The Druids and King Arthur

The Druids
and King Arthur
This page intentionally left blank
The Druids and

King Arthur
A New View of Early Britain
ROBIN MELROSE
McFarland & Company, Inc., Publishers

Jefferson, North Carolina, and London
Illustrations by Diana Gardner
LIBRARY
OF
CONGRESS CATALOGUING-IN-PUBLICATION DATA
Melrose, Robin.
The Druids and King Arthur : a new view of early Britain /
Robin Melrose.
p.
cm.
Includes bibliographical references (p. ) and index.
ISBN 978-0-7864-5890-5
softcover : 50# alkaline paper
. Druids and Druidism. 2. Arthur, King. 3. Great
Britain History Anglo-Saxon period, 4491066. I. Title.
BL910.M45 2011
942.01' 4 dc22
2010043915
British Library cataloguing data are available
2011 Robin Melrose. All rights reserved
No part of this book may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying
or recording, or by any information storage and retrieval system,
without permission in writing from the publisher.
Front cover: The rider-god known as the Thracian Horseman,
found on the Lincolnshire/Nottinghamshire border, near the
village of Brough (Roman Crocolana). British Museum.
Manufactured in the United States of America
McFarland & Company, Inc., Publishers

Box 6, Jeerson, North Carolina 28640
www.mcfarlandpub.com
Table of Contents
Introduction
Chapter 1 The Dragon Star
Chapter 2 The Severed Head and the Bone Cave:

Religion in Roman Britain
18
Chapter 3 Arthurs Voyage: The Spoils of Annwn
42
Chapter 4 Magic Mounds, Sea People and Shape-Shifters:

The Wonderful World of the Mabinogion
66
Chapter 5 Mounds, Mounds, Mounds: Rubbish Heaps,

Hillforts and the Prehistory of Southern England
86
Chapter 6 Visitors from the East
114
Chapter 7 Brutus of Troy Town
155
Chapter 8 Arthur, King of Wessex?
168
Chapter Notes
193
Bibliography
203
Index
211
Introduction
Many books have been written about the Druids, and even more have
been written about King Arthur. However, this book is not really about
the Druids or King Arthur, but about where the people we call the Druids
came from, where the story of Arthur came from, and the role played by
the Druids and Arthur in the creation of a new order after the end of Roman
rule. To understand the beliefs of the Druids, we will be looking at the religion of Roman Britain, an Old Welsh poem called The Spoils of Annwn,
and four medieval Welsh tales known as the Mabinogion. To understand
where the Druids originated, we will be investigating the archeology and
history of Salisbury Plain, the site of Stonehenge, and surrounding areas
from 2300 B.C. until the arrival of the Celts; and we will be exploring legends of the foundation of Britain, the cult of the Thracian Horseman, the
oracle of Dodona who listened to the rustling of oak leaves, the lore and
legends surrounding the star Arcturus and Arcas the mythical founder of
Arcadia, and some basic principles of prehistoric astronomy. From this rich
brew we will demonstrate that the Druids originated from central or eastern
Europe in around 850 B.C., bringing with them the cult of an underworld
deity, a belief in reincarnation, and a keen interest in astronomy. In the
nal chapter we will show that while the Druids tried to protect the people
of Britain from the comet that appeared in A.D. 539 and the outbreak of
plague that followed it with the monster-slaying Arcturus/Arthur, at the
same time a descendant of the families that had ruled Salisbury Plain for
so long joined forces with another Briton to establish the Kingdom of Wessex and so maintain a continuity that was only broken by the Norman Conquest in 1066. The Druids are usually interpreted as oak-seers, but I will
argue that they were in fact doorkeepers who, with their knowledge of
the doors to the underworld and the celestial realms, ruled over the people
of southern England for over a thousand years and guided them through
the transition from Roman Britain to Saxon England.
1
Introduction
In researching this book, I have drawn on the works of a large number

of writers and scholars from the dawn of European literature to the present
day. As far as primary sources go, I have relied on numerous classical Greek
and Roman authors, from Homer and Hesiod to Julius Caesar, Pliny the
Elder and Tacitus, for their insights into the Druids, the Celts and a host of
other subjects. For insights into early British religion, mythology and history,
I have consulted works in Latin, including Nenniuss History of the Britons
and Geoffrey of Monmouths History of the Kings of Britain, and works in
Welsh, including the mysterious and haunting poem The Spoils of Annwn,
which has an early mention of Arthur, and the mythological tales of the
medieval Welsh Mabinogion, which may provide the key to early British history as well as further information on the Druids and Arthur.
As for secondary sources, I am indebted to the work of the archeologist
Barry Cunliffe, who has not only carried out excavations of important sites
like the hillfort of Danebury in Hampshire, but has in all his writings
tried to put a human face to the people whose dwelling-places he has excavated, and brought alive the prehistory of southern England. Another
scholar whose work has contributed to this book is Miranda Green: her
output on the Celts has been prodigious, and I have frequently turned to
her two books on Celtic art and the role of animals in Celtic life and myth
for guidance. I have also made considerable use of John Kochs Celtic Culture: A Historical Encyclopedia, an indispensable work for someone with
an interest in the Celts.
But this book is not only about the Celts, and The All-Knowing God,
by the Italian scholar Raffaele Pettazzoni, rst published in 1955, has proved
an invaluable resource in understanding the religion of the Thracians and
the Thracian Horseman. Graham Andersons King Arthur in Antiquity has
proposed a number of Greek, Macedonian and Scythian sources for the
mythology of Arthur and opened up some novel lines of research; Thomas
McEvilley, in The Shape of Ancient Thought, has provided a plausible framework for the Druid belief in reincarnation; and both Richard Hinckley
Allen (Star Names: Their Lore and Meaning) and F. Graham Millar (The
Celestial David and Goliath) have made me understand how the Druids
and the prehistoric people of southern England may have viewed the heavens. Finally, in the Arthurian sections I have relied on the work of Frank
D. Reno (Historic Figures of the Arthurian Era) for the historical background
to Arthur, Thomas Green (Concepts of Arthur) for the mythological Arthur,
and Barbara Yorke (Wessex in the Early Middle Ages) for the early history
of Wessex.
Introduction
Ultimately I have used all these writers and scholars very much in my
own way, drawing conclusions that those among them who are still alive
might well disagree with. In coming to these conclusions, I had a number
of eureka moments: one when I read in Gibbons Decline and Fall of the
Roman Empire about the comets in the 530s and the ensuing plague, and
linked that to the story of how Uther Pendragon got his name; another
when I read in Cunliffes Iron Age Communities that the All Cannings Cross
pottery of Wiltshire could have been brought there by people from eastern
France about 800 B.C.; and a third when I read in Cunliffe about the horseand-dog burials at Danebury and discovered they might have a Druid connection. The conclusions I have drawn may well be wrong, but I have
tried to provide solid evidence for them wherever that was possible, and
weave all the disparate elements into a coherent narrative that tells the
story of southern England from the time of Stonehenge to the founding
of the kingdom of Wessex in the 6th century A.D., accounting for the
Druids and the story of Arthur. At times I have resorted to etymology (the
history of the origins of words) to prove a point I have a professional
interest in language using the Indo-European Etymological Dictionary of
the great German philologist Julius Pokorny, and Ranko Matasovics Etymological Lexicon of Proto-Celtic, as well as Kochs excellent etymological
notes. Sometimes when history, literature and archeology are silent on a
particular point, language is the only resource we have for peering into the
past. Words have a story to tell, if only we can nd a way of making them
talk, and I am always happy to hear any story that will unlock the secrets
of the distant past. My greatest regret, as a student of language, is that, as
languages have disappeared from southern England, so many stories have
been lost, so many ways of understanding the world have vanished. This
book has been written in an attempt to recover these lost stories, these
vanished ways of viewing the world. If I have succeeded in this endeavor,
then it has all been worthwhile.
Chapter 1
The Dragon Star

The Son of the Dragon Star
Around A.D. 1136 the Welsh priest Geoffrey of Monmouth published
his work Historia Regum Britanniae (History of the Kings of Britain), which
launched the literary career of one of Britains most famous sons, King
Arthur. Geoffreys History is a rich source for the legendary history of
Britain, and we will return to it frequently in the course of our investigation, but for the moment I want to focus on Arthur and his father, Uther
Pendragon. According to Geoffrey, 1 Uther got his name Pendragon when
there appeared a star of wonderful magnitude and brightness, darting
forth a ray, at the end of which was a globe of re in form of a dragon,
out of whose mouth issued forth two rays; one of which seemed to stretch
out itself beyond the extent of Gaul, the other towards the Irish Sea, and
ended in seven lesser rays. Troubled by the star, Uther ordered Merlin to
be called, and Merlin explained the star in these terms: The star, and the
ery dragon under it, signies yourself, and the ray extending towards the
Gallic coast portends that you shall have a most potent son, to whose
power all those kingdoms shall be subject over which the ray reaches. But
the other ray signies a daughter, whose sons and grandsons shall successively enjoy the kingdom of Britain. When Uther became king, he remembered Merlins explanation of the star and
commanded two dragons to be made of gold, in likeness of the dragon
which he had seen at the ray of the star. As soon as they were nished,
which was done with a wonderful nicety of workmanship, he made a present of one to the cathedral church of Winchester, but reserved the other
for himself, to be carried along with him to his wars. From this time,
therefore, he was called Uther Pendragon, which in the British tongue signies the dragons head; the occasion of this appellation being Merlins
predicting, from the appearance of a dragon, that he should be king.
The Druids and King Arthur
Geoffreys explanation of the meaning of Pendragon is not correct

it actually means Chief Dragon but it does not detract from what is a
great story. But is it only a story? Edward Gibbon, in The History of the
Decline and Fall of the Roman Empire,2 records that in the fth year of the
reign of the Eastern Roman Emperor Justinian (A.D. 527A.D. 565), in the
month of September,
a comet was seen during twenty days in the western quarter of the heavens, and which shot its rays into the north (A.D. 531). Eight years afterwards (A.D. 539), while the sun was in Capricorn, another comet appeared
to follow in the sagittary; the size was gradually increasing; the head was
in the east, the tail in the west, and it remained visible above forty days.
The nations, who gazed with astonishment, expected wars and calamities
from their baleful inuence; and these expectations were abundantly
fullled.
Abundantly fullled because in A.D. 542 an outbreak of plague reached

Constantinople from Egypt the so-called Plague of Justinian, which may
have killed up to 100 million people across the world, and reduced the
population of the eastern Mediterranean by a quarter.
Geoffreys description of the star of wonderful magnitude sounds
rather similar to Gibbons description of the comet of A.D. 539. It is unclear
whether Geoffrey would have read Procopius, the Byzantine Greek historian who wrote a history of Justinian, but we can take it that the event
Geoffrey describes is genuine, and it tells us a great deal not only about
Uther, but also about his son Arthur.
Geoffrey, of course, did not invent Arthur he had been known since
at least the time of Historia Brittonum (History of the Britons), a work attributed to the Welsh monk Nennius and written around A.D. 830. Arthur is
mentioned twice in this work, the rst time3 in relation to a series of twelve
battles against the Saxons, the second time4 in a list of marvels or wonders.
Many of the battles are in places that cannot be identied, but those that
can be located range widely over the island of Britain: the site of the seventh
battle, Coit Celidon, is the Caledonian Forest, which once covered the
uplands of Scotland; the ninth battle was waged at the City of Legions,
either Caerleon, near Newport in Wales, or Chester in Cheshire; and the
twelfth battle was at Mons Badonicus (Mount Badon), thought to be somewhere in Oxfordshire or Wiltshire (see Chapter 8).
The rst marvel that mentions Arthur is in a region called Buelt
(Builth Wells, in Powys, eastern Wales). Nennius says there is a mound
of stones there and one stone placed above the pile with the pawprint of
1 The Dragon Star
a dog in it. When Cabal, who was the dog of Arthur, was hunting the
boar Troynt, he impressed his print in the stone, and afterwards Arthur
assembled a stone mound under the stone with the print of his dog, so it
is called Carn Cabal. The second wonder is in the region called Ercing
(now in western Herefordshire): A tomb is located there next to a spring
called Licat Amr; and the name of the man who is buried in the tomb was
called Amr. He was the son of Arthur the soldier, and Arthur himself
killed and buried him in that very place. Men come to measure the grave,
and nd it sometimes six feet in length, sometimes nine, sometimes twelve,
sometimes fteen.
These two chapters from Nennius provide a good summary of the
early Arthur, part warrior, part wonder-worker. A better picture of this
early Arthur can be built up if we look at Welsh texts composed before
the time of Geoffrey of Monmouth, or at least not inuenced by Geoffreys
Arthur. For a fuller discussion of these texts, the reader is directed to
Thomas Greens Concepts of Arthur, and to his excellent website
http://www.arthuriana.co.uk, where much of the following information
was gleaned (see in particular A Bibliographic Guide to Welsh Arthurian
Literature). One of the earliest references to Arthur is in the Marwnad
Cynddylan (The Death-song of Cynddylan), composed immediately after
the death of Cynddylan, ruler of Powys, in A.D. 655. In this poem it is
suggested that Cynddylan and his brothers might be seen as whelps of
great Arthur, a mighty fortress. A similar reference to Arthur is also found
in Y Gododdin, composed sometime between A.D. 800 and A.D. 1000 to
commemorate warriors who died ghting the Angles of Deira and Bernicia
at a place called Catraeth in about A.D. 600: it is said that the warrior
Gwawddur fed black ravens on the rampart of a fort, although he was no
Arthur.
Both of these references seem to depict Arthur as a peerless warrior,
but this is only one aspect of the early Arthur. In the enigmatic poem The
Spoils of Annwn, attributed to Taliesin and possibly composed in A.D. 800
or earlier, Arthur appears as part of a group of people, including the poet
himself, making a journey to Annwn, the Welsh Otherworld (this poem
will be considered in detail in Chapter 3). It is unclear what Arthurs role
is in this journey all we know is that, in the words of the mournful
refrain, except seven none rose from Annwn.5
More aspects of Arthur emerge in Pa gur yv y porthaur? (What man
is the gatekeeper/porter?), composed sometime between A.D. 800 and A.D.
1000. In this poem,6 Arthur is interrogated by a gatekeeper or porter who
will not let him in unless he vouches for those traveling with him. His
companions include Mabon son of Modron, who gures in our discussion
of the next early Arthurian work; Manawydan son of Llyr, who appears
in two branches of the Welsh mythological work known as the Mabinogion
(see Chapter 4); and Cai, better known as Sir Kay, Arthurs foster-brother
and one of the rst Knights of the Round Table. Cai tells us that Arthur
caused blood to ow/ In the hall of Wrnach [or Afarnach]/ Fighting with
a witch; and on the heights of Eidyn/ He fought with champions [or
dog-heads]. Arthur then informs us of Cais exploits: when he would
drink from a horn,/ He would drink as much as four; he slew nine
witches; he went to Ynys Mon/ To destroy lions; but Little protection
did his shield offer/ Against Palugs Cat. There are also accounts of what
sound like real battles, but we get the impression that Arthur and Cai did
not only ght historical foes, but also were also pitted against witches and
monsters.
This impression is reinforced in the last early Arthurian work we will
consider, Culhwch and Olwen.7 This work is thought to have been composed
in the late 11th century, and is the longest of the Arthurian tales predating
Geoffrey of Monmouth. The hero of the story is Culhwch Pig-Run,
because he was born in a swines burrow and the story begins when Culhwchs stepmother tells him that he is destined to marry Olwen, daughter
of the giant Ysbaddaden. Culhwch instantly falls in love with Olwen, and
rides off to see his cousin Arthur to ask him to help him nd her. We learn
that Arthurs court is at Celliwig (Forest-Grove) in Cornwall, and we are
introduced to Arthurs courtiers, whom Culhwch calls on for help one by
one. This list of warriors he calls on is very long and at times downright
silly, as if the copyist had got bored and decided to have some fun, and
includes some obscure names as well as some interesting ones. Among the
interesting names are Cai, mentioned also in Pa gur; Bedwyr, also mentioned
in Pa gur (the Sir Bedivere of Arthurian romances); Gwyn son of Nudd,
who is usually associated with the Otherworldhis father Nudd is the Celtic
god Nodons or Nodens (see Chapter 2); Gwalchmei son of Gwyar, who is
the Sir Gawain of later literature; Gwenhwyvar Chief of Queens, the Guinevere of later tradition, whose name is usually translated as White Phantom
or White Fairy; and Lludd Llaw Eraint (Lludd Silver-Hand), the god
Nodons, and the Welsh equivalent of the Irish Nuada, the rst king of the
Tuatha De Danann (Peoples of the Goddess Danu), a race of people in
Irish mythology, thought to represent the deities of preChristian Ireland.
Arthur then orders some of his warriors, including Cai and Bedwyr,
1 The Dragon Star
to help Culhwch nd the giant Ysbaddaden and his beautiful daughter,

and they set off. Eventually, of course, they nd Ysbaddaden, and Culhwch
asks for the hand of his daughter Olwen. However, the giant is reluctant
to grant Culhwchs request, since he is destined to die on the day Olwen
marries. After three days he nally agrees that Culhwch and Olwen can
marry but only if Culhwch fullls a long list of impossible-sounding
tasks. The rst task is carried out immediately Cai kills the giant Gwrnach and carries off his sword then they return to Arthurs court.
The rest of the tale is taken up by several of the tasks, most notably
the hunt for the great boar Twrch Trwyth, and the preparations that must
be made before this hunt can be successfully undertaken. The hunt for the
great boar is impossible without the dog Drudwyn, and the only person
who can hunt with Drudwyn is Mabon son of Modron, so Arthur and his
men set off rst to nd Mabon. Mabon is the Celtic god Maponos (DivineSon), and Modron is the goddess Matrona (Divine-Mother) see Chapter 2 for further discussion and he is a prisoner somewhere, after being
taken from his mother when he was three nights old. In order to nd
Mabon, Cai and Arthurs interpreter Gwyhyr must seek his whereabouts
from a series of ever older animals: the Ousel of Cilgwri, the Stag of Redynvre, the Owl of Cwm Cawlwyd, the Eagle of Gwern Abwy, and the oldest
of all, the Salmon of Llyn Llyw, who tells them that Mabon is imprisoned
in the walls of Gloucester. Cai and Gwrhyr inform Arthur, who storms
Gloucester with his warriors and frees Mabon.
A number of tasks are carried out before the hunt for the great boar
is described, and one of these in particular stands out as signicant not
only to the story of Arthur, but to British and Welsh mythology in general.
Ysbaddaden had asked Culhwch to get the cauldron of Diwrnach Wyddel
(Diwrnach the Irishman), the steward of Odgar the son of Aedd, king of
Ireland, to boil the meat for his marriage feast. To fulll this task, Arthur
sent an embassy to ask for the cauldron, but the request was refused, so
he boarded his ship Prydwen and sailed to Ireland, where Bedwyr slew
Diwrnach and seized the cauldron. We will return to this important cauldron in Chapters 2 and 3.
The hunt for the great boar probably has its roots in Celtic mythology
and is even referred to in Historia Brittonum, Chapter 73, where the boar is
called Troynt, and Arthurs dog, as in Culhwch and Olwen, is Cabal, which,
curiously, means Horse. In Culhwch and Olwen this hunt is a momentous
event, with Arthur summoning all his warriors from the three islands of
Britain, France, Armorica, and Normandy. This mighty force then sails to
10
Ireland and reaches the place where the great boar is living with his seven
young pigs. The battle wages for over a week, and we learn that Twrch Trwyth
was once a king transformed into a boar for his sins. The battle then switches
to the island of Britainthe great boar can obviously swim and Twrch
Trwyth slays a large number of Arthurs warriors, and Arthur and his men
in turn kill a number of the great boars piglets. In the end they manage to
achieve their objectiveto get the comb, and the razor, and the scissors,
which are between the two ears of Twrch Trwyth, but the great boar escapes
into the deep sea off Cornwall.
The Star Watchers

The early Welsh material on Arthur shows that he and his men were
not only mighty warriors, but regularly did battle with various monsters
and giants. But this book is not only about Arthur it also investigates
that mysterious priesthood called the Druids. The Druids and Arthur may
seem to have little in common, but there is a link between the two: Arthurs
father, according to Geoffrey of Monmouth, was a man who was inspired
by a comet, and the Druids counted among their many accomplishments
an interest in astronomy.
All our knowledge of the Druids comes from classical writers. Perhaps
the most extensive description of them comes from Julius Caesar, in the
book he wrote about his campaigns in Gaul and Britain, Commentarii de
Bello Gallico (Commentaries on the Gallic War). He says of the Druids8:
Throughout all Gaul there are two orders of those men who are of any
rank and dignity: ... of these two orders, one is that of the Druids, the
other that of the knights. The former are engaged in things sacred, conduct the public and the private sacrices, and interpret all matters of religion. To these a large number of the young men resort for the purpose of
instruction, and they [the Druids] are in great honor among them. For
they determine respecting almost all controversies, public and private; and
if any crime has been perpetrated, if murder has been committed, if there
be any dispute about an inheritance, if any about boundaries, these same
persons decide it; they decree rewards and punishments; if any one, either
in a private or public capacity, has not submitted to their decision, they
interdict him from the sacrices. This among them is the most heavy punishment. Those who have been thus interdicted are esteemed in the number of the impious and the criminal: all shun them, and avoid their society
and conversation, lest they receive some evil from their contact; nor is justice administered to them when seeking it, nor is any dignity bestowed on
1 The Dragon Star
11
them. Over all these Druids one presides, who possesses supreme authority
among them. Upon his death, if any individual among the rest is pre-eminent in dignity, he succeeds; but, if there are many equal, the election is
made by the suffrages of the Druids; sometimes they even contend for the
presidency with arms. These assemble at a xed period of the year in a
consecrated place in the territories of the Carnutes, which is reckoned the
central region of the whole of Gaul. Hither all, who have disputes, assemble from every part, and submit to their decrees and determinations. This
institution is supposed to have been devised in Britain, and to have been
brought over from it into Gaul; and now those who desire to gain a more
accurate knowledge of that system generally proceed thither for the purpose of studying it.
From this we learn that the Druids functioned as judges in Gaul, that
there was a supreme Druid, that there was a general assembly of Druids
held in the territories of the Carnutes, who lived between the Seine and
Loire in present-day France, and, most signicantly, that the institution
was devised in Britain, where Gauls went to study the ner points of
Druidism.
Caesar continues in Chapter 14:
The Druids do not go to war, nor pay tribute together with the rest; they
have an exemption from military service and a dispensation in all matters.
Induced by such great advantages, many embrace this profession of their
own accord, and [many] are sent to it by their parents and relations. They
are said there to learn by heart a great number of verses; accordingly some
remain in the course of training twenty years. Nor do they regard it lawful
to commit these to writing, though in almost all other matters, in their
public and private transactions, they use Greek characters. That practice
they seem to me to have adopted for two reasons; because they neither
desire their doctrines to be divulged among the mass of the people, nor
those who learn, to devote themselves the less to the efforts of memory,
relying on writing; since it generally occurs to most men, that, in their
dependence on writing, they relax their diligence in learning thoroughly,
and their employment of the memory. They wish to inculcate this as one
of their leading tenets, that souls do not become extinct, but pass after
death from one body to another, and they think that men by this tenet are
in a great degree excited to valor, the fear of death being disregarded. They
likewise discuss and impart to the youth many things respecting the stars
and their motion, respecting the extent of the world and of our earth,
respecting the nature of things, respecting the power and the majesty of
the immortal gods.
Here we are told that the Druids learn by heart a great number of verses,
that to become a Druid requires twenty years of training, that they believe
12
souls do not become extinct, but pass after death from one body to
another, and that they impart to the youth many things respecting the
stars and their motion, respecting the extent of the world and of our earth,
respecting the nature of things.
In Chapter 17 Caesar says:
They worship as their divinity, Mercury in particular, and have many
images of him, and regard him as the inventor of all arts, they consider
him the guide of their journeys and marches, and believe him to have great
inuence over the acquisition of gain and mercantile transactions. Next to
him they worship Apollo, and Mars, and Jupiter, and Minerva; respecting
these deities they have for the most part the same belief as other nations:
that Apollo averts diseases, that Minerva imparts the invention of manufactures, that Jupiter possesses the sovereignty of the heavenly powers, that
Mars presides over wars.
This is an example of what is called interpretatio Romana, in which Romans

took the gods of conquered people and interpreted them as Roman gods.
We will return to the gods Caesar mentions in Chapter 2 and throughout
the book, but for the moment it will be enough to say that Mercury (the
Greek Hermes) was the messenger of the gods and the god of trade, prot
and commerce; Apollo was the god of light and the sun, of prophecy,
archery, medicine, healing, music, poetry and the arts; Mars was the god
of war; Jupiter was the king of the gods; and Minerva was the virgin goddess of warriors, poetry, medicine, wisdom, commerce, crafts, magic, and
the inventor of music.
The nal piece of information that Caesar gives us is in Chapter 18:
All the Gauls assert that they are descended from the god Dis, and say
that this tradition has been handed down by the Druids. For that reason
they compute the divisions of every season, not by the number of days,
but of nights; they keep birthdays and the beginnings of months and years
in such an order that the day follows the night. Dis or Dis Pater is the
Roman god of the underworld, whose functions were later taken over from
Pluto or Hades, and will be explored further in Chapter 2 and in subsequent chapters.
The next classical writer to offer observations on the Druids is the
Greek historian Diodorus Siculus, who ourished in the 1st century B.C.
and who, as his name indicates, was born in Sicily. In his Historical
Library,9 he is discussing the Gauls:
Among [the Gauls] are also to be found lyric poets whom they call Bards.
These men sing to the accompaniment of instruments which are like lyres,
1 The Dragon Star
13
and their songs may be either of praise or of obloquy. Philosophers, as we

may call them, and men learned in religious affairs are unusually honoured
among them and are called by them Druids. The Gauls likewise make use
of diviners, accounting them worthy of high approbation, and these men
foretell the future by means of the ight or cries of birds and of the
slaughter of sacred animals, and they have all the multitude subservient to
them. They also observe a custom which is especially astonishing and
incredible, in case they are taking thought with respect to matters of great
concern; for in such cases they devote to death a human being and plunge
a dagger into him in the region above the diaphragm, and when the
stricken victim has fallen they read the future from the manner of his fall
and from the twitching of his limbs, as well as from the gushing of the
blood, having learned to place condence in an ancient and long-continued practice of observing such matters. And it is a custom of theirs that no
one should perform a sacrice without a philosopher; for thank-offerings
should be rendered to the gods, they say, by the hands of men who are
experienced in the nature of the divine, and who speak, as it were, the language of the gods, and it is also through the mediation of such men, they
think, that blessings likewise should be sought. Nor is it only in the exigencies of peace, but in their wars as well, that they obey, before all others,
these men and their chanting poets, and such obedience is observed not
only by their friends but also by their enemies; many times, for instance,
when two armies approach each other in battle with swords drawn and
spears thrust forward, these men step forth between them and cause them
to cease, as though having cast a spell over certain kinds of wild beasts.
From this we learn that there are lyric poets in Gaul called bards, as well
as philosophers called Druids. The Gauls, he says, also make use of diviners, and practice human sacrice (Caesar also mentions this), but cannot
carry out any sacrices unless a Druid is present. He also claims that if a
Druid steps between two warring armies, the armies will immediately cease
hostilities.
Another Greek historian who touches on the Druids is Strabo (64
B.C.A.D. 24). In his Geography,10 Strabo writes:
Among all the Gallic peoples, generally speaking, there are three sets of
men who are held in exceptional honour; the Bards, the Vates and the
Druids. The Bards are singers and poets; the Vates, diviners and natural
philosophers; while the Druids, in addition to natural philosophy, study
also moral philosophy. The Druids are considered the most just of men,
and on this account they are entrusted with the decision, not only of the
private disputes, but of the public disputes as well; so that, in former
times, they even arbitrated cases of war and made the opponents stop
when they were about to line up for battle, and the murder cases, in particular, had been turned over to them for decision. Further, when there is
14

a big yield from these cases, there is forthcoming a big yield from the land
too, as they think. However, not only the Druids, but others as well, say
that mens souls, and also the universe, are indestructible, although both
re and water will at some time or other prevail over them.
What Strabo tells us is not very different from the information provided
by Caesar and Diodorus Siculus, except that we learn from Strabo that the
diviners of Diodorus Siculus are called Vates.
Two early Christian writers also have something to say about the
Druids. Hippolytus of Rome (c. A.D. 170c. A.D. 236), who was one of
the most prolic writers of the early Church, wrote of the Druids11: The
Keltic Druids apply themselves thoroughly to the Pythagorean philosophy,
being urged to this pursuit by Zamolxis, the slave of Pythagoras, a Thracian
by birth, who came to these parts after the death of Pythagoras, and gave
them the opportunity of studying the system. Hippolytus may have linked
the Druids to the Greek philosopher Pythagoras (c. 570 B.C.c. 495 B.C.)
because, like the Druids, Pythagoras and his followers believed in reincarnation. Zamolxis, said by Hippolytus to be the slave of Pythagoras, was
actually a god of the Getae, or Thracian Dacians, tribes that occupied
regions south and north of the lower Danube.
Clement of Alexandria (c. A.D. 150A.D. 215) was a Christian theologian who does not mention the Druids directly, but has some interesting
points to make12: Alexander, in his book On the Pythagorean Symbols,
relates that Pythagoras was a pupil of Nazaratus the Assyrian..., and will
have it that, in addition to these, Pythagoras was a hearer of the Galatae
and the Brahmins. The Alexander that Clement is referring to is Alexander
Polyhistor (c. 100 B.C.c. 40 B.C.), who wrote a large number of ethnogeographic and philosophical works. Nazaratus the Assyrian may be
Zoroaster, the ancient Iranian prophet who founded the religion known
as Zoroastrianism, and the Galatae or Galatians are a Celtic tribe who settled in Anatolia (now Turkey) around 270 B.C. Although Clement does
not mention the Druids, he does refer to Pythagoras, linked by Hippolytus
to the Druids, and claims in addition that Pythagoras was inspired by
Zoroaster, the Celtic Galatians and the Brahmins, the priestly caste of
India.
Another intriguing snippet of information comes to us from Pomponius Mela, who wrote De situ orbis (Description of the world) in around
A.D. 43. In one passage in Book 3.2.1819, quoted by John Koch,13 Pomponius writes: [The Druids] claim to know the size of the earth and cosmos, the movements of the heavens and stars, and the will of the gods.
1 The Dragon Star
15
They teach, in caves or hidden groves, many things to the nobles in a

course of instruction lasting up to twenty years. Most of this is repeating
what Caesar had written a century earlier, but the claim that they carried
out their teaching in caves or hidden groves is new.
More about the practices of the Druids can be found in the Natural
History of Pliny the Elder (A.D. 23A.D. 79)14:
I must not omit the veneration shown to mistletoe by the provinces of
Gaul. The Druids (the name they give to their priestly caste) hold nothing
more sacred than mistletoe and the tree on which it grows, provided it is a
hard-oak. They also choose groves of hard-oak for its own quality, nor do
they perform any sacred rites without leaves from these trees, so that from
this practice they are called Druids after the Greek word for oak. For they
believe that anything growing on oak trees is sent by heaven and is a sign
that the tree has been chosen by God himself.
Mistletoe, however, is rarely found on hard-oaks, but when it is discovered, it is collected with great respect on the sixth day of the moon. Then,
greeting the moon with a phrase that in their own language means healing all things, the Druids with due religious observance prepare a sacrice
and a banquet beneath a tree, and bring two white bulls whose horns are
bound for the rst time.
A priest in a white robe climbs the tree and with a golden sickle cuts the
mistletoe, which is caught in a white cloak. Then they sacrice the victims
praying that God may make his gift propitious for those to whom he has
given it. They think that mistletoe given in a drink renders any barren
animal fertile and is an antidote for all poisons.
Further Druid practices in the gathering of medicinal plants are

revealed by Pliny when he is discussing selago, also known as lycopodium
(wolf s foot) or r club moss15:
Similar to savin is the herb known as selago. Care is taken to gather it
without the use of iron, the right hand being passed for the purpose
through the left sleeve of the tunic, as though the gatherer were in the act
of committing a theft. The clothing must be white, the feet bare and
washed clean, and a sacrice of bread and wine must be made before gathering it: it is carried also in a new napkin. The Druids of Gaul have pretended that this plant should be carried about the person as a preservative
against accidents of all kinds, and that the smoke of it is extremely good
for all maladies of the eyes.
The Druids veneration of oak-trees and mistletoe is by now well known,

but several other points are less well known and worth mentioning. Pliny
tells us that the Druids will not perform any sacred rites without leaves
from these trees; they collect the mistletoe on the sixth day of the moon;
16
they cut the mistletoe with a golden sickle; they sacrice two white bulls
after gathering the mistletoe; and they gather selago without the use of
iron.
The reports of all these classical writers seem to raise more questions
than they answer, and the last writer to be considered is no different. The
late 4th-century Roman historian Ammianus Marcellinus, in his Roman
History, gives us this very curious history of the Gauls16:
The ancient writers, in doubt as to the earliest origin of the Gauls, have
left an incomplete account of the matter, but later Timagenes, a true
Greek in accuracy as well as language, collected out of various books these
facts that had been long forgotten; which, following his authority, and
avoiding any obscurity, Ishall state clearly and plainly. Some asserted that
the people rst seen in these regions were Aborigines, called Celts from the
name of a beloved king, and Galatae (for so the Greek language terms the
Gauls) from the name of his mother. Others state that the Dorians, following the earlier Hercules, settled in the lands bordering on the Ocean.
The Drysidae [i.e. Druids] say that a part of the people was in fact indigenous, but that others also poured in from the remote islands and the
regions across the Rhine, driven from their homes by continual wars and
by the inundation of the stormy sea. Some assert that after the destruction
of Troy a few of those who ed from the Greeks and were scattered everywhere occupied those regions, which were then deserted.But the inhabitants of those countries afrm this beyond all else, and Ihave also read it
inscribed upon their monuments, that Hercules, the son of Amphitryon,
hastened to destroy the cruel tyrants Geryon and Tauriscus, of whom one
oppressed Spain, the other, Gaul; and having overcome them both that he
took to wife some high-born women and begat numerous children, who
called by their own names the districts which they ruled.
Timagenes was a Greek historian who ourished in the 1st century B.C.,
and wrote a History of Alexander and a History of the Gauls, now lost; Hercules or Heracles is the Greek demi-god and hero; the Dorians are one of
the three tribes of ancient Greece; Amphitryon was the apparent father of
Heracles (his real father was in fact Zeus, the king of the gods); and Geryon
was a fearsome giant who lived at the western end of the Mediterranean,
whose cattle Hercules had to steal in his Tenth Labor.
There are several intriguing details here. Ammianus was apparently
getting some of his history from the much earlier Timagenes, so there may
be something in what he says. He tells us that the rst inhabitants of Gaul
were Celts, though some assert that Dorians may also have settled there;
the Druids, he says, claim that part of the people were indigenous, but
others came from remote islands (possibly Britain or even Ireland), and
1 The Dragon Star
17
from the regions across the Rhine (that is, from what the Romans called
Germania); some of the inhabitants may be descendants of those who ed
the destruction of Troy; and nally, he claims that some Gauls may be
descended from Hercules, who married a Gaulish noblewoman and
fathered a number of children by her. Ammianus seems to support Caesars
contention that the Druids originated in Britain, and also suggests a connection between Gaul and Troy. This last suggestion, which mirrors the
Roman claim that Rome was founded by descendants of the Trojan hero
Aeneas, will be further addressed in Chapter 5 and beyond when we come
to look at the prehistory of southern England.
Caves and Hidden Groves

