Britain would, within a short period of time, be given command of British troops on the Continent. Plus, we have archaeological evidence that strongly suggests the Deserters' War involved Germany as well, so it makes more sense logistically for Priscus to have taken German troops against the enemy rather than British ones. This is especially true as Priscus prior to leading these detachments was legate of V Macedonica, which was stationed at Potaissa in the province of Dacia Porolissensis (modern Turda).

If the 1500 spearmen who went to Rome were, originally, fighting a rebellion in Armorica, then there are three ways to explain their subsequent conduct. Firstly, as I've previously proposed, Dio might have confused an escort or honor guard sent from the main force to Rome to petition the Emperor for Perennis' removal with the entire force. That is quite plausible. We might also, should we choose to incorporate Herodian's account, have the British troops pursue Maternus to Rome. Finally, we may accept the account at face value and allow for the victorious British legionaries coming to Rome to both recieve honors and to redress grievances. As long as the force was not seen in any way as mutinous, it could have been granted free passage to the capital. It is possible, I suppose, that while the rebellion had been quelled in Armorica, there were still enough of the deserters in other districts to make traveling with a larger force necessary for the protection of the removed British legates.

Do we find ourselves, then, once again with ARMORICOS as a feasible reading for the ARM[...]S of the L. Artorius Castus memorial stone?

BIBLIOGRAPHY

Alcock, Leslie. Arthur’s Britain: History and Archaeology A.D. 367-634. 1990, Penguin, Hammondsworth.

Austen, Paul S. Recent Excavations on Hadrian’s Wall at Burgh-By-Sands in Transactions of the Cumberland & Westmorland Antiquarian & Archaeological Society Vol. XCIV. 1994, Alan Sutton Publishing Ltd, Stroud.

Bannerman, John. Studies in the History of Dalriada. 1974, Scottish Academic Press, Edinburgh and London.

Barber, John, & Elaine Lawes-Martay and Jeremy Milln. The Linear Earthworks of Southern Scotland: Survey and Classification in Transactions of the Dumfriesshire and Galloway Natural History and Antiquarian Society, Series III, LXXIII, 1999.

Bartrum, Peter C. A Welsh Classical Dictionary: People in History and Legend up to about A.D. 1000. 1993, The National Library of Wales.

___________. Welsh Genealogies A.D. 300-1400 (8 vols). 1974, 1980, Cardiff.

Bidwell, Paul (ed). Hadrian’s Wall 1989-1999: A Summary of Recent Excavations and Research prepared for The Twelfth Pilgrimage of Hadrian’s Wall, 14-21 August 1999. 1999, Carlisle: Cumberland and Westmorland Antiquarian and Archaeological Society and the Society of Antiquaries of Newcastle upon Tyne.

Bieler, Ludwig (ed. & tr.). The Patrician Texts in the Book of Armagh. Scriptores Latini Hiberniae 10. 1979, Dublin Institute for Advanced Studies, Dublin.

Breeze, Andrew C. Pennango Near Hawick and Welsh Angau ‘Death’. 2002, Northern History 39.

___________. Some Celtic Place-Names of Scotland: Ptolemy’s Verubium Promontorium, Bede’s Urbs Giudi, Mendick, Minto, and Panlathy. 2004, Scottish Language 23.

Breeze, David J. J. Collingwood Bruce’s Handbook to the Roman Wall. 2006, 14th Edn. Society of Antiquries of Newcastle upon Tyne.

Bromwich, Rachel. Trioedd Ynys Prydein: The Welsh Triads. 1978, 2nd Edn, The University of Wales Press, Cardiff.

________________, & A.O.H. Jarman, & Brynley F. Roberts (eds.). The Arthur of the Welsh. 1999, The University of Wales Press, Cardiff.

________________, & R. Brinley Jones (eds.). Asudiaethau ar yr Hengerdd. 1978, University of Wales Press, Cardiff.

Brown, T. Craig. The History of Selkirkshire or Chronicles of Ettrick Forest, Vol. 1. 1886, David Douglas, Edinburgh.

Cable, James (tr.). The Death of King Arthur. 1975, Penguin Books.

Calise, J.M.P., Pictish Sourcebook: Documents of Medieval Legend and Dark Age History. 2020, Greenwood Publishing Group.

Cameron, Kenneth. The Place-Names of Derbyshire, Vol. 1. 1959, Cambridge University Press, Cambirdge.