Pomponius Mela told us that the Druids carried out their instruction
in caves and hidden groves, and this seems to reect the content of their
teachings: a cave sounds like an ideal place to be initiated into the mysteries
of the underworld god that Caesar called Dis Pater, and a hidden grove
would have been a good place to study the natural world, including the
collection and use of plants like mistletoe and selago, and to learn about
the movements of the heavens and stars. In this book I will be attempting
not only to discover what some of these teachings might have been, but
also to discover where the Druids came from, and what this can tell us
about Arthur. I hope to show that the comet that gave Uther Pendragon
his name also gave us Arthur, and that Arthur was one of the last gifts that
the Druids gave to the British people. One of the last because I will demonstrate that the greatest of the Anglo-Saxon kingdoms may have been
founded by the British, and that England itself may therefore be a legacy
of this mysterious priesthood.
Chapter 2
The Severed Head and the Bone Cave

Religion in Roman Britain
The Conquest of Britain and the Shape

of Roman Britain
The Romans invaded and occupied Britain in A.D. 43, and Britain
remained part of the Roman Empire until around A .D. 410, when the
Roman legions withdrew from the island. The invasion of A.D. 43 was
not the rst Roman contact with Britain: Julius Caesar had twice led
his forces to the island in 55 B.C. and 54 B.C., as part of his Gallic wars.
Caesar believed that Britons were supporting the campaigns of the Gauls
against him, and that Belgae from Gaul were eeing to Belgic settlements
in Britain (the Belgae were probably Germanic tribes ruled by a Celtic
elite and living in what is today northern France, Belgium, Luxembourg
and Germany west of the Rhine). So Caesar may have wished to cut off
this support; if so, the rst invasion of 55 B.C. was unsuccessful, since the
Roman forces gained a beachhead on the Kent coast, but achieved little
else.
The second invasion of 54 B. C . was more successful. Caesar was
accompanied by Mandubracius, the king of the Trinovantes, a tribe inhabiting what is now Essex and Suffolk; Mandubracius had sought Caesars
help after his father had been deposed by the British chief Cassivellaunus.
Caesar marched inland from Kent, crossed the Thames, possibly at Westminster, and besieged Cassivellaunus at his hillfort in Hertfordshire. Mandubracius was restored to power, and Cassivellaunus surrendered, agreeing
to pay an annual tribute to Rome. And both Cassivellaunus and even Mandubracius entered into British and Welsh legendary history, as we will see
later in our discussion of the Mabinogion.
Caesar left no troops in Britain, and it was not until A.D. 43 that the
18
2 The Severed Head and the Bone Cave
19
Romans set about occupying the island. The invasion was met with resistance from Togodumnus and Caratacus, sons of the late king of the Catuvellauni, Cunobelinus. (The Catuvellauni inhabited the modern counties
of Hertfordshire, Bedfordshire and south Cambridgeshire, and were probably the tribe to which Cassivellaunus belonged.) The Catuvellauni were
defeated in two battles on the rivers Medway and Thames, but Caratacus
survived and continued to resist the Romans. He joined two tribes in east
Wales called the Silures and Ordovices, and they were defeated in A.D. 51
by the Roman governor Publius Ostorius Scapula. Members of his family
were captured, but Caratacus managed to escape. He ed north to the
Brigantes, but their queen Cartimandua was loyal to the Romans and
handed him over in chains.
The Silures continued to resist the Roman occupation, and were not
nally conquered until about A.D. 76. Meanwhile, the Druids make their
rst appearance in British history. The Romans obviously considered them
a threat, since in A.D. 60 Gaius Suetonius Paulinus destroyed the Druid
center at Mona, now known as Anglesey, in north Wales. The confrontation
between the Romans and Druids is famously described by the Roman historian Tacitus (c. A.D. 56c. A.D. 117) in his Annals 1:
On the shore stood the opposing army with its dense array of armed warriors, while between the ranks dashed women, in black attire like the
Furies, with hair disheveled, waving brands. All around, the Druids, lifting
up their hands to heaven, and pouring forth dreadful imprecations, scared
our soldiers by the unfamiliar sight, so that, as if their limbs were paralyzed, they stood motionless, and exposed to wounds. Then urged by their
generals appeals and mutual encouragements not to quail before a troop of
frenzied women, they bore the standards onwards, smote down all resistance, and wrapped the foe in the ames of his own brands. A force was
next set over the conquered, and their groves, devoted to inhuman superstitions, were destroyed. They deemed it indeed a duty to cover their altars
with the blood of captives and to consult their deities through human
entrails.
However, Gaius Suetonius Paulinus was unable to consolidate his

conquests in Wales, because he had to deal with the uprising led by
Boudica, also known as Boadicea. She was queen of the Iceni tribe of what
is now Norfolk, the wife of Prasutagus, an Icenian king who had ruled as
a nominally independent ally of Rome. When he died, he left his kingdom
jointly to his daughters and the Roman Empire, but his will was ignored:
the kingdom was annexed as if conquered, Boudica was ogged and her
daughters raped. The uprising that followed is described by Tacitus in his
20
Annals, and also by another Roman historian Cassius Dio (c. A.D. 160c.
A.D. 230) in his Roman History. According to Cassius Dio,2 before joining
battle, Boudica made a stirring speech to her assembled troops: When
she had nished speaking, she employed a species of divination, letting a
hare escape from the fold of her dress; and since it ran on what they considered the auspicious side, the whole multitude shouted with pleasure,
and Boudica, raising her hand toward heaven, said: I thank thee, Andraste,
and call upon thee as woman speaking to woman. This must be one of
the earliest references to British religious practice, and has Boudica calling
on Andraste, the Icenian goddess of victory, who may be related to Andarta,
a warrior goddess worshipped in southern Gaul.
These troops included the Iceni and the Trinovantes, and they
destroyed Camulodunum (Colchester in Essex), formerly the capital of
the Trinovantes, but by then a colonia (a settlement for discharged Roman
soldiers) and the site of a temple to the former emperor Claudius, and
routed a Roman legion sent to relieve the settlement. On hearing news of
the revolt, Suetonius hurried to Londinium (London) and, concluding he
did not have the numbers to defend it, evacuated and abandoned it. It
was burnt to the ground, as was Verulamium (St. Albans in Hertfordshire).
It is thought that between 70,000 and 80,000 people were killed in Camulodunum, Londinium and Verulamium. Suetonius, meanwhile, regrouped
his forces in the West Midlands, and despite being heavily outnumbered,
defeated Boudica in the Battle of Watling Street, somewhere between Londinium and Viroconium (Wroxeter in Shropshire). According to Tacitus,
Boudica poisoned herself, while Cassius Dio says she fell sick and died,
and was given a lavish burial.
The destruction of the Druid groves at Mona and the suppression of
Boudicas revolt was of course not the end of the Roman conquest of
Britain. The Romans pushed north and attempted to bring Scotland into
the Empire, but never succeeded, and the northern border of the Empire
became Hadrians Wall, begun in A.D. 122 well short of the Scottish border.
But what did Roman Britain look like? In order to understand religion in
Roman Britain, we need to know a little of how Britain was organized and
administered.
In general Britain was organized along tribal lines, taking the tribes,
as the Romans found them or perceived them, as the basic administrative
units. The following table gives the tribes of Britain in Roman times,
together with the territory they occupied and the name of the Roman capital and any other town in the territory, going from south to north.
Tribe
Territory
Dumnonii
Durotriges
Devon and Cornwall

Dorset, south Wiltshire,
south Somerset
Southern Hampshire
Belgae
Atrebates
Regnenses
Cantiaci
Catuvellauni
Trinovantes
Iceni
Dobunni
Silures
Demetae
Cornovii
Corieltauvi
Brigantes
Parisii
Carvetii
21
Capital and Other Towns
Isca Dumnoniorum (Exeter in Devon)

Durnovaria (Dorchester in Dorset);
Lindinis (Ilchester in Somerset)
Venta Belgarum (Winchester in
southern Hampshire); Magnus
Portus (Portsmouth); Sorviodunum
(Old Sarum)
Northern Hampshire, Surrey Calleva Atrebatum (Silchester in
northern Hampshire); Cunetio
(Mildenhall)
West Sussex
Noviomagus Reginorum (Chichester)
Kent
Durovernum Cantiacorum
(Canterbury)
Hertfordshire, Bedfordshire,
Verulamium (St. Albans in
Buckinghamshire,
Hertfordshire); Durocina
Oxfordshire
(Dorchester-on-Thames)
Essex and Suffolk
Camulodunum (Colchester in Essex)
Norfolk
Venta Icenorum (Caister Saint
Edmunds in Norfolk)
Gloucestershire,
Corinium Dubonnorum (Cirencester
north Somerset
in Gloucestershire); Aquae Sulis
(Bath); Glevum (Gloucester)
Monmouthshire, Breconshire, Venta Silurum (Caerwent in
Glamorganshire
Monmouthshire); Isca Silurum
(Caerleon, north of Newport)
Pembrokeshire,
Moridunum (Carmarthen);
Carmarthenshire
Luentinum (Dolocauthi Gold
Mines)
Staffordshire, Shropshire,
Viroconium Cornoviorum (Wroxeter
Cheshire
in Shropshire); Deva Victrix
(Chester)
East Midlands
Ratae Corieltauvorum (Leicester);
Lindum Colonia (Lincoln)
North of England, Midlands Eboracum (York)
East Yorkshire
Petuaria (Brough)
Cumbria, north Lancashire
Luguvalium (Carlisle)
Religion in Southern Britain: Temples

There are a number of ways we can investigate religion in Roman Britain:
we can look at the archeological remains of Romano-British temples; we can
22
explore sacred spaces other than temples; and we can draw on the evidence
provided by religious images. For reasons which will become apparent in the
course of this book, we will restrict this examination largely to southern
Britain, more specically to an area stretching from Gloucestershire in the
north through Wiltshire and Somerset to Dorset in the south, focusing on
temples, sacred spaces and religious images that help to give us a coherent
picture of religion in this important area of Britain. To start with, Ill be concentrating on six important temples in our target area.
Uley, Gloucestershire
The rst Romano-British temple I want to look at is the temple at
Uley in Gloucestershire, in the territory of the Dobunni (see Maps 1 and
2). Uley now is a small village situated in a wooded valley in the Cotswold
escarpment, but in Roman times the settlement and temple were on West
Hill, above the modern village of Uley to the south and overlooking the
Severn Valley to the west. The nearest towns to Uley were the colonia of
Glevum 12 miles to the north, and the tribal capital Corinium 16 miles to
the east. To the southeast the Fosse Way (the Roman road from Exeter to
Lincoln) connected Bath and Cirencester, while the road running southwest from Gloucester passed along the Severn Valley below West Hill.
According to Oxford Universitys Centre for the Study of Ancient Documents (CSAD)3 which specializes in the study of curse tablets, which
were found in large numbers at Uley the site was a center of worship
before Roman times. In Uley: History, the CSAD website says:
Beneath the Roman temple are the remains of an earlier shrine, a square
timber structure in a subrectangular ditched enclosure, constructed in the
half century preceding the Roman conquest.... A temple was constructed
in stone in the early second century A.D., [and] continued to be maintained and modied for almost three hundred years.... An aisled timber
building with a semicircular annex was erected on the site of the temple
during the fth century and was rebuilt in stone in the early sixth century.
These structures have been interpreted as a church and baptistery, but the
form and parallels of the buildings are uncertain. Carefully buried outside
the annex was the head of the cult statue of Mercury from the temple,
which must have been curated for at least a century after the collapse of
the building.
The god that Mercury represented must have been very important to the
Dobunni for Christians to have kept the head of a pagan deity (see Figure
1).
23
Map 1: Some important Roman towns and settlements in southern England,

showing the modern counties of Wiltshire, Hampshire, Somerset, Dorset,
Gloucestershire and Oxfordshire.
The CSAD website notes (Uley: deity) that the votive deposits
(sacrices) associated with the preRoman sanctuary
give some clue to the character of the preRoman deity. The presence of
weapons suggests a martial aspect. In the Roman period the identity of
this local god ... seems to have been assimilated with the Roman deity
Mercury. Mercury is named on curses, stone altars and metal plaques and
represented in statues and gurines. The major cult statue probably stood
within the cella (inner chamber of the temple). The god was nude and
slightly larger than life-size, accompanied by a ram and cockerel at his
feet.
We do not know the name of the god worshipped at Uley before the
Romans came, but the Roman god Mercury is widely believed to be the
equivalent of the Celtic god Lugus. Alexei Kondratiev4 discusses the evidence we have for Lugus and his Roman counterpart Mercury, starting
with place-names. The name Lugudunon, says Kondratiev,
24
Map 2: Some important Roman temples in southern England, showing modern counties of Wiltshire, Hampshire, Somerset, Dorset and Gloucestershire.
was given to a very large number of sites (Lyons, Loudun, Laon, Liegnitz,
probably Leiden, etc.) from the later Iron Age. In Old Celtic dunon means
fort (the word has modern cognates in Irish dn fort and Welsh din[as]
city), but the Lugu- element can only be explained by a proper name.
We have no dedications to a god by that name at those sites, but a famous
dedication to the Lugoues by the shoemakers guild of Uxama (Osma) in
Spain; another inscription mentioning the Lugoues from Avenches in
Switzerland; and dedications to Lugubus Arquienobus from Orense and
Lugo in Galicia (northwest Spain) all indicate that the name Lugus was
indeed known. Interestingly, in all these cases the name is given in the plural, as though it referred to a group of divinities rather than to a single
god.
As we saw in Chapter 1, Julius Caesar in his Commentaries on the

Gallic War, Book 6, Chapter 17, said that the Gauls worship as their divinity, Mercury in particular, and have many images of him, and regard him
as the inventor of all arts, they consider him the guide of their journeys
and marches, and believe him to have great inuence over the acquisition
25
of gain and mercantile transactions.

This may explain why there are so few
dedications to Lugus in Roman times,
but so many to Mercury throughout the
Romanized Celtic world: well over 400
dedications to Mercury or one of his
common native titles have been found.
His importance in Gaul and Britain far
exceeded anything that the role of Mercury in Roman religion could have warranted.
Representations of Mercury, like
the one at Uley, often depict him with
his usual classical attributes, but as Kondratiev points out, he also has four traits
which are peculiar to the Celtic world.
Figure 1: Head of the statue of
He is linked with high places (for exam- Mercury from the Roman temple, the Arverni, from the Auvergne ple at Uley in Gloucestershire.
region of France, had a gigantic statue British Museum.
of Mercury seated atop their sacred
mountain, the Puy-de-Dme); he is often represented with three heads;
he is often linked to the antlered god called Cernunnos, of whom more
later; and in the Belgic lands, particularly the Rhineland, Mercury is
armed with a spear and accompanied by his consort Rosmerta, The Great
Provider, a goddess of fertility, who will be discussed later in this chapter.
Kondratiev believes that in these specic representations Mercury and
Rosmerta
are particularly linked to the concept of sovereignty. In Iron Age society
the cohesion of a group around a chieftain was secured and given a sacred
recognition by the means of a communal feast, in which a ritual drink
served by the goddess of the land (a role played by a priestess or by the
chieftains consort) was shared, binding all the participants to their land,
their ruler, and each other. Rosmerta was the divine keeper of the drink of
sovereignty, while the spear-wielding Mercury was the archetype of all
rulers, the Otherworldly protector of the earthly king.
The origin of the name Lugus is much debated. It was earlier thought
to derive from the Indo-European root leuk- light, which gives Latin
lux and English light, but as Kondratiev points out, it may well be derived
from the Indo-European root leugh- or lugh oath, which give Old Irish
26
lu(i)ge and Welsh llw. In support of this, he cites the famous Gaulish text
found at Chamalires in 1971. This text, which is the script of a magicoreligious ritual for obtaining the help of Arvernian Maponos in a military
revolt, concludes with the thrice-repeated formula Luge dessumiis [= dexumiis] (By an oath I make them ready), where the echo of the gods name
in the expression luge could hardly have failed to impress itself on a Celticspeakers ear.
Lydney Park, Gloucestershire

The second temple I want to consider is the temple at Lydney Park
in Gloucestershire, built within the ramparts of an old hillfort, and close
to Iron Age and Roman iron mines. It was close to the main Roman road
that ran from Gloucester (Glevum), 18 miles to the east, to the legionary
fortress of Caerleon and to Venta Silurum (Caerwent), the capital of the
Silures, 12 miles to the southwest.
The temple is generally thought to have been built some time after
A.D. 364, and is dedicated to the god Nodens or Nodons, who gave rise
to the mythological characters Gwyn son of Nudd and Llud Llaw Eraint
that we saw in our earlier discussion of Culhwch and Olwen. The website
of the Centre for the Study of Ancient Documents (CSAD) conrms (Lydney: deity and cult) that
the presiding deity at Lydney is named as Nodens on the single curse
tablet from the site, and on two other metal plaques from the site as
M(ars) Nodons and Nudens Mars. The god is also referred to in an abbreviated form (MN) on a mosaic. The name Nodens is Celtic and its etymology may suggest a possible association with catching or trapping. The
god is perhaps also associated with the river Severn and its tidal bore.
Nodens is only otherwise attested on two statuettes found near Lancaster.
Amongst the votive objects found at the site are dog gurines, some
highly schematic, one amongst the most accomplished pieces of bronze
sculpture from Roman Britain.
CSAD hypothesizes that the dog gurines (see Figure 2) may be related
to hunting, or be associated with healing though as well see in this
chapter, dogs can also be associated with the underworld and notes: A
curative role for the temple might also explain the presence of two possible
offerings to the god, the bone representation of a woman and a hollow
bronze arm. The discovery of an oculists stamp (to be stamped into cakes
of eye medicine) also suggests the presence of a healer.
27
Nettleton Shrub, Wiltshire

This Romano-British temple is in the northwest corner of Wiltshire,
10 miles northeast of Aquae Sulis (Bath), to the west of the Fosse Way
which ran from Corinium (Cirencester) to Aquae Sulis, Lindinis (Ilchester)
and Isca Dumnoniorum (Exeter). Three temples have been discovered at
Nettleton, dating from around A.D. 69 to around A.D. 360. There is a rectangular temple (Nettleton 1), 63 ft. by 21 ft., aligned roughly northsouth with its rear built into the valley side. Finds from this site included
a small bronze candlestick in the form of a cockerel; fragments of stone
slabs bearing sculpted reliefs, smaller than life-size, depicting scenes [interpreted as Diana and Hound and Mercury and Rosmerta]; and glassware,
samian and coarse pottery and many bronze coins dating from Hadrianic
times to the late-4th century.5
Then there is an octagonal temple (Nettleton 3) built on the site of
an earlier, possibly preRoman, circular temple (Nettleton 2). The two
successive temples or shrines were enclosed within a walled precinct or
temenos. Several construction phases have been identied, including a
2nd-century circular shrine, a 3rd-century octagonal stone podium, and
the 4th-century octagonal temple.6
From an altarstone it appears that the temple was dedicated to Apollo
Cunomaglus (Apollo the Hound-Lord), though at one point in the 4th
century alternate chambers were blocked and the plan of the building took
on a cruciform aspect, implying that it was used as a church, before reverting to pagan worship.7 These changes of use reect the history of 4th-century Rome : Christianity became the ofcial religion of Rome under
Constantine around A.D. 313, then paganism was briey restored when
Julian became Emperor in A.D. 355. And all this was happening just at the
time when Roman control over Britain was beginning to weaken, as shown
by the rise of Magnus Maximus, the Iberian general who was proclaimed
Western Roman Emperor
by British troops in A.D.
383, and who ruled over
Figure 2: Bronze statuette
of a dog, said to be an
Irish wolf hound, from the
Roman temple at Lydney
Park in Gloucestershire.
28
Gaul, Britain and Spain until A.D. 388, when he was executed at the command of the Roman Senate.
But who was Cunomaglus? He is not known outside this Wiltshire
shrine, but the pairing with Apollo and the link to Diana suggest that he
was a god of hunting and healing, while the presence of a possible Mercury
and Rosmerta relief suggests a link to Lugus.
Pagans Hill, Somerset

This temple is in north Somerset, near the present-day village of
Chew Stoke, on a promontory overlooking the River Chew. According to
the CSAD website (Pagans Hill: History and Buildings), the coins recovered during the excavation suggest that the temple complex was built in
the later third century A.D.... Occupation continued throughout the fourth
century, with a possible interruption in the middle of the century, and
perhaps ceased in the early fth century, the latest datable coin being of
the emperor Arcadius (A.D. 383408). It is not clear what deity was worshipped at the temple. The CSAD website (Pagans Hill: deity and cult)
says that a clue comes from
four sculptural fragments, parts of the torso of a seated dog with a jewelled
collar perhaps originally c. 0.8m in height. Believed at the time of excavation to date to the 15th or 16th centuries, re-evaluation suggests a Roman
date. Dogs are sometimes associated with the god Apollo in healing cults
across the Roman world, being perhaps suitable spirit guides to those in
the grip of a healing trance.
As the CSAD website points out, the temple at Pagans Hill was octagonal,
like the one at Nettleton Shrub, and since the cult at Pagans Hill seems
to have been linked to dogs, it is tempting to speculate that Cunomaglus
was worshipped at both sites.
Maiden Castle, Dorset

Maiden Castle is an Iron Age hillfort one and a half miles south of
Durnovaria (Dorchester) in the territory of the Durotriges. A temple complex was built on the hill in the late 4th century A.D., nearly two hundred
years after the hillfort was abandoned. The English Heritage website
devoted to Maiden Castle says that the temple consisted of a central room,
surrounded by a passage called an ambulatory, with a portico open to the
weather. Close to the temple was an oval hut, thought to have functioned
29
as a shrine. This was built directly over an Iron Age hut, and may show
continued use of an earlier ritual building. A two-roomed building adjacent
to the temple is believed to have been a priests house.
Perhaps the most interesting nd at this temple site was the statue of a
three-horned bull which the excavator, Sir Mortimer Wheeler, named as Tarvos Trigaranus. Tarvos Trigaranus, the Bull with Three Cranes, is best
known from the bas-relief called the Pillar of the Boatmen, which once stood
in a temple in Lutetia (modern Paris). The bas-relief shows the bull with
three cranes perched on his back, standing under a tree. On an adjacent
panel, the Gaulish god Esus is shown cutting down a tree (possibly a willow)
with an axe. A similar representation, this time with no inscription, is found
on a pillar from Trier in Germany where a man with an axe cuts down a tree
in which are sitting three birds and a bulls head. Anne Ross, in Pagan Celtic
Britain, as quoted by David Blamires,8 says that there are several tales and
traditions both in Scotland and in Ireland of the transformation of women,
especially those of a parsimonious nature, into cranes. Ross suggests that since
the crane (or heron) is a metamorphosis undergone or adopted only by
women, the three cranes of Tarvos Trigaranus might be especially concerned
with otherworld goddesses, being both the form and symbol taken by them.
Jordan Hill, Dorset

This temple is 8 miles south of Durnovaria (Dorchester), in the territory of the Durotriges, and was probably built in the 4th century, on a
hill overlooking Weymouth Bay. However, according to the Visit Weymouth website, nds such as animal bones and bull horns suggest the site
had also been used during Iron Age times. Only the temple foundations
and the base of the walls remain, which are over 3 feet thick and enclose
an area of less than 265 square feet. This site is interesting for a nd in
the southeast corner of the structure a shaft around 12 feet deep containing 2 urns, a spearhead and a sword in a stone cist at its base. Above
this cist were deposited 16 layers of ash and charcoal, each containing the
remains of a bird along with a coin and separated from the next layer with
roong slabs. Some sources say the birds included buzzard, raven, starling
and crow, but Green9 says they were all ravens.
As Kondratiev shows, Lugus was connected in some way with ravens.
In the text De Fluviis, attributed to the Greek historian Plutarch (c. A.D.
46A .D. 120), the writer gives us some information about Lugdunum
(Lyons, the Fort of Lugus). We are told that
30

at the time of the founding of the city certain ravens ew down from the
sky, and were interpreted as a good omen. These were not ordinary ravens,
but had some white feathers in their plumage; and they became the focus
of a prophetic shrine where, after an inquirer had made an offering of food
on an elevated platform, a priest would divine the answer to his query
from the behaviour of the ravens as they went after the food. The main
role of these ravens was therefore ... one of Otherworldly contact, which
was made possible by their unusual appearance: although they were traditional examples of blackness, they nevertheless contained their opposite
(whiteness) within themselves, and could thus offer passage between seemingly opposed realms, even as Lugus/Mercury facilitates such passages.
Representations of the unnamed genius loci of Lugdunum (almost certainly
Lugus himself, since he has typical Mercury attributes) show him
accompanied by ravens.
Religion in Southern Britain: Other Sacred Spaces

Alveston Cave, Gloucestershire
The rst sacred space I want to look at is a cave near Alveston, a
village in South Gloucestershire, a mile south of Thornbury and 13 miles
north of Bristol, not far from the Severn Estuary and from the Lydney
Park temple, and therefore in the territory of the Dobunni (see Map 3).
Bones were discovered in the cave in 1998, and in 2001 the archeology
program Time Team carried out an investigation there.10 The Time Team
website explains the nds: The cavers nds included a large number of
bones: animal, including many dog bones, and human, including a skull
and femur. These were to be added to by similar bones unearthed during
Time Teams excavation. Some showed the distinctive signs of having been
butchered at around the time of death. Others showed clear signs of the
people they belonged to having been suddenly and violently killed.
During their investigation of the cave, the Time Team archeologists
came to the conclusion that these people from around two millennia ago
had indeed met sudden deaths and had indeed then been butchered and
eaten. With radiocarbon dating of the bones from the cave suggesting that
they were all buried around 2,000 years ago, at the very end of the Iron
Age or beginning of the Roman occupation, Time Team was looking at
evidence for the most recent case of cannibalism yet found in Britain by
archaeologists. As the Time Team website goes on to point out, clues as
to why these bones were placed in the cave came from the other nds.
31
These included numerous dog bones, as well as the occasional cattle bone,
a possible vertebra of a bear and wooden twigs. Mark Horton said: This
was a highly structured deposit that can only have got there as a result of
some form of ritual activity. This region was an important centre for
underworld cults during the later Iron Age, some of which survived into
the Roman period in particular the Celtic Hound God, Cunomaglus,
was represented as a dog guarding the underworld in local temple sculpture.
It seems from this that Cunomaglus was associated with an underworld

cult which, in the Iron Age at least, may have included human sacrice
and even cannibalism.
Other Ritual Shafts

Mark Horton refers to an underworld cult in the Gloucestershire
region, and it is worth pausing to ask what he means by that. As the nds
at Jordan Hill show, ritual shafts were obviously important in Roman times,
but in fact they have a long history in southern Britain. As Jodie Lewis
says,11 at Wilsford Shaft, Normanton Down, near Stonehenge, a monument
previously interpreted as a simple pond barrow was found to cover an articial shaft, 100 feet deep and 6 feet in diameter. A range of deposits were
recovered, the earliest being Bronze Age in date, comprising a shale ring,
amber beads and bone pins, along with animal bone and organic materials.
The Wiltshire Council website says12 that Wilsford Shaft could have been
constructed as early as 2300 B.C., and quantities of wooden tubs, bowls,
scoops, amber beads, bone pins, thong and rope, and pottery of DeverellRimbury afnity were found there, with a date of around 1450 B.C.
Ashbee, who excavated the shaft, argued (says Lewis) that the ring,
beads and pins should be interpreted as votive deposits, of a type similar
to those found as grave furniture in Wessex in the Early Bronze Age. He
stresses that there is a very strong possibility that many shafts may exist
adjacent to the major round barrow cemeteries of Wessex and that the
chthonic cult of the Iron Age and Roman periods may well have a much
earlier origin.
We will be exploring ritual shafts in more detail in Chapter 5, so for
the moment Ill restrict myself to saying that ritual shafts continued to be
used in the area around Stonehenge into Roman times. Wessex Archaeology
have excavated a Romano-British village at Buttereld Down near Amesbury, not far from Stonehenge. This village, they say,13 was founded in the
2nd century A.D. but was at its largest in the 3rd and 4th centuries.
32
Map 3: Other sites of religious activity in southern England in prehistoric

and Roman times, in the modern counties of Wiltshire and Gloucestershire.
Houses and yards lined the trackways that met in the centre of the village.
One of the tracks ran between the Roman towns of Durnovaria (Dorchester) to the south-west and Calleva Atrebatum (Silchester) to the northeast. The rectangular houses were made of timber and probably had
thatched roofs.... Votive shafts about 10 feet deep and sited at the edge of
the village were used to communicate with the gods of the underworld.
Offerings to the gods were placed in the shafts. Near the centre of the village there are traces of a building built of stone. It is possible that this
building was a temple. A small metal model of a bird may have decorated
the end of a priests sceptre.
The bird at the end of the priests sceptre is unidentied, but it recalls the
ritually deposited birds at Jordan Hill, and the importance of birds in
Romano-British religion.
Orpheus Mosaics
One classical gure closely associated with the underworld is Orpheus,
who is said to have traveled to the underworld after the death of his wife
33
Eurydice and softened the hearts of Hades and Persephone with his music.
There are at least two Orpheus mosaics in southern Britain, one at Woodchester Roman villa in Gloucestershire, the other at Littlecote Roman villa
in Wiltshire.
In The Romans at Woodchester, Graham Thomas, a local historian,
discusses14 the villa and its mosaic. Building probably began during the
reign of Hadrian (A.D. 117A.D. 138), and the villa was built and rebuilt
over two centuries or more. Giles Clarke believes15 that it was unlikely to
have had an ofcial function. He argues that more likely, the villa was
built and lived in by the descendants of the local preRoman tribal leader.
We dont know why it was built there, but Woodchester was not far from
three important Roman towns, Aquae Sulis (Bath), Corinium (Cirencester)
and Glevum (Gloucester), and was already on the path of an ancient road
that ran between Gloucester and Bath.
As Clarke says,16 the area immediately surrounding Woodchester is
remarkably rich archeologically: there are at least seven other villas within
a ve-mile radius. The great mosaic was made around A.D. 325 by craftsmen from Corinium, with the main design based around Orpheus and his
relationship with nature.
The villa was rst excavated in the 1790s by Samuel Lyson, who published the results of his work in Account of the Roman Antiquities Discovered
at Woodchester in the County of Gloucester. He also found a number of very
ne marble sculptural fragments, including the headless statue of Diana
Luna, the moon goddess, with the sacricial bull at her feet, which are
now in the British Museum. The quality of the carving is exceptional for
statues found in British villas and these nds indicate the luxurious character of the villa.
As Katherine Dunbabin says,17 Woodchester was a villa of exceptional
size and grandeur, at least twenty of whose rooms were decorated with
mosaics. The Orpheus pavement occupied the central hall, c. 50 square
feet. The center here was
an octagon originally containing representations of sh, possibly with a
fountain basin or similar feature. Orpheus was therefore moved into the
next encircling ring, where he sat playing his lyre in the midst of a procession of birds. A second ring, separated by concentric borders, contained
the beasts, walking against a plain white ground, with small trees between
them the beasts included a gryphon, a bear, a leopard, a stag, a tigress, a
lion, a lioness, a boar, a dog and an elephant. Beyond this was an outer
ring, with a eshy scroll, growing from the head of Ocean. This circle is
contained in a square, with four column bases in its corners; in the span-
34

drels are pairs of water-nymphs, shown, unusually, against a dark background.
Littlecote Roman villa is in Littlecote Park, near Ramsbury, Wiltshire,

in the Kennet Valley. In Roman times, it would have been close to the
Roman fortied vicus (settlement) of Cunetio, on the banks of the River
Kennet, and at the crossroads formed by two major Roman roads eastwest from Calleva Atrebatum to Aquae Sulis, and southeast to northwest
from Venta Belgarum to Corinium. The settlement, says Wikipedia, may
have begun life as a small short-lived military establishment guarding a
crossing of the River Kennet. This was replaced by local circular farming
huts around A.D. 70 and a Roman-style rectangular building fty years
later.... After another fty years, this was replaced by a large two-storeyed
winged corridor villa with integral bath suite. This building went through
a number of changes over the subsequent centuries, notably a major
rebuilding around A.D. 270.
Around A.D. 360, agricultural activity seems to have ended and the
complex acquired a religious use. A large barn was converted into a courtyard and a very early triconch (three-shells) hall was built alongside with
its own bath suite. Upon its oor was laid a now famous Orpheus mosaic,
rst discovered in 1727 by the Steward of the Littlecote Park estate. This
change of use, from villa to shrine, has been associated with the pagan
revival under the Emperor Julian, which dates from A.D. 362. Many of the
buildings were demolished or fell into decay around A.D. 400, shortly after
the Theodosius made Christianity the ofcial state religion.
Martin Henig describes the mosaic as follows.18 The central theme of
the mosaic, he says, is
Apollo equated with Orpheus, accompanied by Orpheus regular attribute
of a fox surrounded by personications of the seasons (riding upon animals) which belong with Apollo. Heads of Sol in the apses also refer to
Apollo, but the Suns rays project almost to the edge of the panel and may
also be interpreted as pecten (scallop) shells, recalling the belief in the voyage of the soul over the sea to the Blessed Isles. The other section of the
mosaic is mainly geometric, but a panel with sea-panthers again evokes the
marine thiasos (the ecstatic retinue of Dionysus/Bacchus), while a pair of
facing panthers, one on each side of a chalice, is a well-known Bacchic
device. The mystery at Littlecote seems to have exploited the close relationship between Bacchus (Dionysus), Orpheus and Apollo, and to have
been connected with salvation.
Some further explanation is required here. Henig says elsewhere19 that

Bacchus or Dionysus was a savior-god who was envisaged as leading the
35
dead to a life of triumph, just as he had once rescued Ariadne, the daughter
of King Minos of Crete, on Naxos and carried her away with him to share
the delights of the thiasos. The Blessed Isles exist in both Greek and Celtic
mythology as a blissful paradise for heroes and other favored mortals, and
according to Toynbee,20 the scallop-shell symbolizes the souls journey
across the ocean to this paradise. And of course, Apollo, in the form of
Apollo Cunomaglus, was worshipped not far away from Littlecote in the
temple of Nettleton Shrub.
Religious Images
This survey of religious images and inscriptions will start in the north
of the target region, in Gloucestershire, the territory of the Dobunni. I
mentioned earlier that in Gaul the Gaulish Mercury is often accompanied
by a goddess of fertility called Rosmerta, and although there are no dedications to Rosmerta in Britain, there are images thought to be those of
Rosmerta. There is an image on a relief in Gloucester (Glevum) (from the
Shakespeare Inn, Northgate Street) which has been identied as Rosmerta
and Mercury (see Figure 3): Mercury has his cockerel, caduceus and winged
hat, while Rosmerta has a curious sceptre in one hand and a ladle in the
other, poised over a cylindrical wooden, iron-bound bucket on the
ground.21 In another image, from Bath (Aquae Sulis), Mercury is conventional, with petasos (winged hat), purse and caduceus. Apart from Rosmertas own caduceus (which may instead be a wand of authority), she
rests her right hand on a cylindrical receptacle box, casket, or bucket.
The Celtic character of the couple is enhanced by the presence of three
minute genii cucullati at the base of the stone22 (see Figure 4). The mothergoddess is also associated with war, according to Green. At Daglingworth
near Cirencester (Corinium), an image of the Dubonnic mother-goddess
sits with three hooded godlets, two of whom bear swords.23 A relief from
Kingscote, 3 km east of Uley, associates a throned goddess who holds
bread or fruit with a warrior.24
The genii cucullati (hooded spirits) are found in religious sculpture
throughout the Celtic world, from Austria to Britain. Green notes25 that
the British genii cucullati are rather different from their continental counterparts: They are characterized by their triplicate form, though single
ones occasionally occur; they are invariably dwarves and their phallicism
is not stressed. However, there are overt fertility associations. Thus, the
36
gods are frequently in company with the mother-goddesses, especially in

the territory of the Dobunni, and they often carry eggs. They are found
in both the north and the south of Roman Britain. At Housesteads (Vercovicium, on Hadrians Wall), the trio, who come from a small 3rd century
shrine in the vicus of the fort, are swathed in heavy cloaks reaching to their
feet; the face of the central god is masculine while the other two may be
feminine or one may be older, and the other two may be youths.
The southern British genii cucullati, says Green, are more varied in
their associations. They are linked with healing springs at Springhead,
Kent, ... and at Bath (Aquae Sulis).26 Healing may again be their function
at Lower Slaughter, Gloucestershire, where two schematized groups of
cucullati, one with a worshipper, are depicted and where the context of
the well may imply a curative-water association. But the group with the
devotee are associated with the symbols of a rosette and two ravens, which
may have otherworld associations.27 The Dobunnic cucullati are rmly
linked with the mother-goddess. On a relief from Corinium (Cirencester)
a triplet of hooded spirits receives something from a goddess called Cuda
(which Green claims denotes prosperity).28 Another relief from Corinium
depicts the cucullati in company with a mother nursing a cake or fruit;
and two of the spirits carry swords as if to defend her.29 On another
stone a mother-goddess bears fruit and the accompanying cucullatus an
egg; and two triads at Wycomb, near Uley, also bear these fertility symbols.30 The link between the genii cucullati and the mother goddess is
reinforced by the fact that the mother-goddesses also appear in threes (see
Figure 5).
These images from the territory of the Dobunni are signicant for
our understanding of British religion and later Welsh mythology. The
bucket or tub associated with the mother-goddess has been interpreted as
a cauldron,31 and given the link between the cucullati and fertility, healing
and the Otherworld (the Lower Slaughter cucullati come from deep underground in a well), it can be linked to the cauldron of Diwrnach that we
encountered in our earlier discussion of Culhwch and Olwen, and to the
cauldron of regeneration we will meet when we consider The Spoils of
Annwn and the Second Branch of the Mabinogion.
We cannot talk about cauldrons without mentioning the Gundestrup
Cauldron, not because it is a cauldron, but because it depicts a cauldron
on one of its panels. The Gundestrup Cauldron is a richly decorated silver
vessel thought to date from the 1st century B.C., found in 1891 in a peat
bog near the hamlet of Gundestrup, in the Aars parish in Himmerland,
37
Denmark. The silverworking techniques are consistent with the Thracian

silver-sheet tradition, and it is thought the cauldron was made in Thrace,
an area comprising what is now southeastern Bulgaria, northeastern Greece
and the European part of Turkey. However, the imagery seems to be Celtic,
with some possible Thracian features.
The panel depicting a cauldron is generally known as Plate E. In the
lower half, a line of warriors bearing spears and shields, accompanied by
carnyx players (a carnyx is a bronze war-trumpet used by Iron Age Celts)
march to the left. On the left side, a large gure is immersing a man in a
cauldron, possibly to restore him to life.32 In the upper half, facing away
from the cauldron, are warriors on horseback.
The most famous panel on the cauldron is Plate A. Wikipedia says
that this shows a horned male gure in a seated position. In its right hand,
the gure is holding a torc (neck-ring), and with its left hand, it grips a
horned serpent by the
head. To the left is a
stag with antlers very
similar to the human
gure. Other animals
surround the scene,
canine, feline, bovine,
elephant, and a human
gure riding a sh or a
dolphin. The scene,
says Wikipedia, has
been compared to a
similar seal found at
Mohenjo-Daro in the
Indus Valley (in what
is now Pakistan), and
dating from 2600 B.C.
1900 B.C.). In theories
of Celtic origin, the
gure is often identied
as Cernunnos and
occasionally as Mercury. Cernunnos is Figure 3: Image of Mercury and Rosmerta found
best known from an at the Shakespeare Inn, Northgate Street, Gloucesinscription on the Pil- ter (the Roman town of Glevum).
38
lar of the Boatmen, which I referred to earlier in connection with Tarvos

Trigaranus. Here the antlered god is depicted alongside other Celtic deities
and also Roman gods such as Jupiter, Vulcan and the divine twins Castor
and Pollux. There are few representations of Cernunnos in Britain: a rare
British example at Corinium portrays Cernunnos with two snakes which
form his legs and rear up to eat corn or fruit by his head.33
Three other panels on the Gundestrup Cauldron are worth mentioning. Plate C shows the bust of a bearded gure holding on to a broken
wheel. A smaller leaping gure with a horned helmet also is holding the
rim of the wheel. Under the leaping gure is a horned serpent. The group
is surrounded by grifns and other creatures, some similar to those on
plate B. It is unclear who the central gure is, but in Gaul many representations of a bearded god with a thunderbolt in one hand and a wheel
in the other have been discovered. The Roman poet Lucan (A .D. 39
A . D . 65), in his
Pharsalia (an epic
poem on the civil
war between Julius
Caesar and Pompey), in writing of
the Gauls, refers
to those who
pacify with blood
accursed/Savage
Teutates, Hesus
horrid shrines,/
And Taranis altars
cruel as were those/
Loved by Diana,
goddess of the
north. Based on
this, the iconography from Gaul,
inscriptions to gods
called Taranucno-,
Taranuo-, and
Figure 4: Image of the genii cucullati (hooded spirits Taraino-, and the
of place) with goddess, from the Corinium Museum, existence of a
Welsh word taranu
Cirencester.
39
to thunder, it is possible that the Gaulish god with the thunderbolt and
wheel is Taranis.
The Base Plate is described by Green34 in these terms:
The central gure is a great bull (or, more likely, a wild aurochs), apparently in its death-throes. Attacking it are a small anthropomorphic being,
whose clearly demarcated breasts suggest it as female, and who wields a
knife or sword, and a hunting-dog or wolf; beneath the bull are two other
canids.... The bull or aurochs is itself in high relief and the head rears out
in a dramatically three-dimensional manner; it has holes for a pair of
detachable horns. The beast has a raised dorsal crest and a curious leaf-like
pattern engraved on its neck and shoulders; its body is surrounded by the
leaves of a climbing convolvulus-like plant.
Plate D is also generally interpreted as a bull-slaying scene. Three

bulls are placed in a horizontal line, facing the same direction. They have
massive rumps and short but thick necks. On the lower right side of each
bull, a man is standing in the posture of attacking the bull with a sword.
Under the feet of each bull, by the side of each man, a dog is depicted as
running toward the left while a cat-like creature is running in the same
direction over the back of each bull.
Other Gods and Goddesses

In Chapter 1 I referred to Mabon son of Modron as an important
character in Culhwch and Olwen. Mabon is thought to be the Welsh version
of the god Maponos (Divine Boy or Divine Son), from Gaulish mapos
young boy, son. Maponos, says Wikipedia, is mentioned in inscriptions
from Gaul (Bourbonne-les-Bains in Haute-Marne, northeastern France,
and Chamalires in central France), but is mainly attested in the north of
Britain, at Brampton, near Hadrians Wall, at Corbridge (Roman Coria),
also near Hadrians Wall in the territory of the Brigantes, at Ribchester
(Bremetenacum), a Roman fort in what is now Lancashire, and at Chesterholm (Vindolanda), a Roman fort near the border with Scotland. The
most interesting reference to Maponos is the one from Chamalires mentioned earlier, a magical text which calls on Maponos for help.
Modron is thought to be derived from Matrona, the goddess of the
River Marne in Gaul. She has been linked to the Deae Matres or Deae
Matronae, triple goddesses venerated in eastern Gaul, Germany and upper
Italy. All depictions of the Matres are frontal, says Wikipedia, and they
appear almost exclusively in threes with at least one gure holding a basket
40
Figure 5: Three mother-goddesses from the Corinium Museum, Cirencester.
of fruit in her lap, and the women are either standing or sitting. In some
depictions, the middle gure is depicted with loose hair and wearing a
headband, and the other two wear head dresses. Other motifs include
depictions of sacrice including burning incense, pigs, and bowls lled
with fruit and decorations of fruits, plants and trees. The cult seems
to have spread to Britain, and Henig35 mentions inscriptions and altars
from Dover, Winchester (Venta Belgarum), York (Eboracum), and Londinium. In addition, says Henig,36 the sculptor Sulinus set up dedications
to the Matres Suleviae (Mothers Who Govern Well) at Corinium and
Aquae Sulis. As I said earlier, these triple mother-goddesses can be linked
to the mother-goddess of the Dobunni and the genii cucullati.
It is possible that Maponos and Matrona were venerated mainly in
the north of Britain. The Welsh Triads, a group of 13th-century texts which
preserve fragments of Welsh folklore, mythology and traditional history
41
in groups of three,37 mention Modron (Triad 70) as the mother of Owain

son of Urien, both 6th-century rulers of the northern British kingdom of
Rheged.
Before I conclude this chapter, I would like to consider another divinity who will gure prominently in subsequent chapters, and that is the
goddess Epona. Epona, whose name means Great Mare (cf. Welsh ebol
colt, foal), was protector of horses, donkeys and mules. She was also,
says Wikipedia, a goddess of fertility, as shown by her attributes of a
patera (a shallow dish used for drinking), cornucopia (horn of plenty),
and ears of grain, and by the presence of foals in some sculptures. Henri
Hubert also suggested38 that the goddess and her horses were leaders of
the soul in the after-life. The cult of Epona was widespread in the Roman
Empire between the 1st and 3rd centuries A.D., and is mentioned in The
Golden Ass by Apuleius (c. A.D. 124c. A.D. 180), and in the Satires of Juvenal (c. A.D. 55c. A.D. 138). The worship of Epona, says Green,39 was
most popular in eastern Gaul and the German limes (frontier). Green says
that the most interesting and complete British object pertaining to Eponas
cult is a small bronze gurine said to be from Wiltshire [see Figure 6]
depicting the goddess seated between two ponies.... Lying in Eponas lap
and on a patera held in her right hand are huge ears of corn; on her left
hand the goddess bears a yoke40 (see Figure 6).
This horse-goddess seems to provide one of the keys to prehistoric
British religion, as we will see in Chapters 4, 5 and 6. Another key is provided by the cauldron, the triple goddesses and the genii cucullati of
Gloucestershire, and I will be seeking to throw some light on these in the
next chapter.
Chapter 3
Arthurs Voyage
The Spoils of Annwn
Taliesin and the Taliesin-Poet

The Spoils of Annwn is a poem attributed to Taliesin, and thought to
have been composed between A.D. 800 and A.D. 1000, though, as John
Koch argues,1 some of the grammatical forms indicate that it may have
been composed before A.D. 800. Most of the poems attributed to Taliesin
are found in the Book of Taliesin, which was probably rst compiled about
A.D. 1350. It is generally thought that someone called Taliesin once existed,
but it is highly unlikely he wrote all the poems attributed to him.
The existence of Taliesin is attested in Historia Brittonum (c. A.D.
830), where we read2:
Then Dutigirn at that time fought bravely against the nation of the
Angles. At that time, Talhaiarn Cataguen was famed for poetry, and
Neirin, and Taliesin and Bluchbard, and Cian, who is called Guenith
Guaut, were all famous at the same time in British poetry. The great king,
Mailcun, reigned among the Britons, i.e. in the district of Guenedota,
because his great-great-grandfather, Cunedda, with his twelve sons, had
come before from the left-hand part, i.e. from the country which is called
Manau Gustodin, one hundred and forty-six years before Mailcun reigned,
and expelled the Scots with much slaughter from those countries, and they
never returned again to inhabit them.
Taliesin is mentioned alongside one known poet, Neirin, that is, Aneirin,
a 6th-century poet believed to have written Y Gododdin, which was mentioned in Chapter 1, and three who are now virtually unknown, Talhaiern
Cataguen, better known as Talhaearn Tad Awen (Talhaearn Father of the
Muse), Bluchbard and Cian. He is said by Nennius to have lived at the
time of Dutigirn, who is otherwise unknown, and Mailcun, that is Maelgwn, a 6th-century king of the Welsh kingdom of Gwynedd descended
from Cunedda the founder of Gwynedd.
42
3 Arthurs Voyage
43
A maximum of eleven poems in the Book of Taliesin have been dated

to the 6th century, and many of these praise Urien of Rheged and his son
Owain mab Urien, who both ruled Rheged in the late 6th century. It
appears from this that he was court poet at Rheged, which included what
is now Cumbria in northwest England, and possibly parts of Lancashire
and southern Scotland, and may have been based at Luguvalium (Carlisle).
Other early poems suggest that he may also have been court poet at Powys,
which at the time was centered on the Romano-British town of Viroconium
(Wroxeter in Shropshire).
It is unlikely that the historical Taliesin wrote The Spoils of Annwn
(it is simply too different from the praise-poems he probably did write),
and after I have analyzed the poem, I will consider what it tells us about
the unknown author. I will analyze the poem in sections then as a whole,
using the translation by Sarah Higley.3 A warning: the language of this
poem is Old Welsh, and the poem often alludes to matters that we only
half understand or dont understand at all, so any translation is necessarily
an interpretation. For this reason Higleys translation is likely to differ
markedly from other translations, the best known of which is the 19thcentury one by W.F. Skene.
Analysis of the Poem

The Spoils of Annwn
PART I
I praise the Lord, Prince of the realm, King.
His sovereignty has extended across the worlds tract.
Equipped was the prison of Gweir in the Mound Fortress (Caer Sidi),
throughout the account(?) of Pwyll and Pryderi.
No one before him went into it,
into the heavy blue/gray chain; a faithful servant it held.
And before the spoils of Annwfyn bitterly he sang.
And until Judgment shall last our bardic invocation.
Three fullnesses of Prydwen we went into it.
Except seven none rose up from the Fortress of the Mound (Caer Sidi).
The rst stop in this voyage to Annwn, the Welsh Otherworld, is the
Mound Fortress, Caer Sidi in the original Welsh. To understand the meaning of Caer Sidi, and indeed many other names we meet on the way, we
need look not just at Old Welsh but also at the language family to which
44
Welsh belongs, known as Indo-European. The Indo-European language

family covers a huge area, from Portugal and Ireland in the West to Russia,
India and Pakistan in the East, and includes languages as apparently diverse
as English, German, French, Italian, Greek, Russian, Farsi (the language
of Iran), Urdu (one of the languages of Pakistan) and Hindi (one of the
languages of India). Our understanding of a word in one Indo-European
language can often be helped if we know what words it is related to in other
Indo-European languages and we have tools to help us do that, including
the Indogermanisches Etymologisches Wrterbuch (Indo-European Etymological
Dictionary), rst compiled by the German philologist Julius Pokorny, and
available on the Indo-European Etymological Dictionary Project website.
The word caer means fortress, and is thought to be from the IndoEuropean root kat- hut, shed, which gives us Avestan (Old Persian) kataa
chamber, Iranian kad house, Latin casa hut, Latin castrum fort,
Old English heathor enclosure, prison, Welsh cader fort, Old Irish
cathir town.4 Although this etymology by itself tells us very little, it does
give us an early glimpse of how words get around, in this case apparently
leaping from Avestan, the language of the Zoroastrian scriptures composed
around 1000 B.C., to Latin, German and Celtic.
The word sidi is cognate with the Irish sid or sidh, which according
to Koch5 refers to hills or mounds (often in reality containing prehistoric
burials) conceived of as hollow and the residence of supernatural beings.
In his Dictionary of Continental Celtic Place-Names, Falileyev gives the following information on sid 6: Old Irish sid tumuluus; peace, saidid sits;
Cornish asedhva assembly; Old Welsh in guorsed (place) assembly, Middle
Welsh sed, Modern Welsh sedd seat (from Indo-European root sed- sit,
Latin sedes seat, Sanskrit sadas seat, residence, abode, dwelling, place of
meeting, assembly (especially at a sacrice), Old English set). Sanskrit
is the language of the Indian scriptures, which go back to around 1700
B.C., and of the Rigveda, composed between about 1700 B.C. and 1100 B.C.,
and among the oldest Indo-European religious texts. In the Rigveda the
sadas was a ritual space belonging to the gods where the soma sacrice took
place7 (soma was a stimulant drink poured into re as an offering to the
gods). While I am not suggesting that the British sidi was a sacred space
where sacrices took place, it is clear that the sidi has strong connections
with the Otherworld.
Imprisoned in this Otherworld Mound Fortress is someone called
Gweir. We do not know much about Gweir or Gwair, but he is mentioned
in Triad 52 as one of the Three Exalted Prisoners, along with Mabon son
3 Arthurs Voyage
45
of Modron and Llyr, who

plays a large if passive
role in the Second Branch
of the Mabinogion (see
Chapter 4). Gweir and
Mabon son of Modron
are alike in one way:
Gweir is said to sing bitterly in his prison, just as
Mabon, in Culhwch and
Olwen, made a great
wailing and lamenting
from the dungeon of his
prison in Gloucester. Presumably the spoils of
Annwn before which he
sings bitterly refers to the
cauldron introduced in
Part II.
It seems that Gweir Figure 6: Statuette of the horse-goddess Epona,
is in Caer Sidi through said to be from Wiltshire. British Museum.
the account (ebostol ) of
Pwyll and Pryderi. Pwyll is the king of Dyfed (west Wales) who, in the
First Branch of the Mabinogion, takes over as lord of Annwn for a year and
a day, and Pryderi is his son. It is unclear what the poet means here, since
ebostol (presumably a Christian word) means apostle or epistle, and
neither meaning makes much sense: was Gweir in prison because of messengers (apostles) sent by Pwyll and Pryderi, or was he in prison because
of formal letters (epistles) sent by Pwyll and Pryderi?
The voyage is undertaken in Prydwen (Fair-Face), and only seven
returned a theme taken up in the Second Branch of the Mabinogion,
where only seven return from an expedition to Ireland, including Taliesin,
who by then had become a gure of mythology. Interestingly, however,
the seven returnees in the Second Branch did not include Arthur.
Here is the next section of the poem:
PART II
I am honored in praise. Song was heard
in the Four-Peaked Fortress (Caer Pedryuan), four its revolutions.
46

My poetry, from the cauldron it was uttered.
From the breath of nine maidens it was kindled.
The cauldron of the chief of Annwfyn: what is its fashion?
A dark ridge around its border and pearls.
It does not boil the food of a coward; it has not been destined.
The ashing sword of Lleawch has been lifted to it.
And in the hand of Lleminawc it was left.
And before the door of hell lamps burned.
And when we went with Arthur, brilliant difculty,
except seven none rose up from the Fortress of Mead-Drunkenness
(Caer Vedwit).
The next stop in this otherworldly voyage is Caer Pedryuan, with its
four revolutions; Higley translates it as the Four-Peaked Fortress, but it
can also be translated as Four-Cornered Fortress. Since we know so little
of British mythology what we can glean from Roman Britain is necessarily seen through Roman eyes, and later medieval sources like the
Mabinogion were inuenced by centuries of Christianity I am going far
back into Indo-European prehistory to explain Caer Pedryuan.
In that ancient Indian religious text, the Rigveda, one of the most
important gods is Varuna, king of the gods, the god of the sky, of
waters and the celestial ocean, as well as god of law and of the underworld.
Our knowledge of ancient Indo-European religion, however, does not
come only from the Rigveda, for we also have the Avesta, the sacred texts
of Zoroastrianism, the religion of preIslamic Persia, which were thought
to have been composed between about 1000 B.C. and 500 B.C. According
to Mary Boyce and Frantz Grenet,8 Varuna lost his identity in Zoroastrianism, being invoked as Apam Napat (Grandson of Waters), Ahura
Berezant (High Lord) or the Baga (God). Interestingly, however, there
is a place called Varena in Zoroastrianism: in the sacred book called
the Vendidad,9 the supreme deity Ahura Mazda is said to have created
sixteen good lands, the fourteenth of which is four-cornered Varena,
for which was born Thraetaona, who smote Azi Dahaka [Zohak]. Thraetaona, also known as Fereydun or Apam Napat, is the name of a mythical
Persian king who is said to have killed the dragon Azi Dahaka or Zohak
in Varena, which was said10 to be inhabited by Varenya daevas (ends,
demons).
The epithet four-cornered is echoed in ancient Indian descriptions
of Varuna. Pettazzoni notes11 that Varuna is given the epithet caturanika
four-faced, referring to the four cardinal points or four directions in
which he looks, not as monarch but as sky-god; and the Rigveda speaks12
3 Arthurs Voyage
47
of the fearful four-edged bolt of Mitra-Varuna, using the Sanskrit word

caturasrir four-cornered.
The name Varuna may be derived from the Indo-European root werto close, cover, guard, save,13 which gives Old Indian varati, surrounds,
protects, varutar protector, defender, varutha protection, defense,
Vritra the enveloper, the Vedic demon associated with drought, Avestan
varetha defensive weapon, Old Irish fertae burial mound, Middle Welsh
gwerthyr fortress, German Wehr defense, protection. So it is possible
that the Four-Peaked or Four-Cornered Fortress, with its four revolutions, is not only a very ancient concept, but also one which has some
connection with the heavens, in contrast to Caer Sidi, which seems much
more earth-bound though curiously four-faced Varuna and four-cornered Varena seem to have become a fortress in Britain and a burial mound
in Ireland.
It soon emerges that Caer Pedryuan contains a cauldron. We have
already encountered a cauldron in Culhwch and Olwen, and on one of the
panels of the Gundestrup Cauldron, and we will also come across another
cauldron in the Second Branch of the Mabinogion, a cauldron belonging
to the main character Bran, which the Irish use to regenerate their dead
warriors. It seems that this cauldron is the source of Taliesins poetry, that
it belongs to the (unnamed) chief of Annwn, that it has a dark ridge and
pearls around its border, it does not boil the food of cowards, and it is
kindled by the breath of nine maidens.
Let us start with the nine maidens. If the number nine sounds familiar, it is because in Pa gur (see Chapter 1) we are told by Arthur that Cai
slew nine witches. Surprisingly we know quite a lot about these nine
witches. Pomponius Mela,14 describes an island called Sena (now called
Ile de Sein, off the coast of Britanny). This island, he says, was famed for
its oracle of a Celtic god, to whose cult nine maiden-priestesses were
devoted. The priestesses were known as Senae, and were said to be gifted
with remarkable intelligence. It was believed that they can raise up the
waves of the sea and the winds with their songs, that they can assume the
shape of any animal they choose, that they can cure complaints that to
others are incurable, and that they know and predict the future.
These nine priestesses reappear in the Life of Saint Samson, a 6thcentury saint who was born in Wales but later moved to Brittany, where
he became one of the founding saints of the Breton church. In the Life,
written between 610 and 820, Samson encounters a sorceress in a wood,
pursues her and nally catches up with her. He then asks her15:
48

Who art thou, misshapen one, and of what kind art thou? And she,
greatly trembling, said, I am a sorceress, for indeed the women of my race
and sex, as morally perverse, have to this day been transgressors in your
sight, and now no one of my race is left in this wood save me only. For I
have eight sisters and my mother is still living, and these are not here, but
dwell in a wood more remote, and I was made over to this wilderness by
my husband, and my husband is dead, and on this account I am unable to
withdraw from this wood.
These nine women then reappear in Geoffrey of Monmouths Life of

Merlin, written after the History of the Kings of Britain. Here Geoffrey is
describing Avalon16:
The island of apples which men call The Fortunate Isle gets its name
from the fact that it produces all things of itself; the elds there have no
need of the ploughs of the farmers and all cultivation is lacking except
what nature provides. Of its own accord it produces grain and grapes, and
apple trees grow in its woods from the close-clipped grass. The ground of
its own accord produces everything instead of merely grass, and people live
there a hundred years or more. There nine sisters rule by a pleasing set of
laws those who come to them from our country. She who is rst of them is
more skilled in the healing art, and excels her sisters in the beauty of her
person. Morgen is her name, and she has learned what useful properties all
the herbs contain, so that she can cure sick bodies. She also knows an art
by which to change her shape, and to cleave the air on new wings like
Daedalus.... Thither after the battle of Camlan we took the wounded
Arthur.
The nine women make a nal appearance in Peredur Son of Efrawg,

a 13th-century Welsh version of Perceval, the Story of the Grail, written by
the 12th-century French poet Chrtien de Troyes. In this tale, which has
elements not found in the French original, we read the following17:
Then Peredur rode forward. And above him he beheld a castle, and thitherward he went. And he struck upon the gate with his lance, and then,
behold, a comely auburn-haired youth opened the gate, and he had the
stature of a warrior, and the years of a boy. And when Peredur came into
the hall, there was a tall and stately lady sitting in a chair, and many handmaidens around her; and the lady rejoiced at his coming. And when it was
time, they went to meat. And after their repast was nished, It were well
for thee, chieftain, said she, to go elsewhere to sleep. Wherefore can I
not sleep here? said Peredur. Nine sorceresses are here, my soul, of the
sorceresses of Gloucester, and their father and their mother are with them;
and unless we can make our escape before daybreak, we shall be slain; and
already they have conquered and laid waste all the country, except this one
dwelling. Behold, said Peredur, I will remain here to-night, and if you
3 Arthurs Voyage
49
are in trouble, I will do you what service I can; but harm shall you not
receive from me. So they went to rest. And with the break of day, Peredur
heard a dreadful outcry. And he hastily arose, and went forth in his vest
and his doublet, with his sword about his neck, and he saw a sorceress
overtake one of the watch, who cried out violently. Peredur attacked the
sorceress, and struck her upon the head with his sword, so that he attened
her helmet and her head-piece like a dish upon her head. Thy mercy,
goodly Peredur, son of Evrawc, and the mercy of Heaven. How knowest
thou, hag, that I am Peredur? By destiny, and the foreknowledge that I
should suffer harm from thee. And thou shalt take a horse and armour of
me; and with me thou shalt go to learn chivalry and the use of thy arms.
Said Peredur, Thou shalt have mercy, if thou pledge thy faith thou wilt
never more injure the dominions of the Countess. And Peredur took
surety of this, and with permission of the Countess, he set forth with the
sorceress to the palace of the sorceresses. And there he remained for three
weeks, and then he made choice of a horse and arms, and went his way.
The most intriguing of these references is in Peredur, where the nine

sorceresses are said to come from Gloucester. It is tempting to link the nine
maidens of Annwn to the mother-goddess of the Dobunni, with her bucket,
tub or cauldron, and to the cucullati, the hooded spirits, who almost always
appear in groups of three, and the Matres or Mothers, who likewise always
appear in groups of three and the nine sorceresses of Gloucester seem to
conrm this. And Pomponius Melas allusion to the nine priestesses of Sena
seems to indicate that the nine maidens represent an old tradition in British
religion (as their name suggests, the Bretons originally came from Britain).
This long tradition also shows that the priestesses of paganism became the
witches of Christianity, except in the case of Geoffreys Life of Merlin; perhaps
Geoffrey was familiar with Pomponius Melas description of Sena.
The other interesting feature of the cauldron of Annwn is that it does
not boil the food of cowards, a fact which has led the cauldron to be compared with the Holy Grail. The Holy Grail was rst mentioned in Chrtiens Perceval. In this early Arthurian tale, Perceval is a young man who,
inspired by his encounters with knights, travels to King Arthurs court,
where a young girl predicts greatness for him. He is taunted by Sir Kay,
but receives the vermilion armor of a knight outside Arthurs castle. He
then sets out for adventure, rescues and falls in love with the young princess
Blancheeur, and trains under the older knight Gornemant.
One day at a river, he sees two men in a boat shing. Perceval is
unable to cross, and one of the shermen offers him lodging for the night.
Following the directions of the sherman, Perceval climbs to the top of a
hill, where he sees a square tower and a hall. He enters the hall and is
50
greeted by a handsome nobleman with gray hair, who apologizes for being
unable to rise. A squire enters carrying a sword, and announces that the
lords niece has sent it to him the lord gives the sword to Perceval.
Another squire enters carrying a white lance, from which blood drips.
Perceval refrains from asking about this lance, recalling Gornemants earlier
advice not to be too talkative. More squires bring in candelabras. Then,
in the words of Chrtien18:
A damsel came in with these squires, holding between her two hands a
grail (graal ). She was beautiful, gracious, splendidly garbed, and as she
entered with the grail in her hands, there was such a brilliant light that the
candles lost their brightness, just as the stars do when the moon or sun
rises. After her came a damsel holding a carving-dish of silver. The grail
which preceded her was of rened gold; and it was set with precious stones
of many kinds, the richest and costliest that exist in the sea or in the earth.
Without question those set in the grail surpassed all other jewels.
Perceval fails to ask who is being served by the grail. They dine at an ivory
table. The grail returns, but again Perceval does not ask about it. Later that
night, the nobleman excuses himself and has to be carried off to his bedroom.
The next morning, Perceval discovers that the hall is deserted and everyone
has left. As he rides over the drawbridge, it mysteriously rises up on its own.
He then encounters a maiden weeping beneath an oak tree. She holds a dead
knight, whose head has been cut off by another knight that morning. She
asks where he spent the night, he describes the tower and hall, and she tells
him that he stayed with the Fisher King (the name of the mysterious nobleman). She says the Fisher King was wounded in a battle by a javelin through
both thighs and is still in much pain, and that shing is the only recreation
he has. She rebukes him for not asking any questions, saying that if he had
asked only one question, he would have cured the king.
The Holy Grail was the dish, plate or cup used by Jesus at the Last
Supper and said to possess miraculous powers. The word grail (in French
graal) is derived from the Latin gradalis, meaning a dish brought to the
table in different stages of a meal. For Chrtien the grail was a wide, somewhat deep dish or bowl, and Perceval learns many years later that the grail
of the Fisher King contains a single consecrated host (wafer) that the King
has lived on for twelve years. Like the cauldron of Annwn, the grail is
jewel-encrusted, and just as the cauldron does not boil the food of cowards,
the grail can only be won by those worthy of it (Perceval fails to ask any
questions, and is therefore unable to cure the Fisher King).
Back in The Spoils of Annwn we next learn that the ashing sword
3 Arthurs Voyage
51
of Lleawch has been lifted to [the cauldron], and that it was left in the
hand of Lleminawc. As Higley points out, we can compare this to the
episode in Culhwch and Olwen in which Arthur and his men have to capture a cauldron belonging to the Irish giant Diwrnach Wyddel. They ask
the giant for the cauldron, but he refuses, whereupon: Bedwyr arose and
seized hold of the cauldron, and placed it upon the back of Hygwyd,
Arthurs servant, who was brother, by the mothers side, to Arthurs servant,
Cachamwri. His ofce was always to carry Arthurs cauldron, and to place
re under it. And Llenlleawg Wyddel seized Caledvwlch, and brandished
it. And they slew Diwrnach Wyddel and his company. It appears from
this that Lleawch may be identied with Llenlleawg Wyddel (Llenlleawg
the Irishman), one of the people Culhwch called on for help when he rst
arrived at Arthurs court, and who eventually helped Culhwch to get the
giants cauldron. As for Lleminawc, Higley links this name to Lluch
Llawwynnawc, another knight of Arthurs court called on by Culhwch,
which can be translated as Lluch Windy-Hand or Lluch StrikingHand), and can be seen as another Welsh version of Lugus (compare this
with the Irish epithet for Lugh, Fierce-Striker). As Higley notes, there
does seem to be a good deal of scribal confusion here.
Before this section ends, we are informed for the rst time that Arthur
is involved in this voyage to Annwn, which is now referred to as the Fortress
of Mead-Drunkenness, Caer Vedwit. The role of mead in early British
societies is underlined in the poem Y Gododdin, which mentions Arthur
in passing (see Chapter 1). Mead is mentioned forty-three times; here are
some examples.19 Note that Catraeth is the scene of the battle where so
many warriors were killed:
Men went to Catraeth, keen their war-band.
Pale mead their portion, it was poison.
Men went to Catraeth at dawn:
Their high spirits lessened their life-spans.
They drank mead, gold and sweet, ensnaring;
For a year the minstrels were merry.
Men launched the assault, nourished as one
A year over mead, grand their design.
How sad their tale, insatiable longing,
Bitter their home, no child to cherish it.
Because of wine-feast and mead-feast they charged,
Men famed in ghting, heedless of life.
Bright ranks around cups, they joined to feast.
Wine and mead and bragget, these were theirs.
52
The attitude towards mead, gold and sweet, ensnaring and ultimately poison, is highly ambivalent. However, the fact that one of the
locations in the Otherworld is associated with mead suggests that more is
involved. In Irish mythology there is a queen called Medb, whose name
is related to mead and means intoxicating. She was queen of Connacht
in the Ulster Cycle of Irish mythology, whose husband was Aillil mac
Mata, although she had several husbands before him who were also kings
of Connacht. She was probably originally a sovereignty goddess, whom a
king would ritually marry as part of his inauguration. We will return to
the theme of the sovereignty goddess in Chapter 4 when we look at the
First Branch of the Mabinogion.
One last point to make is that before the Fortress of Mead-Drunkenness is mentioned, the poet refers the doors of hell, in Welsh porth
uffern, apparently based on two Latin words, porta gate and infernus
hell.
PART III
I am honored in praise; song is heard
in the Fortress of Four-Peaks, isle of the strong door.
Flowing water and jet are mingled.
Sparkling wine their liquor before their retinue.
Three fullnesses of Prydwen we went on the sea.
Except seven none rose up from the Fortress of Hardness (Caer Rigor).
We now return in Part III to the Fortress of the Four-Peaks, which

is apparently an island. In another poem attributed to Taliesin, Song Before
the Sons of Llyr,20 the Otherworld is also seen in terms of an island:
Complete is my chair in Caer Siddi,
No one will be aficted with disease or old age that may be in it....
And around its borders are the streams of the ocean.
And the fruitful fountain is above it,
The liquor is sweeter than white wine.
In the Second Branch of the Mabinogion, after the Welsh have

defeated the Irish, the seven survivors, together with Brans head, spend
eighty years on an otherworldly island, Gwales. Before they go, Bran warns
them21: You will be at Gwales in Penfro eighty years. Until you open the
door facing Aber Henvelen on the side facing Cornwall, you will be able
to abide there, along with the head with you uncorrupted. But when you
open that door, you will not be able to remain there. In the end, of course,
one of the seven opens the door and when he looked, suddenly everything
3 Arthurs Voyage
53
they had ever lost loved ones and companions, and all the bad things
that had ever happened to them; and most of all the loss of their king
became as clear as if it had been rushing in towards them. The strong
door, in Welsh pybyrdor, this time with the normal Welsh word dor, seems
to be the door between the Otherworld and the world of the living
strong because in this case it prevents the dead from regretting all that
they have left behind in the world of the living.
In line 25 the poet tells us that Flowing water and jet are mingled,
presumably in or around the isle of the strong door. Jet is a form of lignite
used in jewelry. Jet beads have often been found in Bronze Age barrows,
says the Discover Yorkshire Coast website, and it is thought that the stone
might have been worn to ward off evil spirits. The Romans also had great
liking for jet and jet objects are often found on Roman sites, these including
hair pins, bracelets, medallions and nger rings; jet rings were found in
excavations at Huntcliff and Normanby around 1920 (both these sites are
in Yorkshire). As Higley points out, the mingling of water and jet may
refer to a passage from Isidore of Seville (c. 560636), the Archbishop of
Seville said to be the last scholar of the ancient world. In his Etymologiae
(Etymologies) he writes22: [Jet] is black, at, smooth, and burns when
brought near to re. Dishes cut out of it are not destructible. If burned it
puts serpents to ight, betrays those who are possessed by demons, and
reveals virginity. It is wonderful that it is set on re by water and extinguished with oil.
Part III ends with a reference to the fourth fortress to be mentioned,
the Fortress of Hardness (Caer Rigor). This is sometimes translated as the
Fortress of Kings or Kingly Castle, because in Old Irish the word ri (genitive
rig) means king (compare Gaulish rix, latin rex and Old Indian raj-). The
word comes from the Indo-European root reg-, and Pokorny23 lists as
derived from this root not only rigor and rex, but also Old Irish reg-, rigto stretch out (e.g. the hand), at-reig uplifts oneself, eirge raising up;
Middle Welsh dy-re stands up, rhein stiff, elongated; Old Welsh arcib-renou buried, and Middle Welsh ar-g yu-rein to bury. So the word
may be borrowed from Latin, or may be an old Celtic word. As to what
Caer Rigor means, we will consider that when we try to interpret the poem.
PART IV
I merit not the Lords little men of letters.
Beyond the Glass Fortress (Caer Wydyr) they did not see
the valor of Arthur.
54