Cessford, Craig. Post-Severan Cramond, The Heroic Age, Issue 4 (Winter) 2001. [See Supplementary Online Bibliography]

Clarke, Basil. Calidon and the Caledonian Forest, in The Bulletin of the Board of Celtic Studies, Vol. XXIII, Part III, November 1969.

Coates, Richard. Middle English Badde and Related Puzzles in North-Western European Language Evolution, Vol. 11, February 1988.

___________, On some Controversy surrounding Gewissae/Gewissei, Cerdic and Ceawlin in Nomina 13, 1989-90.

Collingwood, R.G. Explorations at the Roman fort of Burgh-By-Sands in Transactions of the Cumberland & Westmorland Antiquarian & Archaeological Society, Vol. XXII, 1923.

Collingwood, W. G. Arthur's Battles in Antiquity 3, 1929, 292-298.

Cooper, Nicholas J. The Archaeology of the East Midlands: An Archaeological Resource Assessment and Research Agenda, in Leicester Archaeology Monograph No. 13. 2006, University of Leicester, Leicester.

Crawford, O. G. S. Arthur and his Battles in Antiquity 9, 1935, 277.

Curtis, Renee L. The Romance of Tristan. 1994, Oxford University Press, Oxford.

Dark, K.R. A Sub-Roman Re-Defense of Hadrian’s Wall? In Britannia Vol. XXIII, 1993, 111-120.

________, & S.P. New Archaeological and Palynological Evidence for a Sub-Roman Reoccupation of Hadrian’s Wall, in Archaeologia Aeliana, 5th Series, XXIV, 1196, 57-72.

Diehl, Ernst. Inscriptiones Latinae Christianae Veteres Vol. I-III. 1926-1931, Berlin.

Dyer, James. The Penguin Guide to Prehistoric England and Wales. 1982, Penguin Books, New York and London.

Ekwall, Eilert. English River-Names. 1928, Clarendon Press, Oxford.

___________. The Concise Oxford Dictionary of English Place-Names, 4th Edn. 1977, Clarendon Press, Oxford.

Field, P.J.C. King Arthur’s Battles, An Inaugural Lecture, School of English & Linguistics, University of Wales, Bangor, 1995.

Foster, Idris, and Glyn Daniel. Prehistoric and Early Wales. 1965, Routledge and Kegan Paul, London.

Fraser, James E., From Caledonia to Pictland: Scotland to 795. 2009, Oxford University Press.

Frere, Sheppard. Britannia: A History of Roman Britain. 1987, Routledge and Kegan Paul, London and New York.

Gover, J.E.B., A. Mawer & F.M. Stenton (eds). The Place-Names of Wiltshire in Journal of the English Place-Name Society 16. Cambridge: 1939.

Hall, J. R. Clark, & H.D. Merritt. A Concise Anglo-Saxon Dictionary (4th edn). 1969, Cambridge University Press, Cambridge.

Handford, S.A. (ed), & Mattingly, H.(tr). The Agricola and the Germania. 1971, Penguin Classics.

Haycock, Marged (ed). Blodeugerdd Barddas o Ganu Crefyddol Cynnar. 1994, Barddas, Swansea.

Higham, N.J. King Arthur: Myth-Making and History. 2002, London and New York.

__________, & Barri Jones. The Carvetii. 1991, Alan Sutton Publishing Ltd., Wolfboro Falls.

Hunt, August. The Mysteries of Avalon: A Primer on Arthurian Druidism. 2011

Jackson, Kenneth Hurlstone. Arthur’s Battle of Breguoin in Antiquity 23, 1949, 48-9.

______________________, Once Again Arthur’s Battles in Modern Philology 43, 1945-6, 45-57.

______________________, The Site of Mount Badon in Journal of Celtic Studies II, 1953-8, 152-5.

Jarman, A.O.H. Aneirin: Y Gododdin, Britain’s Oldest Herioc Poem. 2005, Gomer Press, Llandysul.

____________, Llyfr Du Caerfyrddin. 1982, University of Wales Press, Cardiff.

Jarrett, Michael G. Non-legionary Troops In Roman Britain: Part One, The Units in Britannia, Vol. XXV, 1995, 35-77.

Kennedy, W.N. Remarks on the Ancient Barrier Called ‘The Catrail’, with Plans in Proceedings of the Society of Antiquaries of Scotland, 1857-1860, 117-121.