Six thousand men stood upon the wall.
It was difcult to speak with their sentinel.
Three fullnesses of Prydwen went with Arthur.
Except seven none rose up from the Fortress of Guts (Hindrance?)
(Caer Golud ).
The Glass Fortress, or Caer Wydyr, has sometimes been compared to

Glastonbury in Somerset. The reason for this can be found in two works
by the 12th-century Norman-Welsh priest and writer Giraldus Cambrensis,
De principis instructione (On the instruction of a prince) and Speculum Ecclesiae (Mirror of the Church). Here is the relevant passage from De principis
instructione 24:
What is now known as Glastonbury used, in ancient times, to he called the
Isle of Avalon. It is virtually an island, for it is completely surrounded by
marshlands. In Welsh it is called Ynys Avallon, which means the Island
of Apples and this fruit used to grow there in great abundance. After the
Battle of Camlann, a noblewoman called Morgan, who was the ruler and
patroness of these parts as well as being a close blood-relation of King
Arthur, carried him off to the island, now known as Glastonbury, so that
his wounds could be cared for. Years ago the district had also been called
Ynys Gutrin in Welsh, that is the Island of Glass.
Here is the relevant passage from Speculum Ecclesiae25:

After the Battle of Camlann ... the body of Arthur, who had been mortally
wounded, was carried off by a certain noble matron, called Morgan, who
was his cousin, to the Isle of Avalon, which is now known as Glastonbury.... It is called Avalon, either from the Welsh word aval, which
means apple, because appletrees and apples are very common there, or
from the name of a certain Vallo who used to rule over the area long ago.
In remote times, the place used to be called Ynys Gutrin in the Welsh
language, that is the Island of Glass, no doubt from the glassy colour of
the river which ows round it in the marshland.
This is obviously inspired by the tale in Geoffrey of Monmouths Life of

Merlin, except that Avalon is now equated with Glastonbury, or rather
Glastonbury Tor, which in the past was surrounded by marshes, and resembled an island.
Something similar to the Glass Fortress is also mentioned in the Historia Brittonum.26 Nennius writes that as the Milesians (the nal inhabitants
of Ireland, according to Irish mythology) were voyaging towards Ireland,
there appeared to them, in the middle of the sea, a tower of glass, the
summit of which seemed covered with men, to whom they often spoke,
but received no answer. This is reminiscent of the scene in the Second
3 Arthurs Voyage
55
Branch of the Mabinogion, where the dead Irish warriors are brought back
to life by Brans cauldron but cannot speak.
At the end of Part IV we move on to another fortress or is it the
same fortress with different names? the Fortress of Guts or Hindrance.
Higley notes that either interpretation is possible, and points out that other
translators have seen it differently, with Loomis opting for Fortress in the
Middle of the Earth, and Koch choosing Concealed Fort. In modern
Welsh golud means wealth, riches; the Anglo-Saxon monk and historian,
Bede, refers to an urbs Coludi, a 6th-century fort near St. Abbs, Berwickshire, where St. Aebbe established a monastery.
PART V
I do not merit little men, slack their shield straps.
They do not know which day who was created (or: created whom?);
what hour of midday (?) Cwy was born.
Who made him who did not go (to the) meadows of Defwy?
They do not know the brindled ox, thick his headband.
Seven score links on his collar.
And when we went with Arthur, dolorous visit,
except seven none rose up from the fortress of Gods Peak
(Caer Vandwy).
At this point the poem becomes especially challenging, both difcult

to translate and with allusions to characters and places whose meaning is
lost (e.g. Cwy and the meadows of Defwy). Two elements do stand out:
the brindled ox and the fortress of Gods Peak, or Caer Vandwy. The
brindled ox may be related to the two dun oxen of Gwlwlyd that Culhwch must get as one of the tasks he has to complete in order to win the
hand of Olwen; or to the yellow and the brindled bull that Culhwch
must obtain; or to the two horned oxen, one of which is beyond, and the
other this side of the peaked mountain, yoked together in the same plough
that he is required to get. The brindled ox may also be related to Tarvos
Trigaranus, whose statue was found at the Roman temple at Maiden Castle
hillfort (see Chapter 2), and to the bulls depicted on two of the panels of
the Gundestrup Cauldron.
The bulls depicted on the Gundestrup Cauldron were either about
to be sacriced or had just been sacriced. Pliny tells us that the druids
sacriced bulls, and such a ceremony is described in The Sick-Bed (Wasting
Sickness) of Cuchulain, a 10th/11th-century Irish text that is part of the
Ulster Cycle of Irish mythology27: A white bull was killed, and one man
would eat his ll of the meat and of the broth, and in his sleep after that
56
meal, a charm of truth would be said over him by four Druids. And whoever he would see in his sleep would be king. The brindled ox is said to
have seven-score links on his collar. In a poem attributed to Taliesin, Angar
Kyfyndawt (Hostile Confederacy), the poet says: Seven score Ogyrven/Are
in the Awen.28 The Welsh word awen means inspiration, and derives
from the Indo-European -uel, meaning to blow, and has the same root
as the Welsh word awel meaning breeze. The meaning of og yrven is
unclear: in another poem attributed to Taliesin, The Chair of the Sovereign, the poet says of awen and og yrven29:
High (is) truth when it shines,
Higher when it speaks.
High when came from the cauldron
The three awens of Gogyrwen (= Ogyrven).
All that is clear is that the little men of letters do not know the brindled
ox, suggesting that the brindled ox has something to do with poetic inspiration.
The next fortress encountered in this otherworldly voyage is the
fortress of Gods Peak, or Caer Vandwy. This is not the only reference to
Caer Vandwy in early Welsh literature: in The Dialogue of Gwyddno Garanhir and Gwynn ap Nudd from The Black Book of Carmarthen, the poet says
(speaking here in the voice of Gwynn)30:
To my sadness
I saw a conict before Caer Vandwy.
Before Caer Vandwy a host I saw,
Shields were shattered and ribs broken
Renowned and splendid was he who made the assault.
Gwynn ap Nudd (Gwynn son of Nodens) rst appears in Culhwch and

Olwen, where he abducted a maiden called Creiddylad, daughter of Llud
Llaw Eraint (Nodens of the Silver Hand) after she eloped with Gwythr
ap Greidawl, Gwyns longtime rival. Gwyn and Gwythr are condemned
to ght, every May Day until Doomsday, a ght which represents the contest between summer and winter. If Gwyn is associated with Caer Vandwy,
then Caer Vandwy may also be associated with the contest between summer
and winter, between life and death.
PART VI
I do not merit little men, slack their will.
They do not know which day the chief was created,
what hour of the midday the owner was born,
3 Arthurs Voyage
57
what animal they keep, silver its head.

When we went with Arthur, sorrowful strife,
except seven none rose up from the Fortress of Enclosedness
(Caer Ochren).
Part VI seems to consist mainly of complaints about the little men, particularly about their lack of knowledge. Line 43 parallels line 35, line 44
is similar to line 36, line 45 is almost identical to line 37, so line 46 may
perhaps be linked to lines 3940, suggesting that the animal with the silver
head is in fact the brindled ox with the thick headband. This section ends
with yet another fortress, Caer Ochren, which Higley translates as Fortress
of Enclosedness. It has been linked to Welsh ochr slope (mountain side),
hill. It also seems that in Roman times Lizard Point in Cornwall was
called Ocrinum Promontorium, and Greek okrin is a jagged point or
prominence (see also Latin ocris a broken, rugged, stony mountain).
PART VII
Monks howl like a choir of dogs
from an encounter with lords who know:
is there one course of wind? is there one course of water?
is there one spark of re of erce tumult?
PART VIII
Monks pack together like young wolves
from an encounter with lords who know.
They do not know when midnight and dawn divide.
Nor wind, what its course, what its onrush,
what place it ravages, what region it strikes.
The grave of the saint is hidden (or: lost, vanishing, in the
Otherworld), both grave and ground (or: champion).
I praise the Lord, great prince,
that I be not sad; Christ endows me.
It now appears that the little men of Parts IVVI are in fact monks, who
are ignorant of some very basic facts of Celtic lore. In line 55 they are said
to not know when midnight and dawn divide, which sounds like a reference to the way the Gauls computed time, as explained by Julius Caesar
(see Chapter 1). Line 58, says Higley, is very difcult to translate, and
could mean:
The grave of the saint is vanishing, both grave and ground.
The grave of the saint is hidden, both grave and champion.
How many saints in the Otherworld, and how many on earth?
58

How many saints lost, and how many altars? or:
How many saints in the void, and how many on earth?
Whatever the line means, says Higley, diuant is a gloomy concept, and
the sense expressed here is of sadness and loss, which is conrmed by the
last line of the poem (that I be not sad). It seems that the poet is lamenting the passing of those who know the old religion (the knowledge of the
Druid, the world view of Celtic religion), and the loss not only of their
material remains, but also of their spiritual legacy. It recalls Chinua
Achebes 1958 novel Things Fall Apart, which is set during the coming of
the white man to Nigeria in the 19th century. Although on the face of it
the setting of this novel could not be more different from the setting of
The Spoils of Annwn, at a deeper level the protagonists of the two are living
through a similar experience. In Things Fall Apart Nwoye, the son of the
main character, Okonkwo, converts to Christianity, and Okonkwo, who
still follows the traditional religion, has a sudden terrible thought31:
Suppose when he died all his male children decided to follow Nwoyes
steps and abandon their ancestors? Okonkwo felt a cold shudder run
through him at the terrible prospects, like the prospect of annihilation.
He saw himself and his fathers crowding round their ancestral shrine waiting in vain for worship and sacrice and nding nothing but ashes of
bygone days, and his children the while praying to the white mans god.
Interpretation of the Poem

The Spoils of Annwn is often interpreted as a military expedition, and
there are certainly links between the poem and Brans expedition to Ireland
in the Second Branch of the Mabinogion (seven survivors, cauldron, strong
door, uncommunicative sentinel). However, a good work of art can be
interpreted on more than one level, and I believe that The Spoils of Annwn
is more about a symbolic voyage than a real one. This idea is certainly not
a new one. The Scottish journalist, folklorist and scholar of the occult,
Lewis Spence, in The Mysteries of Britain, rst published in 1905, says32:
The poem is on the same lines as The Harrying of Hell, the descent into
the gulf, to cow its evil denizens and carry away its secret and treasures.
It is, indeed, part of the ritual of the candidate for adeptship into the
British mysteries.
Let us assume for the sake of argument that The Spoils of Annwn is a
symbolic voyage, an initiation real or metaphorical into Druid knowl-
3 Arthurs Voyage
59
edge or the mysteries of British religion. The rst stop on this symbolic
voyage is Caer Sidi, the Mound-Fortress. This was interpreted as a burial
mound, but as Song Before the Sons of Llyr suggests, Caer Sidi is more than
this, since around its borders are the streams of the ocean./And the fruitful
fountain is above it, suggesting its location is in the sea or even in the
heavens. The Welsh word sidydd means zodiac, so it is not surprising
that Caer Sidi is also seen as a heavenly abode. The Welsh word is similar
to the Latin sidus stars, constellation, which Pokorny33 derives from the
Indo-European root sueid- to shine, which also gives Avestan xwaena
glowing, Lithuanian svidus shining, bright, svidu gleam. For this
reason, it is not clear whether the Welsh word was borrowed from Latin
or developed independently, and whether the Taliesin-poet was thinking
of the zodiac when he referred to Caer Sidi in The Spoils of Annwn.
However, this link is arguably being made by the author of Song
Before the Sons of Llyr, and it may be reinforced by the two lines that
precede the reference to Caer Sidi:
In the festivals of the Distributor, who bestowed gifts upon me.
The chief astrologers received wonderful gifts.
A few lines before, the poet said:

I liberated my lord in the presence of the distributor.
Elphin, the sovereign of greatly aspiring ones.
We saw earlier that Varuna, the sky-god in the Rigveda, was sometimes
called in Old Persian texts Baga, which signies God but literally means
The Distributor.34 If the Distributor of the Taliesin-poet is the sky-god,
then Caer Sidi is surely a celestial dwelling-place.
If Caer Sidi is both an earthly and celestial abode, what does this tell
us about the symbolic voyage described in The Spoils of Annwn? We saw
in Chapter 1 that Hippolytus of Rome linked the Druids to Pythagoras,
so perhaps this question can be answered by looking at the philosophy of
Pythagoras and the Pythagoreans. Pythagoras himself wrote nothing, so
everything we know about his philosophy comes from Pythagoreans like
Philolaus, and later philosophers like Aristotle.
According to the Stanford Encyclopedia of Philosophy, Philolaus (c.
470 B.C.c. 385 B.C.) believed that nature was tted together out of two
opposites which he calls limiters and unlimiteds. The rst thing tted
together, he says, is called the hearth, the central re around which all
heavenly bodies, including the earth, orbit. What he means by tted
together is explained in Fragment 735:
60

... since these beginnings [i.e. limiters and unlimiteds] preexisted and were
neither alike nor even related, it would not have been possible for them to
be ordered, if a harmony had not come upon them.... Like things and
related things did not in addition require any harmony, but things that are
unlike and not even related ... it is necessary that such things be bonded
together by a harmony, if they are going to be held in an order.
Aristotle, in his Metaphysics, after presenting his account of the philosophy of the so-called Pythagoreans, which has strong connections to
the philosophy of Philolaus, turns to others of this same group and
assigns to them what is commonly known as the table of opposites. These
Pythagoreans presented the principles of reality as consisting of ten pairs
of opposites36:
limit
odd
unity
right
male
rest
straight
light
good
square
unlimited
even
plurality
left
female
motion
crooked
darkness
bad
oblong
Pythagoreanism is often linked to Orphism, the mystery religion associated with Orpheus (see Chapter 2 for a discussion of two Orpheus
mosaics in Britain). As Orphism was a mystery religion, very little is known
about its doctrine, but some texts do survive. One of these is the Petelia
tablet, discovered in lower Italy near Sybaris. The poem was written on
thin gold leaf, rolled up and placed in a cylinder hanging from a gold
chain. It was presumably hung around the neck of a dead person as an
amulet. The tablet reads37:
Thou shalt nd out to the left of the House of Hades a Wellspring
And by the side thereof standing a white cypress.
To this Wellspring approach not near.
But thou shalt nd another by the lake of Memory,
Cold water owing forth, and there are guardians before it.
Say I am a child of the Earth and Starry Heaven:
But my race is of heaven alone. This ye know yourselves.
And lo, I am parched with thirst and I perish. Give me quickly
the cold water owing forth from the lake of memory.
And of themselves they will give thee to drink from the holy Well spring.
And thereafter among the other Heroes thou shalt have lordship.
3 Arthurs Voyage
61
It seems likely, therefore, that if Caer Sidi is both of the earth and of the
heavens, it embodies one or more of the opposites proposed by the
Pythagoreans, and represents the possibility of the kind of rebirth symbolic or real hinted at in the Petelia tablet.
Some clue as to the nature of Caer Sidi is given by the plight of Gweir,
the man held prisoner there. Koch suggests38 that Gweir may be a variant
of Gwri, the name given to the infant Pryderi in the First Branch of the
Mabinogion. The Canadian psychoanalyst Dan Merkur, following a suggestion made by W.J. Gruffydd,39 believes that Gweirs imprisonment is
equivalent of that of Pryderi in the Third Branch of the Mabinogion, in
which Pryderi enters an enchanted caer, touches a golden bowl and becomes
stuck to it (see Chapter 4). Merkur interprets the imprisonment thus40:
The statement in The Spoils of the Otherworld that Gweir sang sadly in
front of the spoils or booty of the otherworld suggests that like Pryderi,
Gweir was imprisoned in the presence indeed, imprisoned by the very
object that he sought to carry off.... The meaning of the motif is implicit.
The cauldron or [bowl] imprisoned in that a person could not let go of it
once it had been touched.... The cauldron or [bowl] implicitly contained a
potable and/or edible substance that was addictive.
Merkurs suggestion that the imprisonment of Gweir is a psychological

one, arising from an addiction to a psychedelic substance, is an intriguing
one. Certainly mead is mentioned later in the poem, and we know that
one possible interpretation of the name Belenus (a Celtic god especially
popular in Aquileia, on the Adriatic coast) is the Henbane (God),41
implying that this hallucinogenic substance was used among the Celts.
The use of mead or henbane suggests that Caer Sidi could well be a place
where hell and heaven could be experienced.
The second stage of the symbolic voyage is Caer Pedryuan, the FourPeaked or Four-Cornered Fortress. The Four-Cornered Fortress, we saw
earlier, represents the heavens and as such may be the other face of Caer
Sidi. What it means in the poem, however, can only be dened by what
is found there, namely the cauldron. The cauldron is kindled by the breath
of nine maidens, those gures who keep emerging from the mist for over
a thousand years of Celtic history, in Britanny, Wales, the Fortunate Isle
and Gloucestershire. But what is their connection with re? I may be
stretching a point here, but the answer may lie with the Irish saint Brigit,
who has the same name as the Irish goddess Brigit (Exalted One), whose
British counterpart is Brigantia, the goddess of the Brigantes of northern
Britain. Giraldus Cambrensis tells us42 that at Kildare there is a perpetual
62
re tended by nineteen nuns, who take it in turns to watch the re for

nineteen nights, with Brigit taking her turn every twentieth night. The
re, says Giraldus, is surrounded by a hedge, made of stakes and brushwood, and forming a circle, within which no male can enter. It is only
lawful for women to blow the re, fanning it or using bellows only, and
not with their breath (unlike the nine maidens in the Four-Cornered
Fortress).
Now Brigit was the goddess of poetry and prophecy, and Brigantia
was equated with Minerva, goddess of poetry, wisdom, crafts and magic,
so this would strengthen the argument that we are dealing here with a
cauldron of inspiration. It is said, however, that the cauldron does not boil
the food of a coward, and we can infer from this that a poet must make a
dangerous symbolic voyage into the depths of the Otherworld and eat (or
drink) from the cauldron of regeneration before he can draw on the cauldron of inspiration. If the Four-Cornered Fortress is the high point of the
symbolic voyage, it can only be reached by those brave enough to descend
to the depths.
Until now the poet has in a way been mapping out the symbolic voyage he is about to make, but the voyage proper begins with the next stop,
Caer Vedwit, the Fortress of Mead-Drunkenness. This at rst brings
euphoria and possibly a trance state as we are transported back to the FourCornered Fortress, the isle of the strong door (the gateway to higher
knowledge), and the uniting of opposites in the mingling of water and jet.
But not long after comes the next stage, Caer Rigor, the Fortress of Hardness rigor may be translated as hardness, but also as the stiffness produced by cold, perhaps the stiffness and coldness of a trance state.
This seems to mark a transition from one stage of the symbolic voyage
to another, since it is at this point that the poet begins complaining about
the little men. It is now that the poet reaches Caer Wydyr, the Glass
Fortress, where communication becomes difcult, either because the poet
is in a trance state, or because he is symbolically like a small child who
cannot talk. This is perhaps reinforced by the next stage, Caer Golud, if
it means Fortress of Hindrance.
By now the poets complaints about the little men have intensied,
as if they are preventing him from completing his symbolic voyage. However, he does reach the last two stages, Caer Vandwy, the Fortress of Gods
Peak, and Caer Ochren, the Fortress of Enclosedness or Fortress of the
Sloping Hill. It is impossible to know whether the poet has completed the
symbolic voyage, but judging from Parts VII and VIII, where monks howl
3 Arthurs Voyage
63
like a choir of dogs or pack together like young wolves, and the poet
expresses his sadness and loss, it seems highly unlikely. After all, Caer
Vandwy and Caer Ochren both sound like points in an actual or symbolic
landscape which are difcult to reach (like the Fortress of Hindrance, perhaps) and, once reached, lead on to some higher realm.
In summary, The Spoils of Annwn may on one level be the story of a
military expedition, but on another level it is the account of a symbolic
voyage in search of a metaphorical rebirth through a bringing together of
the dark forces of Caer Sidi, the Mound-Fortress, and the forces of light
represented by Caer Pedryuan, the Four-Cornered Fortress. The voyage
appears to fail, however, just as it failed for Gweir, now imprisoned in the
Otherworld. It fails because the traditions and oral poetry of a culture
whose ultimate prize is a mystical cauldron are being drowned out by
little men who no longer appreciate the beauty of the cauldron and all
that it represents, who have all but forgotten the knowledge accumulated
over centuries and handed down from father to son. The Spoils of Annwn
is probably the closest we will ever get to a Druid text, a eeting glimpse
into a vast and ancient tradition gone beyond recall.
And what of Arthur? In the poem he plays no obvious role, implying
that he is a kind of spirit-guide. We will be returning to Arthur later in
the book, when we try to determine how a warrior and king can also
initiate the Taliesin-poet and his companions into the mysteries of the
Otherworld.
Some Notes on the Author of The Spoils of Annwn

Was the author of The Spoils of Annwn a Druid? We will probably
never know, but this enigmatic poem does give us some clues about its
author. The most important clue is that the poem is plainly not a Christian
poem, although the author pays homage to the Lord at the beginning
and end of the poem, and refers to Christ in the last line. We can infer
from this that the poem was composed in a society where the old British
religion still existed, but was gradually being replaced by the new Christian
religion.
Christianity came to Britain slowly over a period of several hundred
years. The martyrdom of St. Alban, which occurred between about A.D.
200 and A.D. 300, suggests that Christianity was established in Britain
well before it became the ofcial religion of Rome under the Emperor
64
Constantine. We know little of Christianity in Britain after the withdrawal

of Roman forces, but the fact that the British cleric Gildas wrote De excidio
et conquestu Britanniae (On the ruin and conquest of Britain) in the early
6th century, castigating the sinfulness of its rulers, suggests that there was
a sizeable Christian community in Britain at the time.
Another important clue is the reference to the cauldron kindled by
the breath of nine maidens. If I am right, this cauldron and the nine maidens can be associated with Gloucestershire, with the genii cucullati and the
triple mother-goddesses, divine gures before Christianity transformed
them into sorceresses. Gloucestershire was also an important focus of the
underworld cult which seems to have been observed in large parts of Southern England, from Gloucestershire to Dorset. According to the AngloSaxon Chronicle, three towns in the territory of the Dobunni, Corinium
(Cirencester), Glevum (Gloucester) and Aquae Sulis (Bath), fell to the
Saxons in 577, and Gloucestershire was incorporated into the kingdom of
the Hwicce by 600. Interestingly, the Anglo-Saxon monk Bede, in his Historia ecclesiastica gentis Anglorum (Ecclesiastical History of the English People),
completed in about 731, does not tell us how the Hwicce converted to Christianity, and we can only surmise that the Dobunni/Hwicce remained Christian from Roman times (we saw in Chapter 2 that the Roman temple at
Uley became a church, and Dorothy Watts43 demonstrates that there were
probably also very early churches in Woodchester Roman villa, Chedworth
Roman villa, and possibly on the site of St. Mary de Lode, Gloucester).
So it is unlikely that the Taliesin-poet came from Gloucestershire. By
around 650, the only parts of England that remained securely in British
hands were Devon and Cornwall, in the kingdom of Dumnonia, and west
Herefordshire, in the kingdom of Ergyng. The Taliesin-poet could therefore have lived in Ergyng, which was close to Gloucestershire and centered
on the Roman ironworking settlement of Ariconium, but it is more likely
that he lived in Gwent, also close to Gloucestershire and centered on Venta
Silurum (Caerwent) or perhaps Caerleon (Isca Augusta), a Roman legionary
fortress and settlement. Gwent is thought to have been Christianized by
Cadoc, who ourished in the 6th century, and Dubricius, who died in
612. However, Britain only became part of the wider Christian community
when Pope Gregory sent Augustine to England in 597 to convert the pagan
king Aethelberht of the Kingdom of Kent to Christianity. Augustine
became the rst Archbishop of Canterbury, but despite a decree from the
Pope, was unable to extend his authority to the bishops of Wales and
Dumnonia.
3 Arthurs Voyage
65
I see the Taliesin-poet as the descendant of people from Gloucestershire, perhaps even of a Druid priest (the Romans suppressed the Druids,
but they probably survived as an underground movement, and made a
comeback when Roman power in Britain declined and paganism was
restored in the late 4th century). He lived perhaps in Gwent around A.D.
700, but grew up listening to stories of the nine maidens of Gloucestershire
and their magic cauldron, and of heroic exploits in the underworld. His
family may have been Christians, and he himself may nominally have been
a Christian, but his heart was with the old religion and the mysteries of
that land called Annwn.
Chapter 4
Magic Mounds, Sea People

and Shape-Shifters
The Wonderful World of the Mabinogion
Introduction
The tales that make up the Mabinogion (more correctly, Mabinogi)
are thought to have been composed between 1060 and 1200. Here I will
be focusing on the Four Branches of the Mabinogion, which are:
First Branch: Pwyll, Prince of Dyfed
Second Branch: Branwen, Daughter of Llyr
Third Branch: Manawydan, Son of Llyr
Fourth Branch: Math, Son of Mathonwy
The word mabinogion, or rather, mabinogi, originally meant the (collective) material pertaining to the god Maponos.1 It traditionally includes
not only the Four Branches, but also the so-called native tales (derived
from Welsh tradition and legend), one of which, Culhwch and Olwen, we
are already familiar with, and another of which, Lludd and Llefelys, we will
examine later in the book. Finally in the Mabinogion are three romances,
Welsh versions of Arthurian tales one of these is Peredur, Son of Efrawg,
which features the nine sorceresses of Gloucester.
Only one character appears in all four branches, and that is Pryderi,
son of Pwyll; Pryderis mother Rhiannon appears in two branches (the
First and Third), and Pryderis later companion Manawydan also appears
in two branches (the Second and Third). Otherwise the Four Branches
appear to be separate, which obviously presents the analyst with a problem.
So to simplify discussion of the Mabinogion, I am going to approach it as
the story of three sets of characters, the children of Llyr and the children
of Don, whose names are self-explanatory, and the Mound-People, whose
name is less transparent. I will look at each Branch in turn and in order,
drawing on the translation and notes of Will Parker.
66
4 Magic Mounds, Sea People and Shape-Shifters
67
The Mound-People (1):

Pwyll, Pryderi and Rhiannon
The First Branch is called Pwyll, Prince of Dyfed, and this raises two
immediate points: Pwyll, unlike most Welsh heroes, apparently has no
father or mother, and he has an unusual name. Pwyll means mind, spirit,
reason, and Matasovic, in An Etymological Lexicon of Proto-Celtic, links
pwyll to Old Irish ciall and ultimately to the Indo-European root kweysperceive,2 which also gives Old Indian cayati observe, citi understanding, and Old Greek tereo observe. As the First Branch opens, Pwyll,
ruler of the seven cantrefs (administrative divisions) of Dyfed, is out hunting with his dogs and sees another pack of dogs bring down a stag dogs
that were dazzling bright white and with red ears. 3 Pwyll drives away
these strange dogs and lets his dogs feed on the stag. The owner of the
strange dogs turns up and berates Pwyll for his discourtesy. This man, it
emerges, is Arawn, king of Annwn, the Otherworld, and to make amends
to Arawn, Pwyll agrees to take Arawns place for a year and a day then
ght Arawns enemy Hafgan (Summer-Bright). After this he assumes
the title of Pwyll Pen Annwn Pwyll Head of Annwn.
It might appear from this that we are in the same mythical territory
as The Spoils of Annwn, but in fact this Annwn is a harmless place, rendered
safe by centuries of Christianity. However, potentially more dangerous are
the otherworldly events that follow Pwylls sojourn in Annwn. Pwyll is in
Arberth, a chief court of his, when after the rst course of a feast, he
goes for a walk and makes for the top of a mound which was above the
court and was called Gorsedd Arberth. Arberth is modern Narberth in
Pembrokeshire Koch says4 that the name is probably derived from Celtic
are-kwert-, referring to what stands before a dyke or hedge, though a
connection with arberth sacrice is not impossible. A gorsedd is the later
Welsh equivalent of the sidi of The Spoils of Annwn, and therefore a place
combining the darkness of the burial-mound and the dazzling brightness
of the heavenly abode. One of Pwylls court knows this because he says to
Pwyll5: Lord, ... it is a peculiarity of the mound that whatever high-born
man might sit upon it, he will not go away without one of two things:
either wounds or blows, or his witnessing a marvel.
As Pwyll and his court were sitting on the gorsedd, they could see a
woman on a large stately pale-white horse, a garment of shining gold brocaded silk about her, making her way along the track which went past the
mound.6 Pwyll sent one of his men to nd out who this woman was. The
68
horse was proceeding at an even, leisurely pace, so the man went after
her as fast as he was able to on foot, but the greater was his speed, the
further away from him she became.7 The man then fetched a horse and
pursued her, with the same result. The next day Pwyll pursued her on
horseback, and the same thing happened until he spoke to her, whereupon
she stopped and conversed with him. It turned out that her name was Rhiannon (Divine Queen), and she was promised to another man, but wanted
Pwyll to rescue her from this man in a years time, at the court of her father.
As most commentators on the Mabinogion agree, Rhiannon is the
equivalent of the Celtic horse-goddess Epona (see Chapter 2), and a sovereignty goddess, that is, a goddess whom the king must ritually marry in
order to assume the kingship (like the goddess of mead Medb mentioned
in Chapter 3). As we will see in Chapter 6, she can also be linked to the
rider-god known as the Thracian Horseman, whose horse also proceeds
at a walking pace. After the year laid down by Rhiannon, Pwyll goes to
the court of Rhiannons father, and a feast is prepared for him. As they
start their after-dinner drinking, a young man approaches Pwyll and says
he has a request, and Pwyll tells him he will grant him any request he can.
The young man then tells Pwyll he is about to sleep with the woman he
loves, and asks for Rhiannon, along with the provisions and victuals which
are here. Telling Pwyll that there was never a man so slow with his wits
as you were [just] then, Rhiannon gives him a bag, and says she will
arrange to meet the young man Gwawl in a years time to sleep with him,
when Pwyll will also be there with the bag. After the year has passed, Pwyll
goes to the court wearing dull rags and big rag-boots, and during the
after-dinner drinking, he approaches Gwawl and begs him to ll his little
bag with food. But Rhiannons bag is a magic bag, so however much they
threw in, it was no more full than before. Pwyll then advises Gwawl to
press his feet down on the food in the bag and say, Enough has been
placed herein then immediately turns the bag so that Gwawl is head
over heels in the bag, closes the bag and ties it up. Not long after this,
Pwyll sleeps with the sovereignty-goddess Rhiannon.
The magic bag also links Rhiannon to Epona, who was a goddess of
fertility, often shown with a cornucopia. Rhiannons link to Epona is reinforced in the third and last part of the tale. Here Rhiannon gives birth to
a boy, and on the night of his birth, some women are assigned to keep
watch over him. They fall asleep, and when they awake next morning, the
little boy has disappeared. The women then kill a puppy and smear the
blood on Rhiannon, accusing her of killing the baby. Rhiannon is then
69
punished in the following way8: There was a mounting-block by the gate.

She had to sit beside it every day telling anyone coming by the whole story
(of those she supposed did not know it) and offering whichever guest and
stranger would allow themselves to carried, to be carried on her back to the
court. While Rhiannon is serving out her punishment as a horse, her baby
boy is discovered in strange circumstances. One night a certain lord,
Teyrnon (=Great Lord) Twryf Liant, ruler of Gwent, is watching over his
prize mare, who has just given birth to a foal inside his house. There is a
commotion, and Teyrnon goes outside to investigate, and when he returns
he sees a small child in swaddling clothes wrapped in a sheet of brocaded
silk. Teyrnon and his wife name this child Gwri Golden-Hair. Eventually
they realize that this boy must be the lost child of Pwyll and Rhiannon,
and return him to his rightful parents, who rename him Pryderi, usually
interpreted as meaning care, anxiety. The tale ends with Pryderi growing
up, becoming the ruler of Dyfed after the death of Pwyll, then conquering
the three cantrefs of Ystrad Tywi (now Carmarthenshire in southwest Wales)
and the four cantrefs of Ceredigion (now Cardiganshire in mid-west Wales).
After this he marries Cigfa, granddaughter of Gloyw Wallt, the mythical
founder of Gloucester (Caer Gloui, in Historia Brittonum).
Pryderi is clearly linked to horses, like his mother Rhiannon, and is
also linked to Mabon (Maponos), who was stolen from his mother when
he was three nights old. His infant name of Gwri also identies him with
Gweir, the prisoner of The Spoils of Annwn, and his wife Cigfa permits us
to associate him with Gloucester. Pryderi, like Pwyll, is an unusual name,
and while it is conventionally translated as care, anxiety (Welsh pryder),
the Welsh academic Ifor Williams9 links it to an Old Breton word pritiri
loss, which is even more appropriate in the circumstances. An even more
radical interpretation is given by the Czech folklorist Joseph Baudis,10 who
sees Pryderi as derived from the IE root qrt, which appears in Irish cruth,
Welsh pryd form, aspect, the Old Welsh name of Britain Prydain, and
the Irish name of the Picts, Cruithne. This is an intriguing suggestion
which we will return to later in the book.
The Children of the Sea: Bran, Branwen

and Manawydan
Analysis
The Second Branch of the Mabinogion, Branwen, Daughter of Llyr,
focuses on the exploits of the children of Llyr, Bran, also known as Bran
70
the Blessed, his sister Branwen, his brother Manawydan, and his two half
brothers from his mothers side, Nisien and Efnisien. The mother of all
ve is Penarddun, daughter of Beli, also known as Beli Mawr; the father
of Nisien and Efnisien is Euroswydd. As the story opens, Bran is the
crowned king of this Island, and exalted with the crown of London.11
According to Matasovic,12 the name Llyr is from Proto-Celtic liro- sea
ocean, Old Irish ler, Middle Welsh lirou seas, oceans, meaning that his
offspring are the Children of the Sea. Just what this means will become
clear in the course of this chapter.
Before I say anything more about Llyrs children, I need to outline
the story of the Second Branch. One day Bran and his followers were
sitting above the ocean at Harlech (northwest Wales), when they saw a
eet of ships approaching. Matholwch, king of Ireland, wished to marry
Brans sister Branwen, and landed to discuss the matter with Bran. There
was great feast at Abberfraw (on Anglesey, and once the capital of
Gwynedd), and Matholwch slept with Branwen there. The next morning
Efnisien, a quarrelsome man, noticed Matholwchs horses and asked who
they belonged to. When he discovered that they belonged to Matholwch,
and that Matholwch had just slept with his sister, he was angered that his
sister had been given away to Matholwch without his consent, and mutilated Matholwchs horses, the symbol of his royal authority according to
the First Branch. On hearing the news, Matholwch left the court. When
Bran discovered what had happened, he offered to replace the horses that
had been mutilated, and also gave him a silver rod as thick as his nger.
Matholwch returned, but when he and Bran were dining together, Matholwch seemed out of sorts so, thinking the compensation he had given
him was insufcient, Bran also offered Matholwch the cauldron of regeneration that plays such an important part in The Spoils of Annwn.
Matholwch then cheers up, and the two exchange stories about the
origin of the cauldron. Bran says he got it from an Irishman called Llasar
Llaes Gyfewid and his wife Cymidei Kymeinvoll, who had escaped from
the Iron House in Ireland. Matholwch then tells Bran about the history
of these two individuals. One day while hunting he was on top of a tumulus
(gorsedd) above a lake in Ireland called the Lake of the Cauldron. A large,
monstrous man emerged from the lake with a cauldron on his back, accompanied by an even bigger woman. Matholwch took them in, but they
insulted and injured people, and made themselves generally hated. However, the two, plus their children (they reproduced warriors every three
months!) would not go of their own free will, nor could they be forced to
71
go. So Matholwch and his people decided to forge a solid iron chamber,
surround it with charcoal, and re it until it was red hot. The man and
woman escaped and presumably ed to Wales with their cauldron. This
is a fascinating story, and the use of an Iron House to imprison the two
monstrous creatures reminds us that according to Pliny, the Druids of Gaul
gathered selago without the use of iron, suggesting that iron may have
been considered impure. If the Druids shunned the use of iron, it is only
a short step to imagining that they feared it and envisaged it as capable of
being used in the building of a place where they could be imprisoned.
Interestingly, Oosten suggests13 that the Iron House is an inverted cauldron
which kills warriors instead of regenerating them.
After this, Matholwch returned to Ireland with Branwen, who bore
him a son. Eventually people started talking about the humiliation that
Matholwch suffered at the hands of Efnisien, and mocked Matholwch. In
reaction, Matholwch drove Branwen from his bed, and made her work as
a baker in the court. Branwen then raised a starling, tied a letter around
the birds wing, and sent it off to Bran. Once the message was delivered,
Bran decided to attack Ireland the ocean was not extensive then, so he
went by wading, carrying the string-minstrels on his back (Bran, it
should be said, is a giant). The Irish decided to make peace with Bran,
and offered to give the sovereignty of Ireland to Gwern, Branwens son,
and to build Bran a house large enough to contain him. The house was
built, but the Irish laid a trap, hanging a crane skin-bag on each of the
hundred columns of the house with an armed ghting man in each.
Efnisien came ahead of the others, and asked what was in the bags. When
he was told Flour, he felt in each bag and crushed the head of each
hidden man.
Then Bran with his entourage arrived and sat down with the Irish to
conclude the peace deal. Bran and Manawydan then spoke to their nephew
Gwern; Gwern went up to Efnisien to speak to him, and Efnisien threw
him into the re. Branwen tried to go after him, but Bran restrained her;
the Irish lit the re under the cauldron of regeneration, and started regenerating their dead warriors. Efnisien then crawled among the Irish corpses,
was thrown into the cauldron, and stretched himself out until the cauldron
broke into four pieces.
In the end, the British were victorious but, as we saw in Chapter 3,
only seven survived, including Manawydan, Pryderi and Taliesin. Bran,
wounded in his foot with a poisoned spear, ordered the survivors to cut
off his head and bury it in the White Hill in London. But, he said, they
72
would take a long time to get to London: they would spend seven years
feasting in Harlech, with the birds of Rhiannon singing to them; and
they would spend eighty years at Gwales in Penfro (thought to be
Grassholm off the southwest coast of Pembrokeshire). In the meantime,
he said, the head would remain uncorrupted and be as good company to
you as it was at its best when it was ever on me. Once arrived on the coast
of Wales, they discovered that Casswallawn son of Beli had overrun the
Island of Britain and become crowned king in London. They made for
Harlech and feasted there for seven years; as soon as they began to eat and
drink, there came three birds, which began to sing a kind of song to them;
and when they heard that song, every other [tune] seemed unlovely beside
it these were presumably the mysterious birds of Rhiannon. After their
allotted seven years they headed for Gwales, where Of all the grief that
they had witnessed or experienced themselves there was no longer any
memory. Eventually, of course, one of the seven survivors opened the
door to Cornwall which was supposed to remain closed, and they were
forced to leave Gwales and head for London, where Brans head was buried.
Commentary
Who are the children of Llyr? The answer to this question may lie
not in the mythological elements of the tale (the cauldron of regeneration,
the Iron House, the sojourn in the otherworldly Gwales), but in the names
of the children and their relatives. The name Bran means crow or raven
in Welsh, which reminds us that Lugdunum, the city of Lugus, was associated with ravens, and that ravens were found buried in the shaft at Jordan
Hill temple in Dorset (see Chapter 2). However, Koch14 believes that Brans
name also has a historical basis.
In 280279 B.C., an army of Celts, coming from Pannonia (an area
around the Danube), invaded Macedonia (northeast Greece). One division
of the army, led by Bolgios, inicted heavy losses on the Macedonians and
killed their king; another contingent, led by Brennus, urged an attack on
Greece. The Greeks sent their cavalry and light infantry to meet Brennuss
forces at the river Spercheios; the Greeks broke down the bridges, but
Brennus and his army crossed further downriver, where the river formed
a marshy lake. Eventually Brennus defeated the Greeks at Thermopylae,
and marched on to Delphi, home of the famous oracle. There, according
to the Pausanias, 15 the forces of Brennus were defeated, and Brennus,
wounded by a spear or javelin, took his own life. However, Diodorus Sicu-
73
lus16 has a slightly different version: In the mighty battle fought [at Delphi]
he lost tens of thousands of his comrades-in-arms, and Brennus himself
was three times wounded. Weighed down and near to death, he assembled
his host there and spoke to the Gauls. He advised them to kill him and
all the wounded, to burn their waggons, and to return home unburdened.
Pausanias makes it clear that Brennus and his forces were defeated at
Delphi, but apparently rumors circulated that the Celts had managed to
sack Delphi and loot it. Strabo, in his Geography, written over two centuries
later, reports17 that treasures said to have been looted from Delphi were
found in Tolosa (Toulouse) by the Roman statesman and general Caepio
in 105 B.C., though to his cost, for it was on account of having laid hands
on them that Caepio ended his life in misfortunes for he was cast out
by his native land as a temple-robber, and he left behind as his heirs female
children only, who, as it turned out, became prostitutes, as Timagenes has
said, and therefore perished in disgrace. Strabo further says that the treasure found in Tolosa amounted to about fteen thousand talents (a talent
is about 70 lbs), part of it in sacred lakes, mostly gold and silver bullion.
However, he says, treasure was commonplace in Gaul, especially in lakes
but he adds that the temple of Tolosa was hallowed, since it was very
much revered by the inhabitants of the surrounding country, and on this
account the treasures there were excessive, for numerous people had dedicated them and no one dared to lay hands on them.
On the basis of the similarities between the Brennos and Bran stories
(both crossed water to engage the enemy, both were wounded by a missile,
both begged their men to kill them, and the treasures looted from Delphi
were ritually buried in lakes just as Brans head was ritually buried in London), Koch believes that Bran is in one sense Brennos, and that the story
was transmitted orally across the Celtic world until it assumed its present
form in Branwen, Daughter of Llyr.
Manawydans name is cognate with that of the Irish god Manannan
mac Lir, the god of the sea, but Manawydan seems to have no connection
with the sea, or with the Isle of Man, from which his name is said to be
derived. Koch proposes18 that Manawydans name is derived from Mandubracius, the king of the Trinovantes who was deposed by Cassivellaunus
and restored to power by Julius Caesar (see Chapter 2). This proposal is
given some support by the fact that when Bran and the seven survivors
return to Wales, they discover that Bran has been deposed by Casswallawn
son of Beli.
Beli is also the father of Penarddun, the mother of Bran and Man-
74
awyddan. Beli or Beli Mawr, says Koch,19 is a legendary Welsh ancestor at

or near the prehistoric opening of several royal pedigrees. In the Lludd and
Llefelys he is said to be the father of Lludd and Caswallon or Casswallawn.
The belief that he was father of Cassivellaunus probably arose from a phrase
in the Historia Brittonum,20 where the British king who fought against
Caesar is called Bellinus lius Minocanni (Belinus the son of Minocannus), a textual garbling of Adminius Cynobellini Britannorum regis lius
(Adminius son of Cunobelinus king of the Britains). Koch believes that
the name Beli actually derives from Bolgios or Belgius, the name of the
chieftain who fought alongside Brennus in the campaign against the Macedonians and Greeks. Later Bolgios or Belgios may have become identied
with the Belgae of northern Gaul, and been seen as an ancestor-gure.
As to why Bran and Manawydan are children of the sea, this may
have something to do with British prehistory. Nobody knows for certain
when Celts rst settled in Britain (see Chapter 5 for further discussion),
but it is likely to have happened during the period 450 B.C.200 B.C.,
which saw a large-scale migration of people called Celts (it was during this
migration period that Brennus and Bolgios attacked Macedonia and
Greece). Later, after about 100 B.C., Belgae from northern Gaul also settled
in parts of southeast England and from the perspective of those settled
there, these Belgae would have come from the sea. It may be signicant
that in Triads 8 and 52, Llyr is referred to as Llyr Half-Speech, implying
that the sea-people spoke a dialect that was only half-understood. Curiously, East Walloon (Walloon is the French dialect of Belgium) and Coastal
Dutch (Dutch is the other national language of Belgium) share some
phonological (sound) features with Welsh,21 suggesting a particularly close
link between the language of Belgic Gaul and British (the language that
became Breton and Welsh).
The Mound-People (2): Pryderi, Rhiannon,

Manawydan
The Third Branch of the Mabinogion, Manawydan, Son of Llyr, follows on from the Second Branch, with Manawydan and Pryderi burying
Brans head in the White Hill in London. Manawydan does not wish to
live with his cousin Casswallawn who is now king of Britain, so Pryderi
offers him the seven cantrefs of Dyfed, along with marriage to his mother
Rhiannon. The two then return to Dyfed, and Manawydan assumes the
75
rulership of the seven cantrefs of Dyfed, and marries the willing Rhiannon.
One evening Pryderi and his wife Cigfa and Manawydan and his wife
Rhiannon were feasting at Arberth when they decided to take a stroll on
the gorsedd. As one of Pwylls courtiers said in the First Branch, if a highborn man sat on the gorsedd, he would not leave without wounds or
blows or his witnessing a marvel, and this is indeed what happened as
they were sitting there22:
Suddenly there was a clap of thunder and, with such a great clap of thunder, a fall of mist so that no-one could see anyone else. After the mist,
everywhere [was lled] with bright light. And when they looked where
before they would have once seen ocks and herds and dwellings, they
could see nothing at all: neither house, nor animal, nor smoke, nor re,
nor man, nor dwellings; [nothing] except the empty buildings of the
court, deserted, uninhabited, without man or beast within them, their
own companions lost, without them knowing anything about them; [noone left] except the four of them.
So the four of them wandered the countryside, but found nothing, and
for a year lived off hunted meat, sh and wild swarms. Tiring of this, they
made for England and took up saddle-making; Manawydan began to
fashion pommels, and they were coloured in the way he had seen Llassar
Llaes Gyfnewid do with blue azure (Llassar is the escapee from the Iron
House in the Second Branch who gave Bran the cauldron of regeneration).
Manawydans saddles and pommels were so good that the other saddlers
decided to kill him and Pryderi. Hearing about this, the four left for
another town, and started making shields, but the same thing happened;
so they moved on and became shoemakers, with the same results; fed up
by this time, they then moved back to Dyfed. They made for Arberth and
once more lived by foraging and hunting.
One morning Manawydan and Pryderi were hunting, when their dogs
ran ahead into a small copse, then withdrew again swiftly, all bristling and
fearful. Manawydan and Pryderi approached the copse and saw a shining
white boar, which left the copse. They went after the boar, which led
them to a great, towering caer which they had never seen before. The
dogs and boar entered the caer, and Pryderi followed them, despite the
warnings of Manawydan. When he went inside the caer, all he could see
was23
a fountain with marble stonework around it. Beside the fountain [was] a
golden bowl, attached by four chains, which was above [the] marble
76

slab with the chains reaching up into the air, and he could not see the
end of them.
He was inspired at the beauty of the gold, and [at] how good the workmanship of the bowl [was]. He came up to where the bowl was, and laid
hold of it. As soon as he had laid hold of the bowl, his hands stuck to the
bowl, and his two feet to the slab on which he was standing. The power of
speech was taken from him so he could not utter a single word. And thus
he stood [unable to move].
Manawydan then told Rhiannon what had happened. Rhiannon went

inside the caer and got stuck to the bowl like Pryderi, after which there
was a peal of thunder, a mist fell, and the caer disappeared along with Pryderi and Rhiannon.
Manawydan and Cigfa then went back to England, where Manawydan
once more became a shoemaker, with the same results as before. The two
then returned to Dyfed, and this time Manawydan took with him a bushel
of wheat. He planted three elds of wheat, which grew exceptionally well,
but as he was about to harvest two of the elds, he found them stripped
bare. He kept watch on the third eld, and discovered the culprits were
mice. He managed to catch one of the mice, larger and slower than the
rest, and was about to hang it, when a bishop approached and offered
Manawydan large sums of money to release the mouse. It turns out that
the bishop is Llwyd Cil Coed, who put magic on the seven cantrefs of
Dyfed to avenge Gwawl (Rhiannons suitor who was humiliated by Pwyll
and Rhiannon), and the mouse is the bishops wife. Manawydan releases
the mouse, the enchantment is lifted, and Pryderi and Rhiannon return.
As in the First Branch, the focus of the action is the gorsedd or mound,
where the magic is rst put on the seven cantrefs of Dyfed. The central
episode in this chain of magic is the episode where Pryderi and Rhiannon
get stuck to the golden bowl in the mysterious caer an episode we have
commented on in Chapter 3 as being similar to the imprisonment of Gweir
in Annwn. The man responsible for the magic is Llwyd Cil Coed: Parker
says his name means The Grey One of the Wooded Cell, and is thought
to be a corruption of the Irish Lath Mac Celtchair, who is mentioned in
the Irish Metrical Dindshenchas 6324 (dindshenchas means tradition or lore
of places). Liath is the son of Celtchar or Celtchair, who lives in Dun
Lethglaise, also known as the Mound of Down. He was in love with Bri,
the daughter of Midir, who was one of the sons of the great god of Irish
mythology, the Dagda, and lived in the sidh of Bri Leith. Midir refused
to let his daughter see Liath, and she died of a broken heart. If Llwyd is
related to Liath, then he is also one of the Mound-People.
77
It is signicant that Pryderi and Rhiannon are largely victims in the

Third Branch, and it is Manawydan, the Child of the Sea, who is the man
of action he is best at all the jobs they do when they are in England, and
it is he who nally frees Dyfed from its enchantment, and rescues Pryderi
and Rhiannon from their imprisonment.
The Children of the Earth: Gwydion and Arianrhod

Analysis
The Fourth Branch of the Mabinogion, Math, Son of Mathonwy, concerns the children of Don, Arianrhod, Gwydion and Gilfaethwy, together
with their uncle Dons brother Math son of Mathonwy, and Arianrhods son Lleu. At the time, Math was lord of Gwynedd (north Wales),
and Pryderi was lord of the twenty-one cantrefs in the south (Dyfed,
Ceredigion. Ystrad Tywi, mentioned in the Third Branch, plus Morganog,
now Glamorganshire).
Any discussion of the Fourth Branch must start with Don. Don is
often compared to the Irish Danu, mother of the Irish gods, the Tuatha
De Danann (Peoples of Danu), whose name is sometimes linked to the
Indo-European root ad(u)- water current.25 However, in his discussion
of Don, Koch26 follows Carey27 in his interpretation of the Fourth Branch
as a creation myth. This, says Koch, suggests a different etymology for the
name Don, namely that it is cognate with Old Irish genitive, dative and
accusative singular don place, ground, earth, and with Greek khthon
earth. Before I look at the names of the other characters, I need to summarize the story.
Math, it seems, had a strange disability: he could not live except when
he had his feet in the lap of a maiden, unless he had to go to war. His
footholder at the time was Goewin, and Gilfaethwy fell in love with her,
but could do nothing as long as Math was around. Gwydion asked him
what the matter was, but Gilfaethwy was afraid to say anything, because
of Maths ability: whatever whisper, however small, that there might be
between people, once the wind has met it, he will know it.28 But Gwydion
guessed what the problem was, and said he would arrange a war so that
Gilfaethwy could be alone with Goewin. Gwydion approached Math and
told him that Pryderi had a new kind of animal, called a pig, sent to him
by Arawn, lord of Annwn, and that he, Gwydion, would get hold of the
pigs for Math. He and Gilfaethwy then went to Pryderis court disguised
78
as bards, and Gwydion, who was the best storyteller in the world, delighted
the court with entertaining recitals and storytelling. Gwydion then asked
Pryderi for the pigs, but Pryderi said he could not give them to Gwydion
until they had bred twice their number. Thinking Pryderi might be willing
to exchange the pigs, Gwydion then used his magic to create twelve greyhounds, twelve steeds and twelve golden shields. Pryderi agreed to give
Gwydion the pigs in exchange for these items, and Gwydion and Gilfaethwy hurried away with the pigs, for, said Gwydion, the magic would
not last from one day to the next.
As expected, Pryderi then went to war with Gwynedd, and Math set
out with his army, leaving Gilfaethwy alone with Goewin, who was put
to sleep with Gilfaethwy in Maths bed, and slept with against her will.
Meanwhile, the battle between the armies of Math and Pryderi was joined,
and there was great slaughter on both sides. Finally it was agreed that since
Gwydion had caused all the trouble, the matter should be decided by single
combat between Gwydion and Pryderi, and as the tale says, through
strength and valour and aggression and magic and enchantment Gwydion
prevailed, and Pryderi was killed.29 When Math returned to court and
discovered that Goewin had been raped by Gilfaethwy with Gwydions
help, he punished Gwydion and Gilfaethwy in a rather unusual way30:
Since you have been in league together, I will make you fare together and
be mated. You will have the same nature as the beasts whose shapes you
are in; and during this time, they will have offspring so you will have
them too. A year from today, come to me here.
After a year to the day, lo! he could hear an uproar below the wall of the
chamber, with the dogs of the court barking on top of that uproar.
[Go and] see whats outside, said he.
Lord, someone said, I have looked. There is a stag and a hind and a
fawn with them.
At that, he arose and came outside. When he came, what he could see
was the three creatures. The three creatures were a stag, a hind and sturdy
fawn.
The same punishment was then repeated, with the hind becoming a wild
boar, and the stag a sow; after a year, these two returned with their offspring, and were again transformed, with the wild boar becoming a shewolf, and the sow a wolf. A year later, they returned with their offspring,
and were changed back into their human form.
Since Math no longer had a footholder, Gwydion suggested his sister
Arianrhod, daughter of Maths sister Don. First Math had to be sure Arianrhod was a virgin, so he asked her to step over his magic wand. As she
79
did so, she gave birth to a large boy with curly yellow hair, and shortly
after that a little something dropped from her Gwydion picked it up
and wrapped a sheet of brocaded silk around it and hid it in a small chest
at the foot of his bed. The boy was baptized Dylan and immediately made
for the sea. Not long after, Gwydion heard a cry from the chest, opened
it, and saw a little boy. He took the boy to a wet nurse, and the boy grew
rapidly, much as Pryderi did in the First Branch.
Gwydion virtually became the boys father, and one day he took the
boy to see Arianrhod at her home, Caer Arianrhod, a rocky island off the
coast of Gwynedd and opposite the hillfort of Dinas Dinlle (City of Lleus
Fortress). Arianrhod welcomed them, but on discovering that the boy
was hers, became upset and, learning that he had no name, swore an oath
on him, that he would not get a name unless he got it from her. To overcome this curse, Gwydion conjured up a ship, disguised himself and the
boy as shoemakers, and sailed to Caer Arianrhod. When Arianrhod saw
the shoemakers, she ordered some shoes, but Gwydion deliberately made
them too large. When Arianrhod came to complain, the boy was stitching31:
Suddenly, there was a wren alighting on the deck of the boat. The boy
took aim and hit it between the sinew and the bone of its leg. She
laughed.
God knows, said she, the fair one strikes it with a skillful hand!
Aye, he replied, and the wrath of God upon you! He has obtained a
name, and the name is good enough. Lleu Skillful Hand he will be from
now on.
When Arianrhod realized who they were, she swore another oath on the
boy that he would never take arms unless she armed him herself. To
overcome this oath, Gwydion disguised himself and Lleu as bards, and
went to Caer Arianrhod. They spent the night there, and in the morning
Gwydion used his magic to make everyone believe there was an army
approaching Caer Arianrhod. Arianrhod then provided Lleu with all the
arms he needed, and Gwydion dispersed his enchanted army. Angry at
this, Arianrhod swore a destiny on him that he would never get a wife
from any race that is in the world today.
To overcome this destiny, Math and Gwydion decided to conjure a
wife for him out of owers, using our magic and enchantment.32 So they
took the owers of the oak, the owers of the broom, and the owers of
the meadowsweet and from those they called forth the fairest and most
beautiful woman anyone had ever seen. She was baptised with the baptism
80
they practised [back] then, and [the name of ] Blodeuedd was put upon
her. Lleu and Blodeuedd were married, and while Lleu was visiting Math,
Blodeuedd met and fell in love with a passing lord called Gronw. They
decided to kill Lleu, and not long afterward Blodeuedd asked Lleu about
how he might be killed, on the pretense of being worried about him. It
turned out he could only be killed by a spear made under certain conditions, in a bath by the side of a river, while Lleu had one foot on the bath
and one foot on the back of a buck (young male deer). Blodeuedd said she
didnt understand, and asked him to demonstrate. So the bath was set up
by the river, and goats were brought (a misunderstanding on the part of
Blodeuedd). Llleu demonstrated what he meant, and Gronw threw the
specially made spear at Lleu, who took ight in the form of an eagle, and
gave a terrible scream.
On hearing the news, Gwydion set out to nd Lleu. One day he followed a sow, until she stopped to graze. He went under a tree and saw that
the sow was grazing on rotting esh and maggots. He looked up into the
top of the tree.33
When he looked up, he could see an eagle in the top of the tree. When the
eagle shook himself, worms and rotting esh fell from him, and those the
sow was devouring. It occurred to him that the eagle was Lleu, and he
sung an englyn:
An oak grows between two pools,
Dark-black branches sky and glen
If I do not tell a lie
From the owers of Lleu this has come!
The eagle let himself down until he was in the middle of the tree. [Then]
Gwydion sang another englyn:
An oak grows upon a high plain
Rain neither wets it, nor drips upon it
Nine-score strikes has it endured
In its top, Lleu Skillful-Hand.
And then he let himself down until he was on the lowest branch of the
tree. Then [Gwydion] sang an englyn:
Grows an oak upon a steep
The sanctuary of fair lord
Unless I speak falsely:
Lleu will come down into my lap.
And he fell onto Gwydions knee; and then Gwydion struck him with a
magic wand, until he was [back] in his own form. However, no-one had
ever seen a man in a sorrier state. He was nothing but skin and bones.
81
Afterwards, Lleu was restored to health, Gwydion changed Blodeuedd into

an owl, and Lleu threw a spear at Gronw, as Gronw had done to him,
breaking Gronws back and killing him.
Commentary
The rst general point to make about the Fourth Branch is that
Gwydion kills Pryderi in single combat early on in the tale, suggesting
that the Children of the Earth have in some sense replaced the MoundPeople. The second general point is that the Children of the Sea are apparently nowhere to be seen, with two possible qualications: in Triad 35,
Arianrhod is said to be the daughter of Beli Mawr; and in the course of
the Fourth Branch, Lleu is referred to as one of the Three Golden Shoemakers, named in Triad 67 as Casswallawn, Manawydan and Lleu.
As for the main characters, much can be learned by looking at their
names. The name Math was rst studied by the French historian and
philologist Henri dArbois de Jubainville. He says34 that the word math
can be recognized in the term math-ghamhuin bearcub (literally calf of
bear), used in the Irish translation of the Bible to render the Hebrew dob
bear. The word matus bear appears as the rst term in the Gallic mens
names Matu-genos Son of the Bear, that is, son of the Bear-God, and
Matu-murus Great like a Bear, that is, as great as the Bear-God. The
Gallic god Matunus is also derived from this root, and a variant of this,
Matunnos, provided the second element of the Gallo-Roman name of Langres, Ande-matunnum (Great Bear), fortress of the divinized great bear.
There is also an Irish proper name Mac-Mathghamhna, which today is
written as Mac-Mahon, and means son of the bearcub. To this list we
may add, says Boekhoorn,35 Math son of Mathonwy, Math mac mir in
dru, the druid of the Tuatha De Danann according to the Book of Invasions,
and Matgen (Son of the Bear), the corrguine, or sorcerer, of the Tuatha.
It is not clear in what sense Math is a bear, but he can certainly be seen
as a Druid and sorcerer in fact, in Triad 28, Math is said to have taught
one of the Three Great Enchantments of the Island of Britain to Gwydion
son of Don.
Turning now to Gwydion, Koch36 notes that in the genealogy of
Brycheiniog (a kingdom in southeast Wales) contained in the Harleian
3859 manuscript, we nd the Old Welsh name Lou Hen map Guidgen. If
this can be interpreted as Lleu the Old, son of Gwydion, then it means
(1) Gwydion is actually the father of Lleu which would certainly explain
82
Arianrhods violent reaction toward Lleu and Gwydion and (2) the name
Gwydion is derived from Widu-genos Son of a Tree. As for Arianrhod,
her name is usually translated as Silver Wheel, and she is seen in terms
of the moon, especially since the name Lleu is associated with light
(see Welsh lleuad moon37). The eminent Welsh scholar John Rhys (1840
1915) said that Caer Arianrhod is also used in Welsh to denote the
constellation Corona Borealis.38 Lleu is usually identied with the god
Lugus39: his epithet Skillful-Hand is similar to that of the Irish Lugh,
who is called Lamhfhada Long-Arm or Long-Hand; he was celebrated
as a shoemaker, and as we saw earlier, there is a dedication in Roman
Iberia to Lugus from the shoemakers guild; and the second element
of Dinas Dinlle, the Welsh hillfort referred to earlier, is exactly the same
as Lugdunum, but in reverse order. The association between Lleu and
Lugus may seem rather tenuous, but I hope to reinforce it in later chapters.
It was said earlier that Math, Son of Mathonwy is a creation myth,
and this is supported by the fact that Lleu may be the child of a brothersister relationship. Elizabeth Archibald40 notes that in the Greek creation
myth Cronus married his sister Rhea, and was subsequently deposed by
their son Zeus, and in Egyptian myth the brothers Osiris and Set married
their sisters Isis and Nephthys, since the rst stages of creation permit,
indeed necessitate incest. Fee and Leeming tell us41 that in Norse mythology Njordrs sister was said to have borne him his children Freyr and Freya,
and no discernable shame seems attached to this union.
If Math, Son of Mathonwy is a creation myth, then Gwydion and Math
are the creative forces of the tale. They are both Druids when they made
Blodeuedd, one of the ingredients was owers of the oak, and when
Gwydion found the eagle Lleu, he was perched on an oak tree. Math has a
magic wand, and both Math and Gwydion can perform magic. The ancient
Greek and Roman writers said that the Druids studied astronomy, and perhaps the Welsh knew this, because, as Rhys points out, not only is Caer Arianrhod the Corona Borealis, but Llys Don (Dons Court) is the constellation
Cassiopeia, and Caer Gwydion is the Milky Way. Classical writers also
afrmed that the Druids believed in reincarnation, and certainly a good deal
of shape-shifting occurs in this tale, with Math changing Gwydion and Gilfaethwy into a variety of animals, and Lleu taking the shape of an eagle.
This reincarnation or shape-shifting is most noticeable is some of the poems
attributed to Taliesin. In Angar Kyfyndawt (Cruel Bondage or Hostile Alliance),
as translated by Sarah Higley, the poet says42:
83
I was a blue salmon,

I was a dog, I was a stag,
I was a buck on the mountain,
I was a trunk, I was a spade (?)
I was a drinking horn in the hand,
I was a peg in forceps
for a year and a half
I was a speckled white rooster
among chickens in Edinburgh.
I was a steed in a herd,
I was a erce bull,
I was a millers billygoat,
like the nourisher (?)
I was a scaly grain,
it grew on a hill,
I will be reaped, I will be placed
into the chimney I will be driven,
I will be torn from the hand,
by my bagfulls,
a red-clawed hen received me,
combed foe.
I rested nine nights in her womb as a lad.
The next poem in The Book of Taliesin, Kat Godeu (The Battle of the
Trees), continues in a similar vein43:
I have been in a multitude of shapes,
Before I assumed a consistent form.
I have been a sword, narrow, variegated ...
I have been the dullest of stars.
I have been a word among letters,
I have been a book in the origin.
I have been the light of lanterns,
A year and a half.
I have been a continuing bridge,
Over three score Abers.
I have been a course, I have been an eagle.
I have been a coracle in the seas:
I have been compliant in the banquet.
I have been a drop in a shower;
I have been a sword in the grasp of the hand
I have been a shield in battle.
It is hard to know what to make of these transformations: some sound

like possible reincarnations, some sound like shape-shifts, and some
sound like poetic identication with nature (and culture) in all its facets.
84
The Battle of the Trees is a difcult poem. In her notes to the poem, Mary
Jones points out44 that the theme of a battle of the trees can be seen in the
Irish story The Second Battle of Magh Tuiredh, between the Tuatha De
Danann and the Fomorians (the preCeltic gods of Ireland), in which
Lugh said:
And ye, O Be-cuile and O Dianann, said Lugh to his two witches, what
power can ye wield in the battle?
Not hard to tell, said they. We will enchant the trees and the stones
and the sods of the earth, so that they shall become a host under arms
against them, and shall rout them in ight with horror and trembling.
Whatever the battle may be, later in the poem the narrator tells us:
I was enchanted by Math,
Before I became immortal,
I was enchanted by Gwydyon
The great purier of the Brython ...
I was enchanted by the sage
Of sages, in the primitive world.
When I had a being.
He then goes on to say:

I travelled in the earth,
Before I was a procient in learning.
I travelled, I made a circuit,
I slept in a hundred islands
A hundred Caers I have dwelt in.
Ye intelligent Druids,
Declare to Arthur,
What is there more early
Than I that they sing of.
The Taliesin-poets reference to the primitive world does suggest that

he is talking about reincarnation, and evokes the Irish story of Tuan
mac Cairill,45 which rst appeared in the 12th-century Book of the Dun
Cow. This describes how a man who rst came to Ireland as part of
the rst settlement after the biblical ood lived on though all the subsequent phases of its legendary history by passing through the shapes of a
stag, a boar, a cormorant and a salmon; with each metamorphosis his
youth was renewed. The fact that the poem also refers to Math and
Gwydion, to Druids and to Arthur (possibly the Arthur of The Spoils of
Annwn), seems to reinforce the link between the Children of Don and
Druidism.
85
Conclusion
As I said at the beginning of this chapter, the Mabinogion concerns
three sets of characters, the Mound-People, the Children of the Sea (Llyr)
and the Children of the Earth (Don). We saw that from a historical perspective, the Children of the Sea seemed to be linked to gures from early
history, like Brennus, Mandubracius, Bolgios and Cassivellaunus. The
other two sets of characters seem largely ahistorical, though the MoundPeople seem to be associated with the horse-goddess Epona, and the Children of the Earth seem to be connected with Druidism. In the following
chapter I will try to put a face to the Mound-People and the Children of
the Earth by going back deep into the prehistory of southern England.
Chapter 5
Mounds, Mounds, Mounds