Kibler, William W. & Carleton W. Carroll (tr). Chretien De Troyes: Arthurian Romances. 1991, Penguin Books.

Langham, Mike & Colin Wells. Buxton: A Pictorial History. 1993, Phillimore, Chichester.

Leach, John. The Smith God in Roman Britain in Archaeologia Aeliana Series 4 Volume 40, 1962, 171-184.

Loomis, Roger Sherman. Arthurian Literature in the Middle Ages: A Collaborative History. 1959, Oxford University Press, Oxford.

Mattaraso, P.M. The Quest of the Holy Grail. 1969, Penguin Books.

Mawr, Allen. The Place-Names of Northumberland and Durham. 1920, Cambridge University Press, Cambridge.

McCarthy, Mike. Roman Carlisle & the Lands of the Solway. 2002, Tempus Publishing Ltd., Stroud.

Mills, A.D. A Dictionary of English Place-Names. 1991, Oxford University Press, Oxford.

Morris, John. Nennius: British History and The Welsh Annals. 1980, Phillimore, London and Chichester.

___________. The Age of Arthur: A History of the British Isles from 350-650. 1995, Phoenix, London.

Nicolson, Joseph & Richard Burn. History & Antiquities of the County of Westmorland and Cumberland, Vol. 2. 1977.

Ottaway, Patrick. Roman York. 2004, Tempus.

Padel, O.J., & D.N. Parsons (eds). A Commodity of Good Names: Essays in Honour of Margaret Gelling. 2008, Shaun Tyas, Stamford.

Radford, C.A.R. The Early Inscriptions of Dumnonia. 1975, Cornwall Archaeological Society, Redruth.

Rivet, A.L.F., & Colin Smith. The Place-Names of Roman Britain. 1982, B.T. Batsford Ltd., London.

Ross, Anne. Pagan Celtic Britain: Studies in Iconography and Tradition. 1996, Academy Chicago Publishers, Chicago.

Rowland, Jenny. Early Welsh Saga Poetry: a Study and Edition of the Englynion. 1990, D. S. Brewer, Cambridge.

Shotter, David. Romans and Britons in North-West England. 1993, Centre for North-West Regional Studies, University of Lancaster.

Thomas, Charles. And Shall These Mute Stones Speak? Post-Roman Inscriptions in Western Britain. 1994, University of Wales Press, Cardiff.

Tolstoy, Nikolai. Nennius, Chapter Fifty-Six in Bulletin of the Board of Celtic Studies Vol. 19, 1960-2.

Walker, J., & L. Walker & R. Sheppard, assisted by J. Brown, and K. Swainson. Buxton, The Natural Baths: An Assessment for High Peak Borough Council, 2nd edn. Nottingham: Trent & Peak Archaeological Trust, May 1994.

Watson, Godfrey & Goodwife Hot & Others. Northumberland’s Past as Shown in its Place Names. 1970, Oriel Press, Newcastle-Upon-Tyne.

Watson, William J. The History of the Celtic Place-Names of Scotland. 1926, Edinburgh and London.

Williams, Ifor. Canu Aneirin: Poetical Work of Aneirin. 1938, Cardiff.

___________, Enawau Lleoedd 40. 1945, Liverpool.

___________ (ed). The Poems of Taliesin. 1968, The Dublin Institute for Advanced Studies, Dublin.

Winterbottom, Michael. Gildas: The Ruin of Britain and Other Works. 1978, Phillimore, London and Chichester.

Wright, R.P., & K.H. Jackson. A Late Inscription From Wroxeter in The Antiquaries Journal, Vol. XLVIII, 1968

Zimmer, Stefan. The Name of Arthur – a New Etymology in Journal of Celtic Linguistics, 13, 2009, pp. 131-6.

ONLINE RESOURCES

http://darkavalonbooks.posterous.com

http://www.vortigernstudies.org.uk/vortigernhomepage.htm

http://www.britannia.com/history/index.html

http://www.mun.ca/mst/heroicage/

http://www.pastscape.org.uk/

http://www.rcahms.gov.uk/

http://www.rcahmw.gov.uk/hi/rng/Home/

http://www.roman-britain.org

http://www.ucc.ie/celt/

http://www.ucl.ac.uk/archaeology/cisp/database/

Shadows in the Mist: The Quest for a Historical King Arthur

Wednesday, October 9, 2024

COMPLIMENTARY COPY OF MY FINAL (!) ARTHUR NONFICTION BOOK

No comments:

Post a Comment