Rubbish Heaps, Hillforts and the
Prehistory of Southern England
Prelude
If we want to nd the origins of the Mound-People and the Children
of the Earth, one very good place to start looking is southern England, or
rather an area of southern England to the southwest of London stretching
from the Marlborough Downs and Kennet Valley in the northern part of
Wiltshire through the Vale of Pewsey to Salisbury Plain in the south of
Wiltshire, and east into what is now the western part of Hampshire. This
is a good place to look for ancestors because the ancestors certainly left
their mark there, with West Kennet Long Barrow, Avebury Stone Circle,
Silbury Hill and, of course, the most impressive mark of all, Stonehenge.
Ill be focusing on three of these West Kennet Long Barrow, which is
certainly a burial-mound or sidi; Silbury Hill, which legend says is the last
resting-place of King Sil; and Stonehenge, which Geoffrey of Monmouth1
called the Giants Dance, and which he said had been erected to commemorate the 460 British treacherously slain by the Saxon king Hengist.
West Kennet Long Barrow is a Neolithic tomb or barrow, situated on
a prominent chalk ridge near Silbury Hill, and one-and-a-half miles south
of Avebury (see Map 4). According to Wikipedia, archeologists classify
it as a chambered long barrow, and one of the Severn-Cotswold tombs
(other examples are Hetty Peglers Tump, near Uley, Gloucestershire, site
of the much later Roman temple, and Waylands Smithy in Oxfordshire).
It has two pairs of opposing transept chambers and a single terminal
chamber used for burial. The stone burial chambers are located at one end
of one of the longest barrows in Britain at 330 feet: in total it is estimated
that 15,700 man hours were expended in its construction. The entrance
86
5 Mounds, Mounds, Mounds
87
consists of a concave forecourt with a facade made from large slabs of

sarsen stones which were placed to seal entry. Construction of the barrow
began around 3600 B.C., and it was in use until around 2500 B.C.
Julian Thomas says2 that
initial burial deposits included both articulated bodies and clean bones,
and it seems skeletal parts were taken out of the chambers as well as placed
in them. Seemingly the chambers were entered and used over a long
period, with their contents being repeatedly reorganized. If the deposition,
removal and repositioning of human remains were not casual activities,
then the spatial conguration of the chambers would have limited the
numbers of people who could have been present and witnessed these
events. So even if access to the tomb were not generally limited, then this
kind of ritual activity would have generated a form of socially restricted
knowledge.
In terms of the deposition of bones, the westernmost chamber was dominated by the bodies of adult males, while there were many young persons
in the southeast transept.
At some time during the centuries that followed, says Thomas, a
series of secondary deposits were introduced into the chambers, consisting
of alternating layers of clean chalk and burnt material with a higher organic
content.... The dark, organic layers in particular contained large quantities
of broken pottery, stone tools and waste, and human and animal bones.
Piggott, the excavator, noted the diversity of the cultural material in the
secondary deposits, and the presence of Grooved ware and Beaker pottery
at the base of the sequence. The excavator Piggott argued, says Thomas,3
that the material must have accumulated over a lengthy period, before
being placed in the tomb. This meant that the artefacts must have originally been placed in someoffering-house which would receive ritual or
votive offerings including pottery, and contain the debris of ritual meals
including animal bones and the ashes of the hearth. Thomas goes on to
say4 that the artefacts were treated in the same way as the dead bodies: just
as the bodies were brought into the chambers whole, allowed to rot, parts
of them removed, and the remaining elements reorganized in various ways,
so the pottery vessels in the secondary deposits had been broken and the
parts scattered.
Around 2500 B.C., the passage and chamber were lled to the roof
by Beaker people (of whom more later) with earth and stones containing
the secondary deposits excavated by Piggott. Barry Cunliffe notes5 that
the culture of the Beaker people took over only gradually in the Kennet
88
Valley area, after a prolonged struggle with older entrenched orthodoxies

represented by monuments like West Kennet Long Barrow. The rst Beaker
culture burials occurred well away from existing ritual centers, as if these
were reections of incompatible systems of belief and social practice.
Only the later forms of the Beaker culture occur at such ceremonial sites,
and their growing power is demonstrated at West Kennet Long Barrow,
which was blocked with massive stones as if to emphasize the nal assumption of symbolic authority by beaker users.
Silbury Hill lies on the valley oor of the River Kennet, at a point
where the capacity of the stream is increased by contributions from a small
tributary together with water from a spring where the stream changes its
course. The underlying rock is Cretaceous White Chalk, and the area is
essentially part of the extensive Lower Chalk plateau that extends northwards towards Swindon. The origin of the name Silbury is unclear,
though David Field reports6 that the earliest name is Seleburgh, which may
derive from Old English sele hall. At 130 feet high, says Wikipedia, Silbury Hill is the tallest prehistoric human-made mound in Europe, and
one of the largest in the world. It is thought to have taken 18 million
man hours, or 500 men working 15 years to deposit and shape 8,800,000
cubic feet of earth and ll on top of a natural hill.
According to Richard Atkinson, who excavated the site between 1968
and 1970, the hill comprised an initial smallish mound of gravel and turf,
surmounted later by a second mound of chalk with a surrounding ditch,
over which was constructed the massive chalk mound that we see today.
The nal mound, he thought, was built in tiers like a wedding cake, arguing
that the now-invisible lower terraces had become obscured by weathering.7 Atkinson believed that the hill was Neolithic, and two small trenches
cut on the summit (in 2001) by Fachtna McAvoy of English Heritages
Centre for Archaeology revealed for the rst time a fragment of antler
from a secure context, lying against a chalk wall in a deposit of chalk
rubble. This produced a secure radiocarbon date of between 24902340
B.C., placing the mound rmly in the Late Neolithic.
Atkinson thought the hill had been built in tiers, but recent survey
work by English Heritage suggests otherwise. The uppermost terrace,
when circumnavigated, returns to a point several meters below the starting
point in other words, it spirals down. Weathering has obscured the detail
of ledges further down the slope; but if we assume the feature is continuous,
it implies a spiral path all the way from the summit to the base. The idea
of a spiral path, says Field, is attractive. It would not only have provided
89
Map 4: Salisbury Plain and surrounding areas c. 1000 B.C. with Neolithic
monuments, and sites of early eld systems and later ranch boundaries.
a processional route to the top of the hill, but would also have aided the
construction of the mound.
Nobody really knows why Silbury Hill was built in fact, during
Atkinsons and earlier excavations at Silbury, many more Roman and early
medieval nds were produced than Neolithic. In the 19th century, a substantial Roman building was found south of the hill. Cutting into the
mound at ground level in 1867, Wilkinson found a platform just below
the surface on which was a pile of ashes associated with Romano-British
artefacts. Atkinson himself found over 100 Roman coins in the ditch, while
numerous Roman shafts and wells were found nearby.
Then, over the past decade, the Roman evidence increased. A combination of pipeline excavations and aerial photography has revealed the
extent of Romano-British settlement on the lower slopes of Waden Hill
a natural hill next to Silbury. It now seems almost inconceivable that, with
a Romano-British settlement facing the mound, Silbury was not used in
some way. It is possible, says Field, 8 that the mound was once covered
90
with burials, monuments and memorials, and that what has been found
so far is the tip of the iceberg.
Atkinsons medieval evidence
was no less striking. Small postholes containing iron nails, early medieval
potsherds and a silver coin of Ethelred II dating to 1010 suggested that the
terraces had been revetted by posts. He also found an iron spearhead, and
explained the evidence as defence against the Danish invasions not
unreasonably, given the skirmish at nearby East Kennet in 1006 reported in
the Anglo-Saxon Chronicle. He concluded that the original terraces had
been recut and fortied. The postholes, however, are located on the inside
of each terrace and therefore imply revetment rather than defence.
It seems more likely, says Field, that Silbury Hill was used as the site
for a prestigious building, perhaps a Christian building, in the early
medieval period. If the mound had been sacred in the Roman era, it is
possible that it retained its religious attraction in later centuries. He suggests9 that it may have been this prestigious building that gave the mound
its name Seleburgh (Hall-Fortress).
The ditch surrounding Silbury Hill, says Field,10 is often considered
a mere quarry from which the mound material was derived. However, its
circular nature, and the regularity of its rectangular western extension,
indicate that it served more than a functional purpose:
Archeologists have come to see that ditches, even massive ditches around
henges or hillforts, need not always be just utilitarian structures but may
have had a metaphysical function too for example, to keep evil spirits at
bay. The rectangular extension at Silbury, if waterlled, would have served
as a cistern or reservoir. Elsewhere in the world, cisterns have often been
the focus of ritual and ceremony. The mirror-like quality of standing water
may have had symbolic implications too. Given archeologists fascination
with shamanism, it is signicant that mirrors are considered symbols of
shamanic ceremony and power.
The People of Stonehenge (1): The Builders

While Silbury Hill was being built, some 20 miles to the south, on
Salisbury Plain, Stonehenge was also under construction. The Stonehenge
site has a very long history. According to Wikipedias excellent article on
Stonehenge, archeologists have found four or ve large Mesolithic postholes
dating to around 8000 B.C. beneath the nearby tourist car-park. At least
three of the posts were in an east-west alignment which may have had
91
ritual signicance; no parallels are known from Britain at the time but
similar sites have been found in Scandinavia. The rst monument at
Stonehenge itself, says Wikipedia, which is dated to around 3100 B.C.,
consisted of a circular bank and ditch enclosure ... measuring about 360
feet in diameter with a large entrance to the northeast and a smaller one to
the south.... The builders placed the bones of deer and oxen in the bottom
of the ditch as well as some worked int tools. The bones were considerably older than the antler picks used to dig the ditch, and the people who
buried them had looked after them for some time prior to burial.... Within
the outer edge of the enclosed area was a circle of 56 pits, each about 3 ft
3 in. in diameter, known as the Aubrey holes after John Aubrey, the 17thcentury antiquarian who was thought to have rst identied them.
In the two centuries that followed, at least 25 of the Aubrey holes were
used for cremation burials.
The Stonehenge Riverside Project, headed by Mike Parker Pearson
of Shefeld University,11 has carbon-dated a number of burials at Stonehenge, and reports that the cremation burial of an adult found in Aubrey
Hole 32 dates from 30302880 B.C.; the cremation burial of a young or
mature adult from the middle ll of the Stonehenge ditch dates to 2930
2870 B.C.; human skull fragments from the northern ditch ll and the
eastern ditch ll date to 28902630 B.C. and 28802570 B.C. respectively;
the ditch was partly dug out between 2570 and 2340 B.C., and a third cremation burial was placed in this new ditch on the ditchs northern side
this burial is dated to 25702340 B.C. These nds and dates show that
Stonehenge was a cemetery from around its inception until the period of
the sarsens (26552485 B.C .). An estimated 240 people were buried at
Stonehenge over a period of around 500 years, says Parker-Pearson this
amounts to one burial every two years, and suggests they were drawn from
a very small and select living population, interred there because of their
special status as members of an elite dynasty of rulers.
Construction on Stonehenge proper is thought to have begun around
2600 B.C. At this time, two concentric arrays of holes (the Q and R Holes)
were dug in the center of the site. The holes, says Wikipedia, held up to
80 standing stones, only 43 of which can be traced today. The bluestones
(some of which are made of dolerite, an igneous rock), were thought for
much of the 20th century to have been transported by humans from the
Preseli Hills, 160 miles away in modern-day Pembrokeshire in Wales.
Another theory that has recently gained support, is that they were brought
much nearer to the site as glacial erratics by the Irish Sea Glacier.12
92
During the next stage of activity, thought to have taken place between
2600 B.C. and 2400 B.C ., 30 enormous sarsen stones were brought to the
site. They may have come, says Wikipedia,
from a quarry, around 25 miles north of Stonehenge on the Marlborough
Downs, or they may have been collected from a litter of sarsens on the
chalk downs, closer to hand. The stones were dressed and fashioned with
mortise and tenon joints before 30 were erected as a 110-foot-diameter circle of standing stones, with a ring of 30 lintel stones resting on top. The
lintels were tted to one another using another woodworking method, the
tongue and groove joint.... Within this circle stood ve trilithons of
dressed sarsen stone arranged in a horseshoe shape 45 feet across with its
open end facing northeast.
Some time after this, the bluestones appear to have been re-erected.
They were placed within the outer sarsen circle and may have been
trimmed in some way. Like the sarsens, a few have timberworking-style
cuts in them, suggesting that, during this phase, they may have been linked
with lintels and were part of a larger structure. Then, between 2280 B.C.
and 1930 B.C., the bluestones were arranged in a circle between the two
rings of sarsens and in an oval at the center of the inner ring.... Soon afterwards, the northeastern section of this bluestone circle was removed, creating a horseshoe-shaped setting (the Bluestone Horseshoe) which mirrored
the shape of the central sarsen trilithons. The last known construction at
Stonehenge was around 1600 B.C., with the digging of the so-called Y and
Z holes, two rings of concentric (though irregular) circuits of near-identical
pits (30 pits in each circuit) cut around the outside of the Sarsen Circle.
Who built Stonehenge and where did they live? We know little about
the people who rst erected the bluestones and sarsens, except that they
buried their leaders at Stonehenge, but we know more about those who
carried out later work on Stonehenge, thanks to the famous Amesbury
Archer and the lesser-known Boscombe Bowmen. The Amesbury Archer
was found in 2002 at Amesbury, three miles southeast of Stonehenge,
when archeologists from Wessex Archaeology were excavating in advance
of a housing scheme. The grave they uncovered, dated to around 2300
B.C ., contained the richest array of items ever found from this period.
Around 100 objects were found, including the complete skeleton of a man,
three copper knives, two small gold hair tresses, two sandstone wristguards
to protect his wrists from the bow string, 16 int arrowheads and ve
Beaker pots.13
According to the Wessex Archaeology website, this makes the grave
93
the richest Bronze Age nd in Britain there are ten times the usual
number of nds from other graves. The gold dated to as early as 2,470
B.C. and is the earliest found in Britain. It seems likely that the objects
were buried with the man ... for his use in the next life. Tests on the
enamel found on the Archers teeth revealed that he grew up in central
Europe. They could not reveal how long he had lived in Britain, only that
he must have lived in the Alps region while a child, either Switzerland,
Austria or Germany.
The grave of the Boscombe Bowmen was discovered on Boscombe
Down, not far from Stonehenge, during the digging of a trench for a new
water pipe in May 2003. The grave, dating to around 2300 B.C., contained
the remains of seven individuals three adult males, a teenage male and
three children. Wessex Archaeology notes: Matching the seven individuals
were eight pots. This is the greatest number of people from a single Beaker
grave in Britain and it is also the greatest number of Beaker pots from one
grave.14 Seven of the eight pots are decorated all over, six with cord and
one with plaited cord. Plaited cord on Beaker pots is extremely rare in
Britain, and one of the very few British nds was in the grave of the Amesbury Archer. The other nds include ve barbed and tanged arrowheads
giving the name the Boscombe Bowmen some other int tools, scrapers
and akes, a boars tusk and a toggle.
Wessex Archaeology notes that in continental Europe tusks are often
found in the same grave as stones used for metalworking, like the one
found in the grave of the Amesbury Archer. Sometimes the tusks and
stones have been found next to each other together, however: Only one
other bone toggle has been found in Britain, from a later, and rich, Bronze
Age burial at Barnack, Cambridgeshire. Most toggles have been found in
continental Europe. These nds usually have a central perforation rather
than a loop. They have mainly been thought of as decorative pendants but
they may have fastened clothing or a hair ornament. The enamel on the
Bowmens teeth shows that the men probably grew up in the Lake District
of northwest England, north Wales or southwest Wales, despite their links
to the Amesbury Archer and continental Europe.
But where did this Beaker culture come from? The Bell-Beaker culture, as it is formally known, was a cultural movement that spread over
wide parts of Europe between 2800 B.C. and 1900 B.C., from the upper
Danube in the east, and Germany between the Elbe and the Rhine,
through the Low Countries and France (except for the central massif ), to
Spain, Portugal, Britain and Ireland in the west. Cunliffe15 sees the Bell-
94
Beaker Phenomenon as part of the breakdown of traditional social structures and the emergence of a more mobile way of life that began in northern
Europe after 3000 B.C. The name comes from the characteristic drinking
vessel with its inverted bell-shaped prole, which carries incised decoration
in horizontal zones around the body (see Figure 7). These pots were typically placed in single male burials, often accompanied by weaponry and
covered by a circular mound. It has been demonstrated from pollen grains
found in the bottom of beakers, says Cunliffe,16 that they were also used
for something like mead, avored with herbs such as meadowsweet (whose
owers were used to make Blodeuedd in the Fourth Branch of the Mabinogion) or wild fruits. It was these Beaker people, some from as far away as
the Alps, others indigenous adapters of the culture, who blocked West
Kennet Long Barrow with massive stones, and were associated with the
elaboration and refurbishment of Stonehenge around 2000 B.C.17
While all this work was going on at Stonehenge, where did the huge
labor force live? The answer to this question is to be found in Durrington
Walls, the site of a Neolithic village and henge enclosure two miles northeast of Stonehenge. At some point around 2600 B.C., says Wikipedia, a
large timber circle now known as the Southern circle was constructed. The
circle was orientated southeast towards the sunrise on the midwinter solstice and consisted of 4 large concentric circles of postholes, which would
have held extremely large standing timbers. Some time later, perhaps 200
years after the circle was rst constructed, another two concentric rings
were added and the henge enclosure was constructed. A ditch some 18 feet
deep was dug, and the earth used to create a large outer bank some 100
feet wide and presumably several feet high. Mike Parker Pearson believes
that Durrington Walls was a complementary structure to Stonehenge
evidenced by the similar solstice alignments. He suggests18 that the timber
circle at Durrington Walls represented life and a land of the living, whilst
Stonehenge and the down around it, encircled by burial mounds, represented a land of the dead. The two were connected by the River Avon and
their respective avenues, and a procession route from one to the other represented the transition from life to death.
Durrington Walls was a land of the living not only because the henge
was made of timber, but also because there was a village there. It was a
large circular village of many hundreds of houses.... This settlement, with
a circumference of almost a mile, would have been the largest village
known in northwest Europe. If the density of houses from the 20042007
excavations (one house per 400sq ft) is representative of their packing
95
around the settlement then we might expect over 300 houses to survive
beneath the henge banks.19
Preliminary results from environmental analyses suggest that this was a

seasonal settlement. The absence of carbonized grain or quern stones and
the lack of bones from neonatal pigs and cattle, together with the evidence
for culling of pigs in the midwinter period, suggest that people journeyed
here with their pre-prepared foodstuffs and animals only at certain times
of the year.20
The People of Stonehenge (2): The Stonehenge God

The Amesbury Archer and the Boscombe Bowmen tell us a good deal
about Stonehenge around 2300 B.C., but what of later years? As it happens,
we can learn a great deal about the successors of the Amesbury Archer by
examining another burial place, the Normanton Down Barrows. The
English Heritage website on Stonehenge says that the round barrow cemetery at Normanton dates from around 2600 to 1600 B.C. and it is one of
the most impressive groups of burial mounds in the Stonehenge landscape.
This barrow cemetery lies to the south of Stonehenge. It contains Bronze
Age round barrows of several types, as well as an earlier Neolithic long
barrow. There is a clear line of sight from these barrows to Stonehenge,
just over half a mile away.
The earliest ritual use of this site is most likely to have been the construction of the long barrow between 3400 and 3000 B.C. Related to it,
and dating from roughly the same time, is an enclosure used for preparing
the dead for burial. The body was left in the open air and progressively
cleaned of all esh by the wind and the birds, leaving the bones ready for
the burial a process of de-eshing known as excarnation.
According to Stuart Needham, Andrew Lawson and Ann Woodward,21
the Normanton Down cemetery can be related to the rst ve phases which
Needham has dened for the British Bronze Age. Particular care needs to be
given to interpreting the earliest Beaker burials (Period 1, that is, 2500 B.C.
2300 B.C.) since they are frequently not associated with any surviving mound,
and there could be unknown graves almost anywhere. Nevertheless, they suggest early Beaker graves were deliberately sited in the zone of much older long
or oval barrows to the west. Later Beaker burials (Periods 23, that is, 2300
B.C.1700 B.C.) reinforce this pattern; some were sooner or later covered by
mounds and thus came to be targeted by the early excavators.
96
The picture, Needham et al. say,22 changes radically in the next phase
(Period 3, that is 2050 B.C.1700 B.C.), actually named after Bush Barrow,
which will be discussed shortly.
Setting aside the late Beaker graves, there are now arguably three foci
emerging, most strikingly the famous Normanton linear cemetery along
the main ridge. Two graves yielded both daggers and belt hooks (Bush
itself, and Wilsford G23, where archeologists have found a unique bronze
hook corroded against the dagger), and two had just belt-hooks (G15,
G18). In addition, there are two graves with classic rich ornament sets (G7,
G8) and another with matching bead types (G16).
Assessment of whether the pattern then changes is blurred by the less

well dated burials, but the diagnostic Camerton-Snowshill burials (Period
4, that is 1700 B.C.1500 B.C., and named after a particular type of dagger)
are differently sited. More dramatic, they say, is the marked shift seen in
the Middle Bronze Age (15001150 B.C.). By the end of the early Bronze
Age (Periods 14) the great majority of barrows would have existed, but
there are some clusters of small mounds with evidence for burials in
Deverel-Rimbury urns (the name of a particular type of pottery) and, by
implication, Middle Bronze Age construction. Only one cluster is attached
to a major concentration of earlier barrows, that in the Lake group. There
seems to have been an almost complete retreat from Normanton Down;
the focus of attention has shifted to the bounded eld domains lying just
to the west where there is also settlement evidence and the remarkable
Wilsford shaft (a very deep feature interpreted as a well or a ritual shaft),
which we mentioned in Chapter 2.
The most famous burial at Normanton Down is undoubtedly the one
at Bush Barrow, thought to date from between 1900 B.C . and 1700 B.C. It
was excavated in 1808 by Sir Richard Colt Hoare and William Cunnington.
As James Dyer says23:
On the oor of the barrow lay the skeleton of a stout and tall man, a
Wessex chieftain. Near his head was a wooden shield with bronze decorations. An axe had been wrapped in cloth and placed by his shoulder. At
his side lay two bronze daggers, one with a wooden hilt decorated with
thousands of minute gold pins, each 0.04 inches long. There was a small
bronze dagger by his right hand, and near his leg a very curious perforated stone. This seems to have been a stone mace with a head made
from a perforated fossil tubularia (a plant-like animal closely related to
jellysh) and a handle of wood decorated with bone rings. Although the
mans clothes had not survived, some of the objects sewn on to them were
found.
97
On his chest had been a lozenge-shaped plate of sheet gold measuring 7.3
inches (see Figure 8); his belt had a hook of hammered gold with nely
engraved ornament.
Mike Pitts24 has an interesting theory about the man in the Bush Barrow burial. He points out that in the grave there were three large metal
daggers and one axehead. While daggers are not uncommon in these graves,
axes are rare. But it is a combination we see in the center of Stonehenge:
a single dagger carved on the face of stone 53, surrounded by a group of
at least 14 axes, and further axes on the outer faces of sarsens 3, 4 and
5. It is, says Pitts, as if the stones have become a symbolic grave, the
mythical corpse represented by carvings of precious and distinctive objects
normally buried with people granted elaborate and expensive funerary rituals.
Interestingly, there are thirty axes carved on sarsens 3, 4 and 5, the
same as the number of megaliths in the ring. Stonehenge, says Pitts, has
been usurped by a man and his retinue, accompanied by objects of mortal
existence. The stones are no longer ancestors, but merely an extraordinary
version of something everyday, the burial mound. The man is an extraordinary transformation of humanity, something alien to earlier cosmologies:
a god. His symbolic tomb is enclosed by two rings of thirty graves (the
Y and Z circles), matching the number of stones and axe carvings, just as
barrows for the living are surrounded by a ditch. Beyond Stonehenge, the
hundreds of round barrows are also arranged in two broad bands, as if in
echo of the central barrow, so that the entire cemetery landscape becomes
a symbol of this divine focus.
Controlling the Land: Southern England

in the Middle and Late Bronze Age
The rich culture which produced the Bush Barrow burial did not
exist in isolation from the rest of Europe. Amber beads found on Normanton Down are thought to have come from the Baltic, and faience beads
from later graves may have come from somewhere in Europe or the
Mediterranean, possibly exchanged for Cornish tin, one of the basic ingredients of bronze. Some of these trade links may have stretched over long
distances. Harding notes25 that study of particular amber forms has shown
that there is a specic connection between southern Greece (Mycenae) and
southern England, since the so-called amber spacer-plates of both areas
98
were fashioned in identical ways (spacer-plates were used to separate the

strands of a multi-strand necklace); Sciama and Eicher also point out26
that a gold-framed amber disc found in a tomb at Knossos in Crete so
closely matches Wessex examples that it could be from the hand of the
same goldsmith.
These long-distance trade links remind us that when the elite of
Stonehenge were being buried on Normanton Down, other civilizations
were ourishing in the Mediterranean. Egypt was already an ancient civilization when the Bush Barrow god was laid to rest, using hieroglyphics
to record its history and produce a variety of literary texts; the Minoan
civilization of Crete was ourishing and keeping records in the undeciphered Linear A script; the Hittite Empire dominated Anatolia for most
of the second millennium B.C. and used a script based on the Akkadian
cuneiform of Mesopotamia to produce a variety of texts; and from 1600
B.C. the Mycenean Greeks dominated Greece and later Crete, using the
Linear B script to keep records in an ancient form of Greek.
What the Stonehenge people knew of these civilizations is unclear,
but it is possible that an event in the eastern Mediterranean in the middle
of the second millennium B.C. did have a profound impact on them. Some
time around this date (possibly around 1628 B.C.), the volcano of Santorini
(Thera) erupted. This eruption is believed to have been one of the largest
volcanic events on Earth in recorded history, throwing into the stratosphere
approximately 24 cubic miles of material, four times the amount of material ejected by Krakatoa. We do not
know what effect this cataclysmic eruption had on the people of Stonehenge,
but we can make an educated guess.
The dendrochronologist Mike
Baillie has studied oaks growing on the
surface of Irish raised bogs and has concluded that there was a large volcanic
dust-veil event in 1628 B.C.27 He notes
that ice-core acidity in Greenland peaks
at 1644 B.C. 20, which coincides with
the widespread occurrence of narrowest
growth rings in Irish bog oaks. In fact,
Figure 7: A typical beaker pot,
like those found in the graves of six trees on four different bogs produced
the Amesbury Archer and the their narrowest growth rings in 1624
B. C . and such extreme single-year
Boscombe Bowmen.
99
events occur only nine times in 6000 years. However, it is clear that the
event at 1624 B.C. is not restricted to a single year, for 1624 B.C. is the
low point of a reduction in growth which in general appears to start around
1628 B.C. and lasts for a number of years. This tree ring evidence is supported by data from English bog oaks: Croston Moss in Lancashire has
oaks from 3198 B.C.1682 B.C., and from 1584 B.C.970 B.C., implying a
catastrophic event between 1682 B.C . and 1584 B.C., while the Hasholme
chronology spans 1687 B.C. to 1362 B.C. and shows an extremely narrow
band of rings in the 1620s B.C .
We dont know what life was like in the 1620s B.C. for the people of
Stonehenge, but we can get some idea from A.D. 1816, the Year Without a
Summer, thought to have been caused by the huge volcanic eruption of
Mount Tambora on the island of Sumbawa, Indonesia, in 1815. Europe,
still recuperating from the Napoleonic Wars, says Wikipedia, suffered
from food shortages. Food riots broke out in Britain and France and grain
warehouses were looted. The violence was worst in landlocked Switzerland,
where famine caused the government to declare a national emergency.
Huge storms, abnormal rainfall with oodings of the major rivers of Europe
(including the Rhine) are attributed to the event, as was the frost setting
in during August 1816. Hungary experienced brown snow, and Italy experienced something similar, with red snow falling throughout the year.
If the people of
Stonehenge experienced anything like
this, then it would
not be surprising if,
keen astronomers as
they were, they took
this as a sign from the
gods and instituted
changes in their way
of life. And changes
did occur after 1600
B.C. There are indications that patterns of
worship changed
by around 1450 B.C., Figure 8: Image of the lozenge-shaped plate of sheet
votive offerings were gold found in the Bush Barrow burial near Stonebeing placed in Wils- henge.
100
ford Shaft, near Stonehenge, similar to those placed in the graves on Normanton Down (see Chapter 2). But the way of life of the people also
changed in a much more radical way. Barry Cunliffe, in Landscape with
People, says that in the middle of the second millennium B.C. a major
shift away from monument-building occurred. Instead, he says,28 people
focused on control of the land, with the creation of huge areas of regularly
laid out elds often contained by linear earthworks. One such eld system
that has been studied is at Windy Dido, on the western boundary of
Hampshire near Quarley. The elds were rst identied in the 1920s when
they were photographed from the air; further work has shown that the
system covers 90 hectares and was laid out within large rectangular blocks
running downslope from a linear ditch occupying the crest of a ridge to
the east.29 As Cunliffe says, 30 the magnitude of the scheme and the
extreme order with which it was realized implies social organization at a
high level motivated by a central coercive power.
The Windy Dido elds do not exist in isolation, for similar eld systems belonging to roughly the same period have been identied close to
later hillforts at Danebury (west Hampshire) and Woolbury (to the east
of Widy Dido), and are well known on the eastern fringes of Salisbury
Plain and on the Marlborough Downs to the north.31 The impression is
given that over a comparatively short period of time, communities began
to take control of the land to a degree never before contemplated.32 We
are witnessing, says Cunliffe, the deliberate creation of a new landscape
but one which appears to have complemented the old by leaving many of
the ancestral barrow cemeteries intact in areas of pastureland beyond the
boundaries of the arable.33 It is difcult to know why this reordering of
the land occurred, but one reason could be the need for greater efciency
in arable production, which could have brought about more far-reaching
changes in society, forcing the community to focus more closely on the
fertility of the land and the deities who controlled it34 (which might
explain the votive offerings in the Wilsford Shaft). And status, once represented by the richness grave goods, might now have been displayed in
the ability of the lineage to control land by containing it within boundaries after all, a great expanse of ordered and well-tended elds is
an impressive sight redolent of power.35
Towards the end of the second millennium, says Cunliffe,36
another signicant change to the Wessex landscape can be recognized with
the creation of extensive systems of linear ditches carving up the landscape
into huge blocks. These ranch boundaries or linears clearly represent a
101
new concept of land allotment. Whatever practical functions they may

have performed in terms perhaps of livestock management they were
above all a symbol of social reordering on an extensive scale. The sight of
these thin white lines seared through the landscape will have impressed on
the observer the centralizing power of the central authority.
Many of the linears relate to prominent hilltops, and ve or six centuries

later some of these hilltops, like Sidbury, on the eastern edge of Salisbury
Plain, Clearbury, near Salisbury, and Quarley were enhanced by enclosing
defences, creating what are generally referred to as hillforts ... which suggests that the symbolism of linears was being further reinforced at the
time. Interestingly, some of the new linear ditches cut straight through
the earlier eld systems, perhaps signifying a cancellation of the values and
authority inherent in the preceding system.37
All this happened in the 12th century B.C. or thereabouts, but was
not conned to the WiltshireHampshire area: on Dartmoor a complex
landscape of linear boundaries, enclosures, settlements and elds created
in the middle of the second millennium was widely abandoned, and a
similar large-scale abandonment is evident over much of the uplands of
northern Britain.38 What was the reason for this second reordering of the
land? The usual explanation is that there was slight increase in rainfall,
leading to the development of peat bogs, particularly in upland areas. But
Cunliffe points out39 that there was also a volcanic eruption at Hekla in
Iceland in 1159 B.C. Mike Baillie40 was able to recognize a recurring pattern
of a comparatively thick ring for 1160 B.C. followed by ten very narrow
rings representing a decade during which there was virtually no tree ring
growth. As in 1628 B.C., this volcanic eruption could well have had a profound effect on the people of southern England, forcing them once more
to institute changes in their way of life. This second reordering of land
use may have reected a greater reliance on stock management as opposed
to crop growing, as a result of the failure of the large-scale agrarian
system introduced, with so much effort, in the middle of the second millennium.41 It is possible that after ve centuries or more of intensive use
the
thin downland soils showed signs of exhaustion.... Declining yields
matched by a rise in population would have caused much social stress perhaps leading to fears that the chthonic deities controlling fertility had
ceased to favor humanity. Add to this the possibility of a decade of dust
clouds in the upper atmosphere affecting the power of the sun and one can
begin to picture the sense of foreboding that may have been generated
requiring precipitate communal action.42
102
Interlude: At Home and Abroad

Were the people who organized the distribution of land after 1628
and reorganized it after 1159 B.C., the same people as those whose
ancestors were buried in Normanton Down cemetery? We may never know,
but what we do know is that Salisbury Plain underwent the same transformations as the surrounding areas, with the rectangular eld systems of
around 1500 B.C . succeeded by the later ranch boundaries.43 We also
know that, at some time during the period when the rectangular eld systems were being replaced by the ranch boundaries, possibly around 1100
B.C., the Avenue, the ceremonial route leading to Stonehenge, was lengthened by another mile and a quarter towards the River Avon.44 And at
around the same time, two Bronze Age barrows45 were constructed on the
hill by the River Avon that was later (around 500 B.C.46) to become the
hillfort of Vespasians Camp.
All this the change in the eld system, the new work on the Avenue
at Stonehenge and the building of barrows on the hill by the River Avon
suggests a people under pressure. This pressure may have been purely internal, given the change in the climate and the disruption that must have
ensued, but we cannot entirely rule out external inuences. We dont know
whether the people of southern England knew much about events in the
Mediterranean, though we know they had long-distance trade links which
went as far as Mycenae. For a great deal was going on in the Mediterranean
at the time: the Mycenean civilization collapsed around 1100 B.C., the Hittite Empire of Anatolia disappeared around 1160 B.C., Troy was destroyed
around 1190 B.C., and all written language vanished from Europe for the
next four centuries. The reason for the so-called Bronze Age Collapse is
unknown, but as in Britain, climate may have been a factor. It seems that
there was a period of global cooling from approximately 1250 B.C. to 750
B.C.47 There is evidence that the Hittite world suffered a severe and prolonged famine in the last decades of the Hittite kingdom, and the Hittites
became increasingly dependent on shipments of grain from abroad, probably from the reign of Hattusili III onward, that is, from around 1267
B.C.48 Herodotus, in The Histories, written some 700 years after the event,
reports that when the Cretans (Mycenean Greeks) returned from the Trojan
war, famine and pestilence fell upon them, and destroyed both the men
and the cattle.49
The famine in Anatolia and the famine and pestilence in Crete are
signs of a general disruption of life in the Mediterranean, underlined by
B.C.,
103
the emergence of the mysterious Sea Peoples. This term is derived from
the term used by the ancient Egyptians to describe the people who threatened them in two major attacks c. 1210 B.C. and c. 1180 B.C.50 The Pharaoh
Mereneptah tells of his victory c. 1210 in the western desert over Libyans
who had brought with them as allies Sherden-people, Sheklesh-people,
Aqaiwasha-people of the foreign lands of the Sea ... Aqaiwasha the foreigners of the Sea. In a list of the northern enemies of Ramses III (c. 1180
B.C.) a Sherden chief is called Sherden of the Sea; with him are the chief
of the Pulisati (Philistine) foes. Another inscription of Ramses III mentions as northern enemies peoples of the sea, literally the foreign lands,
the isles who sailed over against his lands, and they included Philistines
and Tursha from the midst of the sea. Finally, in the Harris papyrus in
the British Museum, Ramses II says: I overthrew all who transgressed the
boundaries of Egypt, coming from their lands. I slew the Danuna from
their isles, the Tjekkeru and Philistines ... the Sherden and Weshesh of the
Sea were made as if non-existent.
Who were all the groups mentioned as being Sea Peoples? The Sherden
have been linked to Sardinia, the Sheklesh to Sicily; the Aqaiwasha and
Danuna are interpreted as Achaeans and Danaans, that is Greeks; the
Philistines are thought to have originated in the Aegean before they settled
on the southern coast of Canaan; the Tjekkeru and the Weshesh are
obscure; the Tursa are thought to be the Tyrsenoi, or Tyrrhenians, who
according to Herodotus,51 migrated from Lydia to Italy to become the Etruscans (Tusci).52 Clearly there were large movements of people in the
Mediterranean, from Canaan in the east to Italy and Sardinia in the west,
all driven by the deteriorating climate and resultant famines to seek new
opportunities elsewhere. Whether any of these people reached as far as
Britain is a question we will be considering in the next chapter.
The Midden Builders of the Late

Bronze Age and Early Iron Age
Meanwhile, in southern England the ranch boundaries or linears
were being constructed, and to nd out what happened after this, we need
to go to the Wiltshire village of Potterne, one mile south of Devizes (see
Map 5). In an area to the north of the village known as Blackberry, excavations between 1982 and 1985 unearthed a huge prehistoric midden (rubbish heap) that covered an area of 8.6 acres and survives to a depth of 7
104
feet. A meticulous study of the midden, says Cunliffe,53 showed it to have

accumulated over a period of some 500 years, from 1100 B.C.600 B.C., as
the result of penning livestock and adding domestic rubbish to the animal
waste and bedding material. The excavators identied fourteen successive
zones (layers) at Potterne for which a number of radiocarbon dates are
available.54 Zones 1411 produced plain pottery of the Late Bronze Age
transitional assemblage and are dated to 1100850/800, while zones 102
produced typical All Cannings Cross decorated ware for which the date
range 850/800600 is indicated.
All Cannings Cross pottery (see Figure 9) is named after a farm near
the village of All Cannings in the Vale of Pewsey, Wiltshire, 6 miles east
of Devizes, and is found throughout central southern England, centered
on the Wessex chalklands.55 There are two distinctive types of All Cannings Cross pottery: the large jar with an evenly out-curved rim and a
rounded shoulder and body which in most cases is decorated with incised
and stamped motifs, often in the form of stab-lled geometric shapes;
and the bipartite bowl with beaded rim and sharp shoulder angles,
between which are bands of decoration, either stamped or incised. The
striking, highly decorated pottery of the All Cannings Cross group has
no precedent in Wessex, says Cunliffe56:
While some of the ceramic forms can be explained away as indigenous
copies of exotic metal types, the weak proled jars with stabbed decoration
stand out as something very new that cannot be easily paralleled except
among Late Urneld assemblages in eastern France and western Germany.... Thus it is probable that an Urneld element lies behind the origins of the Early All Cannings Cross group, but it is impossible to dene
the nature of the link more precisely. There is no rm evidence for a folk
movement of any magnitude into the area, but limited inltrations cannot
be ruled out.
This intriguing suggestion that the All Cannings Cross pottery may have
come about as the result of limited inltrations, possibly from eastern
France or western Germany, is a matter we will return to in Chapter 6.
The midden at Potterne is not the only one in the area. John Barrett
and David McOmish,57 have carried out exploratory work at All Cannings
Cross and report evidence of a substantial midden less than half a mile to
the east of All Cannings Cross. But the best known midden outside the
one at Potterne is the midden at East Chisenbury, on Salisbury Plain, 6
miles to the north of Stonehenge.58 The East Chisenbury midden is circular, and spreads over an area 670 feet in diameter, 8.412.4 acres in
105
extent. The mound is 7 feet deep in places, and around 216,000 cubic feet
of material still survive. Preliminary results of the excavations carried out
there in 1992 show that the mound had built up rapidly as a complex
construction rich in cultural residue dating from between 800 B.C. and
600 B.C. Several thousand shards of pottery have been found, and display
characteristics which are entirely consistent with structured deposition.
On the composition of the mound, McOmish notes:
Initial study of the well-preserved faunal assemblage points to disproportionately large numbers of foetal or neonatal sheep; other species such as
cattle, pig and deer are well represented. Human remains, including two
fragments of skull, were uncovered. One of these had apparently been
placed deliberately within the mound, surrounded by shards of pottery
from the same vessel and a small block of sarsen stone. Linear scarring and
teeth-marks on its outer surface are inexplicable. Copious quantities of
coprolites, including human examples, were also noted.
What was the purpose of these mounds? Some have suggested they
were created by cycles of feasting, but Cunliffe59 believes there is another
explanation. He hypothesizes that the accumulation of refuse represents the
desire to create communal control over societys midden material thus commanding the potential fertility embedded within it. He notes that in earlier
archeological contexts in Wessex there is ample evidence of the use of domestic midden material in complex burial rituals. What is exceptional about the
8th- to 7th-century Wessex middens is their scale and the degree of social
cohesion that their construction implies. They must represent an intense,
but comparatively short-lived, communal response to a perceived need in
some way associated with an overwhelming desire to control fertility.
The huge mounds were not the only step taken by the people of southern England to control fertility. As Cunliffe says,60 marl digging became a
widespread phenomenon in the rst millennium (marl is a lime-rich mud
or mudstone), and shallow chalk quarries are frequently found on settlement
sites. It seems that many areas of the downland were covered with clayey
acid soils which needed to be broken down and neutralized if they were to
become productive. Interestingly, old chalk diggings were frequently used
in Late Bronze and Early Iron Age Wessex for the disposal of the dead.61
Of course, during this period the people of southern England left
other reminders of their way of life besides refuse mounds, especially in
the way of bronze artefacts. From the middle of the 12th century B.C. until
the end of the 11th, says Cunliffe,62 Britain south of the Humber was
served by bronze smiths casting a range of new implements in bronze with
106
a high lead content. The new types constitute what is known as the
Wilburton complex (Wilburton is a village in Cambridgeshire): they
include leaf-shaped swords, simple socketed spearheads, and socketed axes.
The largest nd of Wilburton artefacts is the Isleham Hoard from Cambridgeshire, buried around 1000 B.C., and containing some 6500 fragments
of bronze, including weapons of the Wilburton complex type, fragments
of bronze cauldrons and a series of harness ttings.
At the end of the 10th century B.C., south and east Britain developed
a series of new types in parallel with western areas of France, with which
close contacts were maintained. The new assemblage, known after the site
of Ewart Park (Northumberland), incorporates Continental material of
the Carps-Tongue Sword Complex, including the long sword with narrowed point which gives the complex its name, knives of hogs back and
triangular form, various types of socketed axe, socketed knives, pegged
and socketed spearheads, and various chisels and gouges.63
In addition to these cast bronze implements, vessels and shields of
beaten bronze make their appearance in the British Isles from the 10th
century, or perhaps a little earlier, as the result of far-ung contacts with
central Europe.64 Among these were bronze cauldrons, the earliest of
which, from Colchester (Essex) and Shipton (Oxfordshire), are thought
to have originated at the end of the second millennium in the wake of
contacts with central Europe and were probably manufactured in southern
Britain.65 Beaten bronze shields also appeared in the 10th or 9th century
B.C., and were widely distributed over most of the British Isles by the 8th
century recent work, says Cunliffe, indicates a possible Hungarian origin
for the type, possibly transmitted via Denmark.66
Other exotic types found their way in at this time, principally from
central and northern Europe. Among these must be mentioned the rst
appearance of bronze ttings appropriate to horse harness, which surely
implies the introduction of more sophisticated methods of harnessing and
possibly the greatly increased practice of horse-riding. Such developments
are largely in parallel with those discernible in Europe in the Hallstatt B
phase of the Urneld culture (1000 B.C.750 B.C.).67
Hillforts and Iron Age Southern England

By the 7th century, says Cunliffe,68 a more stable pattern of agricultural production was under way, and settlements now become evident
107
and long-lasting after several centuries during which evidence of where

people actually lived is somewhat sparse. But in the years following 700
B.C .,69 a rash of defended sites or hillforts spread across a large area of
southern Britain, from Kent and Sussex in the south, westward through
Wiltshire, Gloucestershire, Hereford and Worcester, and then up through
the Welsh Marches into North Wales. During the period 600 B.C .400
B.C., says Kristiansen, densely packed hillforts emerged in Wessex, in the
Danebury area, with distances of only 6 miles between each.70
The hillforts of southern Britain fall into two types small defended
village-like hillfort settlements and large hill-top enclosures. 71 The
defended village-like hillforts followed the tradition of hillfort building
which emerged in the 8th or 7th century B.C., at sites like Breddin (Powys)
and Crickley Hill (Gloucestershire). These sites were fairly small and were
set in easily defensible positions hill-tops, promontories and escarpments72; all have well-constructed ramparts, and many had elaborate gatetowers over the entranceways. Inside these forts occupation was usually
fairly dense. At Crickley Hill, two phases can be recognized: the early
phase was characterized by rectangular houses together with 4-post struc-
Map 5: Southern England c. 700 B.C . with sites of ritual middens (rubbish
heaps) in the modern county of Wiltshire.
108
tures ... generally interpreted as raised granaries73; later the rectangular

houses were replaced by round houses up to 49 feet in diameter. Further
south at Danebury (west Hampshire), the early phase, radiocarbon dated
to about 500 B.C. 80, contained many roundhouses, but instead of 4posters, which were rather few in number, circular pits or silos were dug
into the solid chalk.74 It was estimated that the total grain capacity of the
investigated pits was about 1521 cubic yards, and that the hillfort as a whole
had the capacity to store 807 cubic yards of grain, enough to feed over
1000 people for a year.75
Large hill-top enclosures, generally over 40 acres in area, defended
by one or more lines of ramparts..., but more often dened by natural
slopes,76 include Bathampton, east of Bath, Balksbury, near Andover,
Hampshire, and Ogbury, near Salisbury in Wiltshire. The function of
these large enclosures is unclear their great size and apparent low density
of occupation suggest they were stock enclosures and storage places.77
The earlier preoccupation with fertility was not forgotten with the
building of the hillforts. Cunliffe, during his excavation of Danebury hillfort, uncovered a large number of storage pits or underground silos, and
notes that they are common on all settlement sites. It had earlier been
thought that they were constructed to protect the grain from raid, but
Cunliffe believes78 that the deposition of grain in a storage pit was a deliberate act involving the placing of the seed in the protective custody of the
chthonic deities during the dangerous liminal period between the time it
was harvested and the moment it was sown. The evidence for this is that
once the pit was emptied and the seed was in the soil, the pit became the
focus of acts of propitiation, involving the structured deposition of parts
of animals, sets of artefacts and other material acts which may have
continued up to the time of the harvest and may have coincided with seasonal festivals. One such propitiatory offering at Danebury79 involves a
Figure 9: All Cannings Cross pottery like that found throughout Wiltshire,
and dating to around 800 B.C.
109
dog and a horse which has been partially dismembered, its head and one
foreleg being placed separately against the pit side.
This ritual dismemberment of the horse in the Danebury pit may be
highly signicant. Anne Ross notes80 that in Ireland a bull-feast, tarbfeis, was used to determine by mantic means the rightful successor to the
king of Tara: a bull was killed, and a druid ate of its esh and drank of
the broth in which it was cooked. The druids sang a spell of truth over
him, and in his dreams he would see the rightful king. Sometimes the
prophet had to be wrapped in the hide of the slaughtered animal. The
ritual slaughter of three bulls seems to be taking place on one plate of the
Gundestrup Cauldron (see Chapter 2), where three warriors hold swords
to the throats of huge animals, their immense size in comparison to the
men suggesting their own divinity. Three dogs bound beneath the bulls.
There can be little doubt, says Ross, that animal sacrice took place in
Ireland, and she quotes Anne Woodward as saying that many of the sacriced
beasts would have provided meat for feasting, and the numbers of bodies and
parts of bodies that have survived on Iron Age sites indicate that the practice
of animal sacrice in Celtic society must have been widespread. All the evidence, says Ross, supports Woodwards contentionCunliffes discovery of
the heads and legs of horses in various contexts at Danebury is suggestive
of the heads and hooves ritual where the body of the animal was consumed
and the hide used to wrap the seer in preparation for his mantic sleep.81
But were the people who built Danebury and other early hillforts
Celts? We saw in Chapter 4 that Celts probably settled in Britain in the
period 450 B.C. to 200 B.C., and the archeological evidence seems to support this. One very distinctive culture which shows Celtic (La Tne
inuences) is the Arras culture of Yorkshire, characterized by inhumations
with the remains of carts sometimes the carts were placed upright, or
the carts were dismantled, the wheels being placed at; grave goods include
iron cart-tires, three-link horse-bits, a bula (brooch) of the Mnsingen
type (Mnsingen is the site of a La Tne cemetery in Switzerland), and a
bracelet similar to those found in Alsace and Burgundy in early La Tne
contexts. There is a distinct possibility, says Cunliffe,82 that the Arras culture arose as the result of a folk movement into eastern Yorkshire late in
the fth or early in the fourth century. These people may have come from
Burgundy via the Paris region (the Celts of the Paris region, the Parisii,
have the same name as those of east Yorkshire).
Meanwhile, in the southeast of Britain, from the 5th century on, the
ceramic assemblages, says Cunliffe,83
110

show strong inuences from the adjacent Continent from northern France
and the Low Countries. This is most vividly demonstrated by the decorated and painted wares of the Highstead-Dollands Moor group.... The
British examples have close similarities with the vases carenes and vases piriformes of the Continental La Tne cultures. The inuence of these
ceramic innovations was felt most strongly in the Thames valley and eastern England, contributing to the regional assemblages called here the Long
Wittenham-Allens Pit group, the Chinnor-Wandlebury group, the Highstead-Dollands-Moor group, and the Darmsden-Linton group. Further
south in Sussex and west in Hampshire and Wiltshire, elements of the new
types, though apparent, are more dimly reected in contemporary ceramic
developments.
The mechanism by which these Continental styles were introduced

into Britain remains unclear. While it is possible a limited folk movement
took place bringing new peoples to Kent and the Thames valley region it
is no less likely that new ceramic styles emerged as the result of intensifying
exchange relations between the two sides of the channel.84
It seems unlikely therefore that the early hillforts were built by Celts,
though Celtic elites or fear of Celtic invaders may have played a part
in the so-called developed hillforts. In the 4th century B.C., says Darvill,85
many of the early hillforts like Crickley Hill were abandoned and not
reoccupied while others were elaborated and extended. These developed
hillforts were larger than the early hillforts. Their defenses often follow
the natural contours of a suitable hill, and the area enclosed may be as
much as 25 acres. Multiple ramparts are usual, and careful use of existing
slopes and natural features to exaggerate the scale of these ramparts suggests
that ostentation was as important as defence. At Danebury, soon after
400 B.C ., a second line of ramparts was constructed, and the eastern
entrance was particularly heavily defended, with a strategically placed
command post with a clear view over the entire entrance area and ideally
suited for slingers.86 At Danebury, as at many sites of this period,
clay sling shot or suitable natural pebbles have been found in considerable
quantities, emphasizing the importance of this form of warfare. The
interior of the site was densely occupied with circular houses set around
the inner edge of the rampart.... Some areas of the interior were given
over to storage, either in pits or in 4-posters. Roadways ran through
the site and these were maintained through the period of occupation.87
The implications, says Darvill, are that occupation was continuous,
intensive and under the control of a strong centralized power. Other
developed hillforts include Beacon Hill and St. Catherines Hill in Hamp-
111
shire, Maiden Castle in Dorset, Cadbury Castle in Somerset, and

Barbury Castle, Fosbury, Yarnbury and Castle Ditches in Wiltshire (see
Map 6).
In the Late Iron Age, says Cunliffe,88 Britain can be divided into
three broad zones: a core comprising the south-east which shared many
cultural characteristics with the Continent, a periphery comprising an arc
of coin-issuing tribes stretching from Dorset to Lincolnshire; and beyond,
that is the rest of Britain west and north of the periphery where coinage
had not been introduced into the socio-economic system. The core consists of four tribes, the Trinovantes, Catuvellauni, Atrebates and Cantii,
who shared a number of cultural aspects with the Belgic areas of northern
Gaul ceramic technology, burial rites, economy and socio-political
structure showed only slight variation from one end of the region to the
other.89 In the past it was believed that the area was settled by the immigrant Belgae referred to by Caesar, but Cunliffe says that a good case can
be made that the undoubted similarities result from regular and intensive
social and economic intercourse between the tribes on either side of the
Channel.90 The culture of these four tribes is usually referred to as the
Aylesford-Swarling culture, named after two cemetries in Kent, and dates
from around 50 B.C. The tribe of greatest interest to us is the Atrebates,
who occupied West Sussex, west Surrey, Hampshire, Berkshire and northeast Wiltshire. Their tribal capital, of course, was Calleva Atrebatum
(Silchester in Hampshire), founded around 50 B.C., and Cunetio, on the
River Kennet and near the site of the Iron Age hillfort of Forest Hill, was
also in their territory.
The four coin-issuing tribes of the periphery are the Durotriges,
Dobunni, Iceni and Corieltauvi (see Chapter 2). The Durotriges were a
close-knit confederacy, says Cunliffe,91 centered upon modern Dorset. To
the east and north
their boundaries with the Atrebates seem to have been marked by the
Avon, which ows through Salisbury Plain, and its tributary the Wylye,
which ows to the south of Salisbury Plain. The New Forest may have
served as something of a buffer zone between the two tribes. Further west
Durotrigan coins and pottery extend along the valleys of the Yeo and Parrett, in Somerset, and the River Brue in Somerset would approximate to a
northern boundary with the Dobunni
The Dobunni92 were centered on Gloucestershire, extending into

north Somerset, down to the River Brue, north and west Wiltshire,
Oxfordshire west of the Cherwell, and most of Worcestershire.
112
Conclusion
We set out at the beginning of this chapter in search of Mound-People, and we found large numbers of them, from the Neolithic builders of
West Kennet Long Barrow, Silbury Hill and Stonehenge to the Late Bronze
Age/Early Iron Age midden builders and the Iron Age hillfort builders.
We also found early signs of the horse-goddess Epona and Rhiannon of
the Mabinogion, and of Cunomaglus, the Hound-Lord of Nettleton Shrub.
We also saw clear evidence of external inuences in British prehistory: the
Beaker people from the Continent and the Amesbury Archer from the
Swiss or Austrian Alps; the Arras culture of east Yorkshire from Celtic
Gaul; the La Tne ceramic forms from Gaul; and the Belgic inuences
from northern Gaul. It also became clear that a number of powerful families dominated Wiltshire and east Hampshire from around 2300 B.C., initially through Stonehenge, later through the control of land, and later still
through the control of fertility and the construction of hillforts.
We have been left with some important questions. The rst concerns
the extent of Celtic inuence in England. Considering how little has
remained of Celtic language and even place-names outside Cumbria (where
Cumbric was spoken until the 12th century) and Cornwall (where Cornish
was spoken until the 18th century), was Celtic spoken only by an elite in
southern England, and only for a relatively short space of time before the
Romans then the Anglo-Saxons became the dominant force? The second
relates to the Mound-People themselves. What became of those who built
and watched over Stonehenge for so long, and for those who supervised
the building of the middens and hillforts? Did they become the Durotriges,
whose territory seems to have included Stonehenge, but who, according
to Cunliffe,93 remained an isolated body, with an impoverished coinage,
showing little sign of wealth accumulation or the emergence of a dominant
lite? Did they become (or were they conquered by) the Atrebates, whose
territory seemed to include the northern and eastern fringes of Salisbury
Plain? Or did they throw in their lot with the Belgic immigrants who settled south of Danebury around the hillforts of St. Catharines Hill or
Orams Arbour, in an area which later became Venta Belgarum/Winchester?94 Or did the people of Stonehenge carry on as normal at Vespasians
Camp? Dennis Price95 believes that the 4th century B.C. Greek geographer
and explorer, Pytheas of Massilia, who made a voyage of exploration to
northwest Europe about 325 B.C., visited Stonehenge. The writings of
Pytheas have not survived, but Diodorus Siculus quotes a contemporary
113
of Pytheas, Hecataeus of Abdera, in the passage that follows,96 where he

is discussing Britain:
There is also on the island both a magnicent sacred precinct of Apollo
and a notable temple which is adorned with many votive offerings and is
spherical in shape. Furthermore, a city is there which is sacred to this god,
and the majority of its inhabitants are players on the cithara; and these
continually play on this instrument in the temple and sing hymns of praise
to the god, glorifying his deeds.... And the kings of this city and the
supervisors of the sacred precinct are called Boreadae, since they are
descendants of Boreas, and the succession to these positions is always kept
in their family.
Price contends that the sacred precinct of Apollo refers to Stonehenge,

and that the city where the supervisors of the sacred precinct live refers
to Vespasians Camp.
The nal question raised by this brief prehistory of southern England
is this: who were the limited inltrations, the small number of people
who introduced All Cannings Cross pottery to southern England from
eastern France or western Germany, who perhaps became the midden
builders of Wiltshire, and who may have been responsible for the construction of the early hillforts? In the next chapter we will attempt to
answer this question and reveal the identity not only of the MoundPeople and the Children of Don, but also of those shadowy gures who
haunt the pages of The Spoils of Annwn and the Mabinogion, the Druids.
Chapter 6
Visitors from the East

Brutus in Devon
Did a small number of people from eastern France or western Germany settle in Wiltshire around 800 B.C. or earlier, bringing with them
the All Cannings Cross pottery? There are various ways we can approach
this question, and the rst course Ill be pursuing is to look at the legendary
early history of Britain, using as my sources three works in Latin, the
Historia Brittonum of Nennius, Geoffrey of Monmouths Historia Regum
Britanniae, and the Itinerarium Cambriae (Itinerary Through Wales) of
Giraldus Cambrensis; and two works in medieval Welsh, Lludd and Llefelys
and the Welsh Triads.
In Chapter 10 of the Historia Brittonum, written around A.D. 830,
Nennius gives his version of the founding of Britain. In Homers Iliad one
of the leaders of the Dardanians of Troy is Aeneas, a principal lieutenant
of Hector, son of the Trojan king Priam. In the Aeneid, written by the
Roman poet Virgil around 25 B.C., Aeneas survives the destruction of Troy,
ees with his son Ascanius and eventually settles in Italy, where he is welcomed by Latinus, king of the Latins, and marries Lavinia, daughter of
Latinus, eventually becoming the ancestor of Romulus and Remus, who
founded Rome. Nennius writes1:
Ascanius married a wife, who conceived and became pregnant. And
Aeneas, having been informed that his daughter-in-law was pregnant,
ordered his son to send his magician to examine his wife, whether the
child conceived were male or female. The magician came and examined
the wife and pronounced it to be a son, who should become the most
valiant among the Italians, and the most beloved of all men. In consequence of this prediction, the magician was put to death by Ascanius; but
it happened that the mother of the child dying at its birth, he was named
Brutus; and after a certain interval agreeably to what the magician had
foretold, whilst he was playing with some others he shot his father with an
arrow, not intentionally but by accident. He was, for this cause, expelled
114
6 Visitors from the East
115
from Italy, and came to the islands of the Tyrrhene sea, when he was exiled
on account of the death of Turnus, slain by Aeneas. He then went among
the Gauls, and built the city of Turones, called Turnis. At length he came
to this island, named from him Britannia, dwelt there, and lled it with
his own descendants, and it has been inhabited from that time to the present period.
Geoffrey of Monmouth, in Historia Regum Britanniae, devotes the

whole of Book 1 to the story of Brutus. He tells us2 of the circumstances
in which Brutus and his men left Gaul and landed in Britain:
So without further delay, with the consent of his company, he repaired to
the eet, and loading it with the riches and spoils he had taken, set sail
with a fair wind towards the promised island, and arrived on the coast of
Totnes.
The island was then called Albion, and inhabited by none but a few
giants. Notwithstanding this, the pleasant situation of the places, the
plenty of rivers abounding with sh, and the engaging prospect of its
woods, made Brutus and his company very desirous to x their habitation
Map 6: Developed hillforts in southern England c. 100 B.C., showing the

modern counties of Wiltshire, Hampshire, Somerset and Dorset.
116

in it. They therefore passed through all the provinces, forced the giants to
y into the caves of the mountains, and divided the country among them
according to the directions of their commander. After this they began to
till the ground and build houses, so that in a little time the country looked
like a place that had been long inhabited. At last Brutus called the island
after his own name Britain, and his companions Britons; for by these
means he desired to perpetuate the memory of his name.
The most interesting detail supplied by Geoffrey is that Brutus and

his men landed on the coast of Totnes. Totnes today is a market town
at the head of the estuary of the River Dart in Devon, founded by the Saxons around A.D. 900. There are few signs of settlement activity before this
time. The earliest signs of occupation near Totnes are a Neolithic cooking
site at Hazard Hill some 2 miles to the west, excavated in the 1950s. Two
Bronze Age axes have been found between Totnes and Dartington, 2 miles
from Totnes. There are Bronze Age earthworks at Dartington North Wood,
a Scheduled Ancient Monument, where evidence of Iron Age and RomanoBritish settlement was also found. The nearest Roman capital was at Isca
Dumnoniorum (Exeter), but there is evidence of some form of Roman
settlement at Totnes.3
However, not far from Totnes is Dartmoor, with human activity
going back at least 12,000 years, as shown by analysis of ancient pollens
and charcoal and from the early styles of int tools still found today.4 As
the website of the Dartmoor National Park Authority also notes, more
concrete evidence of human activity can also be found in the chambered
tombs, burial chambers built of granite and covered by along earth mounds,
which may be up to 6000 years old. There are only a few of these; more
numerous are the cairns, or burial places (there are over 1500 of them),
which are mostly circular, and date to between 2500 B.C. and 1000 B.C.
Prehistoric Dartmoor obviously enjoyed a rich religious life, for it
also contains the largest number of ceremonial sites in northwest Europe,
including stone rows, which often terminate in a burial cairn, single standing stones, and stone circles. The people who built these cairns and stone
rows lived in stone-built round houses; over 5000 of these have been
found, with occupation dating from around 1500 B.C. Like the inhabitants
of Hampshire and Wiltshire, those living on Dartmoor devised their own
rectangular eld system, with low stony earth-covered boundaries called
reaves.
By 1000 B.C. Dartmoors climate had turned wetter and cooler (as
happened on and around Salisbury Plain) with its once fertile soil becom-
117
ing depleted and replaced by a buildup of peat. This led to the slow abandonment of settlement on the higher slopes with evidence of continued
occupation limited to a small number of sites.5
In Roman times Devon was occupied by the Dumnonii, and after
the end of the Roman period, the kingdom of Dumnonia survived in
Devon until around A.D. 800. So it is possible that Geoffrey was drawing
on old Dumnonian traditions about the former inhabitants of Dartmoor.
Interestingly, Geoffrey says that Brutus and his company forced the giants
living there to y into the caves of the mountains if by caves he means
burial mounds, then these giants are the equivalent of the Mound-People
living in their sidi or gorsedd.
Among the most prominent geological features of Dartmoor are the
tors, rock outcrops formed by weathering, usually found on or near the
summit of a hill. The word tor is one of the few Celtic words that survives
in English, but its origin is unclear, though it is thought to be cognate
with Latin turris tower. Curiously, the Latin turris is apparently unrelated
to any other Indo-European word except for the Greek tursis. Equally curious is the fact, rst noted by the German linguist Paul Kretschmer (see
Wikipedia article Tower), that both turris and tursis appear to be connected with place-names from Lydia, an Anatolian kingdom which developed after the collapse of the Hittite Empire, and with the people called
in Greek Tyrsenoi or Tyrrhenioi (Tyrsenians, Tyrrhenians), as well as with
Tusci (from *Turs-ci), the Greek and Latin names for the Etruscans.
But what is the connection between the Lydians and the Etruscans?
The Greek historian Herodotus, writing in about 430 B.C., says6 that as a
result of a prolonged famine, the Lydians decided to split in two, with half
staying in Lydia and the other half setting off for other lands, nally ending
up in Italy, where they became the Tyrrhenians (that is, Etruscans) (see
Map 7). The Dutch Professor of Comparative Indo-European Linguistics,
Robert Beekes,7 believes that at the time of the Trojan War, the Lydians
lived further north in Anatolia, in Maionia, south of the Sea of Marmara,
and were later pushed south by the Phrygians, who came from the southern
Balkans. Maionia borders on Troas, the region where Troy was located.
Beekes believes that there were two cities in the Troas, one called Wilusa
by the Hittites, which was inhabited by people speaking an Anatolian
(Indo-European) language, and one called Taruisa by the Hittites, where
the people spoke a nonIndo-European language. After the Trojan war,
under pressure from the Phrygians and the famine that resulted from the
global cooling we spoke of in Chapter 5, the people of Taruisa migrated
118
to Italy, and their name Tyrsenoi is derived from Taruisa, while their
brothers in Maionia migrated south into the area of present-day Manisa
and the interior of Izmir province to become the Lydians.
Does this mean that tor and turris are Anatolian words? Not necessarily. The eminent scholar of Iranian languages, H.W. Bailey,8 believes
that turris may be derived from one of the Iranian languages, which were
spoken north of the Black Sea and possibly as far west as the Danube by
peoples known as the Scythians, Sarmatians and Cimmerians. Bailey is
discussing Khotanese, an Iranian language spoken in northwest China,
and the word tturaka. He says9 that the word means cover of a quiver,
and he links it to Ossetic tura (where represents a voiced velar fricative
similar to the nal sound in Scottish loch) meaning court, forecourt,
vestibule, balcony. He links both words to Sanskrit torana arch, gateway,
and Khotanese ttora top, and notes that the base Indo-Iranian tav-: tuhas a wide range of meanings: it would seem to have meant place over,
upon, or around. Finally he notes that Ossetic tura as an enclosed place
suggests a connection with Greek tursis, Lat. turris tower (Ossetic is an
Iranian language spoken in North and South Ossetia, which border on the
Republic of Georgia).
Beekess contention that the Etruscans migrated from Taruisa in the
Troas region of northwest Anatolia, and Baileys hypothesis that the Latin
turris is derived from an ancient Iranian word used by tribes north of the
Black Sea, adds further weight to the theory, put forward in Chapter 5,
that the Bronze Age Collapse coincided with mass population movements
in the Mediterranean. It also casts new light on the legend of Brutus of
Troy, raising the possibility that he gave his name not just to Britain but,
more prosaically, to the tors of Dartmoor.
Of course, it is unlikely that the name Britain is derived from a man
called Brutus. The name was rst used by the Greek philosopher Aristotle
in the 4th century B.C., when the people were referred to as Pritteni. As
we implied in Chapter 4, the name Britain is usually derived from the
Indo-European root kwer-,10 which gives Old Irish cruth shape, Welsh
pryd shape, time, and perhaps Old Irish Cru(i)thin, Middle Welsh Prydyn,
Picts, Welsh Prydain Britain, making the British the tattooed people,
and prompting Baudis to suggest that the Mabinogions Pryderi is in effect
Mr. Britain. But Pokorny himself has cast doubt on this etymology. In
a discussion of the etymology of the name of the Swiss town Prttigau, in
Rheto-Romance Val Partns, in older times Pertennis, Pokorny11 links Pritteni to an Illyrian name, from the Indo-European root prteno-, related to
119
Old Indian prt- Kampf (ght, struggle) and Kampfer (ghter, warrior). This etymology apparently refers to Pokornys entry12 for the root
per- to hit, which gives Old Indian prt-/prtana ght, struggle, Avestan
peret- ght, struggle, battle, and Latin premo push, press. Both prt
(ght, battle) and prtana (battle, contest, strife) are listed in Sanskrit
dictionaries, and Old Persian partara battle occurs in a text from the
tomb of Darius (486 B.C.).13 As was the case with tor, we are getting more
hints of a long-distance connection between Britain and central or eastern
Europe.
Elidorus in the Underworld

Giraldus Cambrensis (A.D. 1146c. A.D. 1223) was the son of a powerful Anglo-Norman baron, Guillaume de Barri, the grandson of a Welsh
princess, and a maternal nephew of David FitzGerald, the Bishop of St.
Davids. In A.D. 1184 he became a royal clerk and chaplain to King Henry
II of England. He was chosen to accompany one of the kings sons, John,
on an expedition to Ireland, and published a report of what he saw there
in Topographia Hibernica (Topography of Ireland) (1188). He was then
selected to accompany the Archbishop of Canterbury, Baldwin of Exeter,
on a tour of Wales, the object being a recruitment campaign for the Third
Crusade. His account of that journey, the Itinerarium Cambriae (Itinerary
through Wales) (1191) was followed by the Descriptio Cambriae (Description
of Wales) in 1194.
In the Itinerarium Cambriae Giraldus was traveling in the Gower
Peninsula in Glamorgan, and spent the night at Swansea castle. He then
tells a very curious story14:
A short time before our days, a circumstance worthy of note occurred in
these parts, which Elidorus, a priest, most strenuously afrmed had
befallen himself. When a youth of twelve years, and learning his letters,
... in order to avoid the discipline and frequent stripes inicted on him by
his preceptor, he ran away, and concealed himself under the hollow bank
of a river. After fasting in that situation for two days, two little men of
pigmy stature appeared to him, saying, If you will come with us, we will
lead you into a country full of delights and sports. Assenting and rising
up, he followed his guides through a path, at rst subterraneous and dark,
into a most beautiful country, adorned with rivers and meadows, woods
and plains, but obscure, and not illuminated with the full light of the sun.
All the days were cloudy, and the nights extremely dark, on account of the
120

absence of the moon and stars. The boy was brought before the king, and
introduced to him in the presence of the court; who, having examined him
for a long time, delivered him to his son, who was then a boy. These men
were of the smallest stature, but very well proportioned in their make; they
were all of a fair complexion, with luxuriant hair falling over their shoulders
like that of women. They had horses and greyhounds adapted to their size.
They neither ate esh nor sh, but lived on milk diet, made up into messes
with saffron. They never took an oath, for they detested nothing so much
as lies. As often as they returned from our upper hemisphere, they reprobated our ambition, indelities, and inconstancies; they had no form of
public worship, being strict lovers and reverers, as it seemed, of truth.
Elidorus stayed with these people for some time, frequently returning to
our hemisphere to see his mother. His mother asked him to bring a present of gold, with which that region abounded, so he stole, while playing
with the kings son, the golden ball with which he used to divert himself.
He then hurried to his mothers house, pursued by two pigmies, who
took the ball from him and departed, showing the boy every mark of contempt and derision. When the boy tried to return to the underground
world, he found no sign of the passage that led there, though he searched
for it on the banks of the river for nearly a year. Eventually Elidorus became
a priest, but whenever David II, bishop of St. Davids, talked to him in
his later years of these events, he could never speak of his experience without shedding tears. And, says Giraldus:
He had made himself acquainted with the language of that nation, the
words of which, in his younger days, he used to recite, which, as the
bishop often had informed me, were very conformable to the Greek idiom.
When they asked for water, they said Ydor ydorum, which meant bring
water, for Ydor in their language, as well as in the Greek, signies water...;
and Dur also; in the British language, signies water. When they wanted
salt they said, Halgein ydorum, bring salt: salt is called hal in Greek, and
Halen in British, for that language, from the length of time which the
Britons (then called Trojans, and afterwards Britons, from Brito, their
leader) remained in Greece after the destruction of Troy, became, in many
instances, similar to the Greek.
The story of Elidorus is obviously drawn from the same legend or

legends that inspired Nennius and Geoffrey of Monmouth, except that
Geoffreys giants eeing to the mountain caves have been replaced by pigmies living underground. What is new in Giraldus is the information
that the pigmies spoke a language very conformable to the Greek
idiom a folk-memory, perhaps, of an Indo-European language spoken
in Britain before the Celts came.
121
The Three Oppressions of Britain

The story of Lludd and Llefelys is one of the so-called native tales of
the Mabinogion, and is thought to have been rst written down between
A.D. 1225 and A.D. 1250, though it may date to the 11th century. Lludd is
Nodens or Nodons, the god of the Roman temple of Lydney Park in
Gloucestershire, and the rst element of Llefelys is Lugus, as in Lleu of
the Fourth Branch of the Mabinogion.15 In this tale, Lludd is the son of
Beli, and king of the Island of Britain, and Llefelys his brother is king of
France, having married the daughter of the former king. During Lludds
reign, three gormesoedd, variously translated as plagues or oppressions,
fall on the Island of Britain16:
The rst was a certain race that came, and was called the Coranians; and
so great was their knowledge, that there was no discourse upon the face of
the Island, however low it might be spoken, but what, if the wind met it,
it was known to them. And through this they could not be injured.
The second plague was a shriek which came on every May-eve, over
every hearth in the Island of Britain. And this went through peoples
hearts, and so seared them, that the men lost their hue and their strength,
and the women their children, and the young men and the maidens lost
their senses, and all the animals and trees and the earth and the waters,
were left barren.
The third plague was, that however much of provisions and food might
be prepared in the kings courts, were there even so much as a years provision of meat and drink, none of it could ever be found, except what was
consumed in the rst night.
Not knowing what to do about the Coranians, or what might be causing

the other two plagues, Lludd consulted his brother Llefelys, the two speaking through a brass horn so the Coranians could not hear them. Llefelys
told Lludd that to overcome the Coranians, he would give him some
insects, which Lludd should mash in water and throw over the Coranians.
And the second plague, said he, that is in thy dominion, behold it is a
dragon. And another dragon of a foreign race is ghting with it, and striving to overcome it. And therefore does your dragon make a fearful outcry.... The cause of the third plague, said he, is a mighty man of magic,
who takes thy meat and thy drink and thy store. And he through illusions
and charms causes every one to sleep.
In order to overcome these two plagues, Llefelys made the following suggestions. In the case of the dragons, he should measure the island, and at
122
Map 7: The Classical World at and after the time of the Trojan War, showing
Lydian and Etruscan migrations from the area around Troy.
its exact central point, dig a large pit, and put a cauldron of mead in the
pit covered with satin. Eventually the ghting dragons would fall into the
cauldron, drink the mead and go to sleep, whereupon they could be buried
in a strong place. In the case of the food, Lludd should watch over it, with
a cauldron of cold water by his side to stop himself from falling asleep.
Lludd followed this advice : he mashed the insects in water and
destroyed all the Coranians; and he trapped the ghting dragons in the
cauldron of mead, and buried them at Dinas Emrys in northwest Wales.
He then prepared a large banquet, and with a cauldron of water at his
side, he waited. After a time
he heard many surpassing fascinations and various songs. And drowsiness
urged him to sleep. Upon this, lest he should be hindered from his purpose and be overcome by sleep, he went often into the water. And at last,
behold, a man of vast size, clad in strong, heavy armour, came in, bearing
a hamper. And, as he was wont, he put all the food and provisions of meat
and drink into the hamper, and proceeded to go with it forth. And nothing was ever more wonderful to Lludd, than that the hamper should hold
so much.
And thereupon King Lludd went after him and spoke unto him thus.
Stop, stop, said he, though thou hast done many insults and much spoil
123
erewhile, thou shalt not do so any more, unless thy skill in arms and thy
prowess be greater than mine.
Then he instantly put down the hamper on the oor, and awaited him.
And a erce encounter was between them, so that the glittering re ew
out from their arms. And at the last Lludd grappled with him, and fate
bestowed the victory on Lludd. And he threw the plague to the earth.
The man of vast size then promised to atone for all the wrongs he had
done to Lludd, and to become his faithful vassal.
Before I comment on this tale, I would like rst to present one of the
Welsh Triads Triad 36,17 which is relevant to Lludd and Llefelys:
Triad 36
Three oppressions that came to this Island, and not one of them went
back:
One of them (was) the people of the Coraniaid, who came here in the
time of Caswallawn (= Lludd?) son of Beli: and not one of them went
back. And they came from Arabia.
The second Oppression: the Gwyddyl Fchti. And not one of them
went back.
The third Oppression: the Saxons, with Horsa and Hengist as their
leaders.
Triad 36 speaks of the Saxons as one of the oppressions, and these can
readily be linked to the ghting dragons. In the Historia Brittonum,18 Vortigern decides to build a citadel, and collects together building materials,
but these keep disappearing overnight. So he calls together his wise men,
who tell him: You must nd a child born without a father, put him to
death, and sprinkle with his blood the ground on which the citadel is to
be built, or you will never accomplish your purpose. So Vortigern sends
out messengers to nd a child without a father, who nd one and bring
him before the king. When the king explains to the boy what his wise men
told him, the boy reveals to them that under the ground where the citadel
is to be built there is a pool, in the pool are two vases, in the vases is a
folded tent, and in the tent are two sleeping serpents who begin to struggle
with each other:
The white one, raising himself up, threw down the other into the middle
of the tent and sometimes drove him to the edge of it; and this was
repeated thrice. At length the red one, apparently the weaker of the two,
recovering his strength, expelled the white one from the tent; and the latter being pursued through the pool by the red one, disappeared. Then the
boy, asking the wise men what was signied by this wonderful omen, and
124

they expressing their ignorance, he said to the king, I will now unfold to
you the meaning of this mystery. The pool is the emblem of this world,
and the tent that of your kingdom: the two serpents are two dragons; the
red serpent is your dragon, but the white serpent is the dragon of the people who occupy several provinces and districts of Britain, even almost from
sea to sea: at length, however, our people shall rise and drive away the
Saxon race from beyond the sea, whence they originally came.
One of the three oppressions the ghting dragons therefore refers to

the Saxons, who certainly came and did not go back. This implies that
the man of vast size who stole all the food can be identied with the Gwyddyl Fchti, or Gaelic Picts, the name used by the medieval Welsh to designate the Scottish Picts. But who were the Picts?
The fact that the food thief appears to be a giant and uses music to
cast a spell over people suggests that the Picts belongs to mythological
times. This combination of food, music and thievery calls to mind a passage
from the Deipnosophists (c. A.D. 230) of the Greek writer Athenaeus, where
he says, quoting Poseidonius (135 B.C .51 B.C .)19:
Poseidonius of Apameia, in the twenty-third book of his histories, says,
The Celts, even when they make war, take about with them companions
to dine with them, whom they call parasites. And these men celebrate
their praises before large companies assembled together, and also to private
individuals who are willing to listen to them; they have also a description
of people called Bards, who make them music; and these are poets, who
recite their praises with songs.
The link to a mythological past is reinforced by the magic hamper carried

by the large man, which resembles the magic bag that Rhiannon gave to
Pwyll in the First Branch of the Mabinogion, and the cornucopia of the
goddess of Gloucestershire.
The Picts in some sense do belong to mythological time. They were
a confederation of tribes living in the north and east of Scotland from
before the Roman conquest of Britain until the 10th century. The AngloSaxon historian Bede, writing in the 8th century A.D., tells us20 that Britain
was home to ve distinct languages: Church Latin, English, Welsh, Gaelic
and Pictish. Bede also relates a legend about the Picts. He says they were
descended from a colony of seafarers from Scythia who took possession of
northern Britain. The 15th-century Scottish history, the Scotichronicon, on
the other hand, says that the Picts came from the coasts of the Baltic Sea.21
Is there any substance to the legend that the Picts were descended
from Scythians? Dunbavin says22 that W.J. Watson dismissed the legends
as learned embellishment, based on the mention of picti Agathyrsi
125
(painted Agathyrsi) and picti Geloni (painted Geloni) in the Aeneid of

Virgil. Herodotus, says Dunbavin,23
locates the Agathyrsi on the Black Sea above the Danube, and the Geloni,
he says, were Greeks who settled on the upper Dnieper in present-day
Ukraine. All these tribes were, at that time, tributary to the Iranian-speaking Scythians. Herodotus also describes the social organization of the
Agathyrsi: they held their women in common and were all therefore brothers, as of a single family. This is remarkably similar to the customs attributed to the British tribes by Julius Caesar, Dio Cassius, Solinus and St.
Jerome.
According to Dunbavin,24 the 4th-century Roman writer Solinus

records that the Agathyrsi dyed their hair and faces blue, just as we are
told that the Britons did. Indeed, the practice of tattooing and body
painting was not restricted to the British Isles. The tattooing of animal
gures on the skin was apparently a widespread custom throughout the
Scythian region, as such depictions have been found on the body of a
Scythian chieftain excavated at Pazyryrk, in the Altai region of Russia.
Interestingly, the Roman historian Pliny implies that the Scythians lived
as far north as the Baltic coast,25 which the Scotichronicon gave as the homeland of the Picts.
Of the three oppressions, the most obscure is that of the Coranians,
or Coraniaid as they are more generally known. Their name is generally
linked to Welsh cor(r) dwarf 26; Koch also suggests a connection with
Old Irish corrguinecht magic, wizardry, and corrguinech magician, sorcerer. The Coraniaid are said to have the ability to hear any utterance
that met the wind an ability shared by Math in the Fourth Branch of
the Mabinogion. Asdis R. Magnusdottir27 links the Coraniaid to the Norse
guardian of the gods, Heimdallur, who can hear grass growing, and possesses a horn which can be heard throughout the world.
The method used to destroy the Coraniaid is puzzling. The insects
that Llefelys recommends be crushed in water and sprinkled over the Coraniaid apparently are not found in Britain, and could be a reference to
Spanish y, which is found only in southern Europe, including France.
Spanish y is a species of blister beetle which secretes a poison called cantharidin cantharidin can cause blisters, and if accidentally ingested by
horses can be toxic. The fact that the crushed insects can kill the Coraniaid
but cant harm Lludds people suggests that there may be a close association
between the Coraniaid and horses which, in magical terms, makes them
especially vulnerable to cantharidin.
126
Perhaps most puzzling of all is that the Coraniaid are said to come
from Arabia. In the Third Branch of the Mabinogion (see Chapter 4),
Manawydan, when he became a saddle-maker, made pommels colored
with blue azure, as he had seen Llassar do. In medieval times, blue pigment was made from lapis lazuli, which originally came from Afghanistan the word lazuli comes from the Arabic lazaward, which ultimately
comes from the Persian lazhvard, the name of a place where lapis lazuli
was mined. This word is also the origin of the color azure, and may also
be the origin of the name Llassar. It seems likely therefore that the Coraniaid were thought to come from the same place as lapis lazuli, here glossed
as Arabia, but more plausibly signifying the vast expanse of eastern Europe
where Scythians and other Iranian-speaking tribes lived. It was Llassar,
after all, who escaped from the Iron House in Ireland and brought Bran
the cauldron of regeneration and as was pointed out in Chapter 5, the
technology for making cauldrons of beaten bronze probably originated in
central Europe.
Thracians and Cimmerians

The legendary history of Britain speaks of Trojans, Greeks, Scythians
and people from the Baltic, and there are tantalizing hints that there is
indeed some connection between Britain and lands far to the east. To make
sense of this, we need to look at a culture usually referred to as Thraco
Cimmerian, and associated with two ethnic groups, the Thracians and the
Cimmerians. For Homer (Odyssey 1.14), the Cimmerians were a people
living in a mythical land of fog and darkness on the fringes of the habitable
world28; while the Thracians, a more esh-and-blood people, were allies
of the Trojans in the Iliad.
The Thracians were an Indo-European people speaking a language
that is now extinct, and centered on present-day Bulgaria, the European
part of Turkey, and parts of Romania, Serbia, Greece and the Republic of
Macedonia. The Thracians were neighbors of the Greeks and gure prominently in the historical record, unlike the Cimmerians. Herodotus29 says
that the Cimmerians originally inhabited the region north of the Caucasus
and the Black Sea, in what is now Ukraine and Russia, but were driven
out by the Scythians. The Assyrian annals of 714 B.C. report that a people
called the Gimirri helped the forces of Sargon II, king of Assyria, defeat
the kingdom of Urartu, in the Lake Van area of Anatolia; at the time, the
127
Gimirri seem to have lived in northwest Iran, but later settled in what is
now the Republic of Georgia.
The concept of a ThracoCimmerian culture was originally proposed
by Paul Reinecke in 1925, and reinforced by researchers in the 1930s.30
These researchers saw the presence of equestrian bronzes and weapons in
central and southeastern European graves and hoards that showed East
European steppe-bound analogies as a reection of the invasion of mounted
warriors. These putative invaders were identied with the historic Cimmerians who were thought to have caused major disruptions in the cultural
development of the Carpathian basin and beyond into Central Europe,
superimposing themselves on the prevailing Urneld culture. They were
linked to the Thracians because parts of the Black Sea coast were under
Thracian inuence at the time, the Thracians were rich in mineral
resources, especially iron ores, and they were renowned for their horses
and mounted warriors.31
Where the Cimmerians originated and where they lived still remains
a matter of debate. Kristian Kristiansen notes32 that from about 900 B.C.,
rich, well-organized kingdoms or chiefdoms developed in the Caucasus,
interacting with the civilizations to the south. Here, he says, we also nd
typical horse-bits and cheek-pieces of early ThracoCimmerian type.
During the 9th and 8th centuries, says Kristiansen, a grouping or complex
of related cultural traits ... were expanding into Central Europe which
can be identied as ThracoCimmerian. Among the artefacts belonging
to this complex, says Bouzek, are horse-bits and Cimmerian bimetallic
daggers, with bronze handles and iron blades, which are found as far west
as Silesia (now part of Poland), Moravia (now part of Czechoslovakia),
Leibnitz (Austria), Neundorf (in Lausitz, in the eastern German state of
Brandenburg)33 (see Map 8).
Horses and wagons were an important feature of the ThracoCimmerian culture, which brought with it some important technological innovations. Kristiansen points out34 that ThracoCimmerian horse-bits were
of the two-joint type, meant for riding, as opposed to the rigid Urneld
bits, which were more suitable for traction, as they are hard on the mouth
of the horse. Second, compared to Urneld bits with a diameter normally
of 3 inches, the new types are 4 to 4 and a half inches, implying larger
breeds of horses. Third, the wagons and wheels demonstrate a newer and
more complex technology. From this, says Kristiansen, we may conclude
that riding had now become fully developed in warfare, demanding the
specialized breeding and training of horses.
128
The process of expansion says Kristiansen,35 was highly complex,

trade alternating with elite conquest. In Hungary, pastoral overlords
brought with them a new burial ritual and a completely new material culture, while in Slovenia there was more a mixture of foreign and local traditions. A new princely social organization emerged in the 8th century,
characterized by large fortied settlements and tumulus burials, as at Sticna
(Slovenia), denburg/Sopron (in Hungary, on the Austrian border) and
Novo Mesto (Slovenia). Large tumulus barrows, another aspect of the new
eastern ideology, were raised outside new royal courts at Sticna and Novo
Mesto ... to signal the presence of new elites, as at Seddin (in Branbenburg)
and Voldtofte (in Denmark) in the north during the same period.36
It seems therefore that the ThracoCimmerians were horse-riding
nomadic warriors who spread westward from the steppes of present-day
Ukraine and Russia, bringing with them new types of weapons and horseriding technologies. The inuence of ThracoCimmerian culture seems
to have been strongest in Hungary, but also extended as far west as Austria
and even Switzerland a ThracoCimmerian bronze horse-bit has been
found at the settlement of Zrich-Alpenquai on the shores of Lake
Zrich.37 Obviously the ThracoCimmerian culture did not spread as far
as Britain, but it may well have inuenced Britain indirectly. The All Cannings Cross pottery discussed in Chapter 5 may well have been introduced
by a small number of settlers from eastern France. Nancy K. Sandars says38
that this pottery comes from the Jura region of France, which borders on
Switzerland, not far from the westernmost spread of ThracoCimmerian
bronzes.
It was noted in Chapter 5 that the introduction of the All Cannings
Cross pottery was accompanied by the construction of large-scale middens
in places like Potterne and East Chisenbury. But these were not the only
changes that occurred. The period between 900 B.C. and 600 B.C. saw the
emergence of the rst hillforts large hilltop enclosures and small strongly
defended forts.39 The hilltop enclosures include Danebury, site of a later
hillfort, and Balksbury, both in west Hampshire. Balksbury has been
extensively excavated, and work has shown that
the enclosure bank and ditch was built in the 9th8th centuries B.C., and
the enclosure continued in use for about two centuries, during which time
the bank and ditch was refurbished on at least two occasions. Internally
the only signicant features of this phase to be identied were a number of
small four-post granaries and a few lightly-built circular huts. The
most interesting aspect of the recent work has been the examination of the
129
build-up of colluvium, containing midden material, against the inside of

the enclosing bank. Analysis suggested that the high organic component of
the deposit probably derived from animal waste and other organic material
brought in for fodder and litter.40
The evidence suggests that

the primary function of these hilltop enclosures was pastoral, to provide
corral space for livestock at certain times during the year. In this context
the four-post structures could be interpreted as fodder ricks, while the
light circular buildings could have provided shelter for those tending the
beasts. The size of the enclosures might suggest that they served large communities, and this raises the possibility that they were places where the
community could gather at certain times during the year for ceremonies
and feasting, when the more practical tasks of culling, castration and the
redistribution of stock were being undertaken. Some supporting evidence
for this comes from Balksbury, where it was found that the colluvium contained midden material possibly derived from feasting. As to the size of the
territory to which the enclosure belonged, it may be relevant to note that
some of the pottery found in these deposits came from as far away as 610
miles.41
The small strongly defended forts include Budbury (west Wiltshire);

Lidbury Camp, near Enford, Wiltshire, not far from the midden at East
Chisenbury; and Olivers Camp, near Devizes, Wiltshire, in the general
area of All Cannings Cross and the midden at Potterne (see Map 9). The
sites are of less than 7 acres in extent, they favour ridge-end locations
(although some are found in less defensible and more open central downland settings) and often have more than one line of defence. All seem to
have been intensively occupied with the exception of Olivers Camp, which
produced comparatively little material and no major internal structures.42
Cunliffe suggests that they might have been elite settlements of some
kind, the prominent location and impressive defences being the symbols
of elite status distinguishing them from contemporary farmsteads. 43
According to Cunliffe, the overall impression gained from this and other
evidence is that this was a period of transition in the course of which the
economic, social and belief systems changed rapidly.44
Before I leave the Cimmerians, I need to say something about legends
that link the Cimmerians to the Germanic tribe known as the Cimbri,
and to the Cymry, or Welsh. The Cimbri lived in Jutland, Demark, in the
area known as Himmerland, where the Gundsetrup Cauldron was found,
and the classical writers Strabo, Diodorus Siculus and Plutarch all link
them to the Cimmerians. The etymology of Cimmerian and Cimbri is
130
unknown. One possibility is that Cimbri is from the Indo-European root

kei to lie down,45 which gives Greek keitai lies, kome village, Hittite
kihthta (kitta) and ki-it-ta-ri (kittari) lies, Latin civis townsman, Gothic
haims village, Old Norse heimr homeland, world, Old High German
heim homeland, house, dwelling. Other scholars46 have seen the root as
tkei settle, found, and also derived from it Sanskrit kseti dwell. Thus
Cimmerian and Cimbri would mean settlers, villagers.
The other possible etymology of Cimmerian is linked to the etymology of the Welsh Cymry. This is said to be derived from the Celtic words
com plus brog; brog is cognate with Latin margo frontier, margin, Gothic
marka frontier, Old High German marcha limit, boundary, Old
English mearc boundary, Armenian marz border, land, Ukrainian meza
border, margin, Persian marz border, land47; com is cognate with Greek
koinos shared, common. The earliest mention of the Persian word marz
is during the Persian Sassanid Empire (A.D. 205A.D. 651), when there
was a position called marzban, that is, governor of a frontier province.
However, given the fact that Sassanid Persian is not likely to have been
inuenced by Celtic or Latin, it is probable that the Middle Persian and
Armenian words marz either developed independently, or borrowed from
each other. Therefore, it would seem that the name Cimmerian developed
from Indo-European words similar to kom and marz, and meant those
sharing a common land. So even if the Cimmerians, Cimbri and Cymry
are not related, at least two of the names have a similar Indo-European
derivation.
The Thracian Horseman

What Cunliffe says about Wessex between 900 B.C. and 700 B.C. suggests that the people who brought the All Cannings Cross pottery probably
lived in small strongly defended forts at Lidbury Camp and Olivers Camp,
and kept their stock and conducted communal feasts in hilltop enclosures
at Danebury and Balksbury. If Cunliffes interpretation is to be believed,
they worshipped chthonic deities fertility gods by constructing huge
middens at East Chisenbury and Potterne, where all that the earth produced was returned to the earth. If I understand Cunliffe correctly, it was
these same people who, from around 600 B.C., built the hillfort at Danebury, on the site of the earlier hilltop enclosure, digging out pits in which
grain was stored, placing the grain under the protection of these same
131
chthonic deities, and later making sacrices to the gods in these pits,
including the ritually dismembered horse referred to in Chapter 5.
Continuity seems to have been maintained at Danebury until the end
of the 4th century, when there is evidence of a widespread re followed
by diminished use.48 Cunliffe believes that this is evidence of social disruption: it coincides with a major change in pottery style recognizable
over a considerable area, perhaps indicating a social dislocation of more
than regional signicance. After this distinctive horizon many of the early
hillforts show no sign of any further use, while others continued to be utilized, their enclosing earthworks being refurbished. The simplest explanation of the phenomenon, says Cunliffe,49 is that there was a widespread
social crisis brought about perhaps by the emergence of competing polities.
Once it was resolved some of the old polities, who had maintained their
integrity and dominance, continued while others were disbanded or
absorbed. This could explain the abandonment of some of the hillforts
and the development of others. Although Cunliffe does not say this, one
factor here could be the emergence of Celtic elites in the Thames Valley
and eastern England, where pottery styles were strongly inuenced by
Continental La Tne cultures from northern France and the Low Countries, and conict could easily have arisen in the Berkshire Downs and the
Cotswolds, at the frontier between two groups of people represented by
the more conservative saucepan pot continuum of the south of England50
and the innovative La Tne pottery north of the Thames.
All this points to the fact that the culture which predominated in
southern England until around 400 B.C. or later was not Celtic and had
probably evolved from the culture which made the All Cannings Cross
pottery and accumulated the massive middens. We have two pointers to
the nature of this culture: it appears to have worshipped chthonic deities,
and it appears to have included horses in its religious ritual. Lets start
with the horses, which clearly played an important part in the life of the
Thracians and Cimmerians who may have indirectly inuenced the makers
of the All Cannings Cross pottery.
We know nothing of the religion of the Cimmerians, but a good deal
about Thracian religious beliefs. Horse-gods have a long history among
Indo-European peoples: Linear B tablets from Pylos in southern Greece
(see Map 10), dated to around 1200 B.C ., mention a po-ti-ni-ya i-que-ya,
or Potnia Ikkweia, Mistress of Horses51; and the ancient Indian religious
text the Rigveda refers many times to the divine twin horsemen known as
the Ashvins. But in Thracian religion or mythology the rst mention of
132
Map 8: Furthest extent of ThracoCimmerian migrations westward early in

the 1st millennium B.C.
a gure connected with horses is found in the Iliad of Homer, thought to

have been composed around 800 B.C. In Book 10, Dolon, an ally of the
Trojans, says to Odysseus52:
If youre keen to inltrate
the Trojan army, over there are Thracians.
fresh troops, new arrivals, furthest distant
from the rest, among them their king Rhesus,
son of Eioneus. His horses are the best.
the nest and largest ones Ive ever seen,
whiter than snow, as fast as the winds.
His chariot is nely built with gold
and silver. He came here with his armour
an amazing sight huge and made of gold.
Its not appropriate for mortal men
to wear such armour, only deathless gods.
Homer implies that Rhesus was a god or at least god-like gure, and this
is borne out by the later cult, attested from the 4th century B.C., of the
rider-god called the Thracian Horseman, also known as Heros (Hero).
Robert Turcan53 notes that this being is depicted on stelae found in all
regions of Thrace, including Odessos and Romula (see Map 11). In these
depictions, mostly from the time of the Roman Empire, the Thracian
horseman advances, sometimes at a walk (see Figure 10), sometimes at a
gallop, towards an altar erected in front of a tree around which a snake
133
is entwined54 (the walking pace recalls the horse of Rhiannon when Pwyll
sees her for the rst time). We also see him leaping forward with a spear
in his hand to hunt boar, or astride a lion. Sometimes he holds game a
doe or a hare, for instance which he has ushed out and killed. A dog
usually accompanies him55 (as in the ritual depositions at Danebury).
Raffaele Pettazzoni points out56 that that on at least ve stelae (from
Philippopolis and Oiskos) the Thracian Horseman, or Thracian Rider, as
Pettazzoni calls him, has three heads, and in one stela he has two heads.
One inscription with the Rider includes a Thracian word interpreted as
panthopto, that is, all-seeing, whose Greek equivalent panoptes is used to
describe Helios, the Sun. According to Pettazzoni, there are numerous
indications that the Thracians worshipped the sun: in a fragment of Sophocles tragedy Tereus, dating from the 5th century B.C., the sun is invoked
with the words: O Sun, whose splendour is most revered by the horseloving Thracians57; and according to a myth preserved in the astronomical
work the Catasterismi attributed to Eratosthenes, a Greek astronomer of
Map 9: Small forts and large hilltop enclosures in the vicinity of Salisbury
Plain around 800 B.C.
134
the 3rd century B.C., and probably derived from a lost play of the 5th century B.C. by Aeschylus, the Bassarids, Orpheus, originally a Thracian, used
to rise while it was still dark to climb before dawn to the summit of Mt.
Pangaion and salute the sun as it rose.
However, Herodotus58 says that the Thracians worship no gods but
Ares, Dionysus, and Artemis. Their princes, however, unlike the rest of
their countrymen, worship Hermes above all gods and swear only by him,
claiming him for their ancestor. No mention is made of the Thracian
Horseman, and Pettazzoni tries to explain why. He believes that ordinary
Thracians worshipped the god that Herodotus calls Dionysus, while aristocrats worshipped the god interpreted as Hermes.59 A parallel can be
found, he says, among the Scythians, who, according to Herodotus, had
a certain number of national divinities, including an Apollo, while the
Royal Scythians, the clan to which the kings and lords belonged, venerated
another god.60 The opposition between the religion of Dionysus and the
religion of the Sun is underlined in the Bassarids of Aeschylus, where
Dionysus gets his revenge on the sun-worshipping Orpheus by having him
killed by his followers, the Bassarids.
It appears from this that the sun-god was interpreted as Hermes, but
why? Pettazzoni believes that an infernal aspect is common to both Hermes
as psychopompos (soul-guide) and to the Sun, as setting daily in the west
and vanishing beneath the horizon to lighten the underground world of
the dead during the night and reappear at sunrise in the east.61 This helps
to explain the Greek interpretation of the Persian god Mithra. In inscriptions from the tomb of Antiochos of Commagene (a kingdom in what is
now south-central Turkey), the sun-god Mithra is identied not only
with Helios and Apollo, but also with Hermes. Interestingly, Mithra
was invoked by the rulers of Persia to witness their oaths and asseverations,62 just as the Thracians swore all their oaths by Hermes. Pettazzoni
concludes that the Hermes of Herodotus is the Thracian Horseman,
sometimes described on inscriptions as genikos or geniakos, Progenitor of
the Thracians, and as all-seeing sun-god, well placed to ensure that all
oaths were kept by those who made them.63
Is there any connection between Rhesus of the Iliad, the cult of the
Thracian Horseman, and the dismembered horse found in the storage pit
at Danebury? I think there is, and I am not alone in thinking so. David
Rankin,64 following on from Henrich Wagners 1971 work, Studies in the
Origins of the Celts and of Early Celtic Civilization, links Rhesus and the
Thracian Horseman if not to Danebury and the dismembered horse, then
135
to the Irish god the Dagda, also known as Eochaid Ollathair (Horse GreatFather), and links the Thracian Hermes to Lugh in the Irish tradition.
Of course from the British and Welsh perspective, the closest match to
the Thracian Rider-god is Rhiannon, the horse goddess that both Pwyll
and Manawydan marry. The concept of marriage to a horse-goddess is
given extra piquancy by that indefatigable traveler, Giraldus Cambrensis,
in his Topography of Ireland. He is traveling in the northwest of Ulster, in
the area of Kenelcunill (Cenel Conaill), when he comes across a people
accustomed to appoint its king with a rite altogether outlandish and
abominable65:
When the whole people of that land has been gathered together in one
place, a white mare is brought forward into the middle of the assembly.
He who is to be inaugurated, not as a chief, but as a beast, not as a king,
but as an outlaw, has bestial intercourse with her before all, professing
himself to be a beast also. The mare is then killed immediately, cut up into
pieces, and boiled in water. A bath is prepared for the man afterwards in
the same water. He sits in the bath surrounded by all his people, and all,
he and they, eat of the meat of the mare which is brought to them. He
quaffs and drinks of the broth in which he is bathed, not in any cup, or
using his hand, but just dipping his mouth into it round about him. When
this unrighteous rite has been carried out, his kingship and dominion have
been conferred.
As Daniel Bray points out,66 this ritual of horse-sacrice has distinct parallels in the Vedic inauguration rite, the asvamedha, which
involved the sacricial slaughter of a stallion, whereupon the queen symbolically slept with it under covers, and afterwards the esh was cooked
and eaten by the assembled people. As this activity was carried out, hymns
were sung to praise the horse, to direct its dismemberment, and to petition
favourable returns, phrased as follows: Let this racehorse bring us good
cattle and good horses, male children and all-nourishing wealth. Let Aditi
make us free from sin. Let the horse with our offerings achieve sovereign
power for us.
The Druids of Dodona

We saw in Chapter 5 that the deposition of the dismembered horse
in the storage pit at Danebury was linked by Anne Ross to the heads and
hooves ritual of the bull-feast, rst described in the Irish epic tale The
Destruction of Da Dergas Hostel (c. A.D. 1100). Presumably the heads and
hooves ritual and the kingship ritual described by Giraldus were super-
136
Map 10: The Mycenean world around 1200 B.C., showing settlements in Crete,
Greece and Asia Minor (Anatolia).
vised by priests of some sort, and we must assume that the priests who
supervised these rituals in the 12th century A.D. were in a sense the same
priests who presided over the ritual deposition of the dismembered horse
at Danebury in the 6th century B.C. I will now argue that these priests
were Druids, and that this priesthood was born in the 8th century B.C.
around the middens and small strongly defended forts of Wiltshire.
The rst piece of evidence for this claim lies in the etymology of the
name. In fact the etymology of druid is disputed, but the most common
explanation is that it means something like oak-seer. As West points
out,67 this interpretation of the name links the Druids to the ancient oracle
of Dodona in Epirus, northwestern Greece, which, says West, is an Illyrian
137
rather than a Greek institution. Susan Guettel Cole says68 that the oracle
of Dodona, located in a distant valley deep in Epirus,
had the reputation of great antiquity. Priests called Selloi went barefoot
with feet unwashed. Zeus himself, possibly represented by the great oak
tree that grew in his sanctuary, was attended by three priestesses and
accompanied by doves. Ancient traditions differ in the precise details. Just
how responses were delivered is not divulged; some believe the gods will
was interpreted from the rustling of the oak leaves, others from the cries of
the doves. Enquiries were inscribed on small lead tablets, written by the
petitioners themselves.
Dodona is mentioned in the Iliad 69 when Achilles prays to Lord Zeus,

Dodonean, Pelasgian Zeus; you that live far away and rule over wintry
Dodona. Sotiris Dakaris, who excavated Dodona, dates the beginning
of the cult from the late earlyHelladic to early middleHelladic, hence
to about 2000 B.C. He also points to the close connection of the dove with
Figure 10: Image of the rider-god, also known as the Thracian Horseman,
found on the Lincolnshire/Nottinghamshire border, near the village of Brough
(Roman Crocolana). British Museum. Note the exaggerated walking pace of
the horse.
138
Cretan-Mycenean religion, in which the dove is venerated as a divine symbol and sacred animal,70 and believes that Zeus rst became established
at Dodona in the 13th century B.C.
This was not the only sacred oak in Greece. On Mount Lycaeus in
Arcadia, where Zeus is said to have been reared, there is a spring called
Hagno, named after one of the nymphs who reared Zeus. Pausanias tells
the following story about this spring71:
Should a drought persist for a long time, and the seeds in the earth and
the trees wither, then the priest of Lycaean Zeus, after praying towards the
water and making the usual sacrices, lowers an oak branch to the surface
of the spring, not letting it sink deep. When the water has been stirred up
there rises a vapor, like mist; after a time the mist becomes cloud, gathers
to itself other clouds, and makes rain fall on the land of the Arcadians.
Classical writers linked the Druids to bards and vates, and the etymology of these two words is instructive. The word bard is said to derive
from *g wer- to raise the voice, to praise, which also gives Avestan (aibi)
jaretay praiser and Sanskrit jaritar singer, praiser.72 On the other hand,
Raimo Anttila says73 that Enrico Campanile derives bard from Indo-European *gwer-dhe-o-s the one who puts up (makes) praise(s), which Campanile links to Sanskrit giro dha and Avestan garo da, Vedic giram dha and
Avestan garem da. These derivations are supported by other sources. In
the Sanskrit Lexicon, the meaning of gir is given as addressing, invoking,
praising; invocation, addressing with praise, praise, verse. According
to the Old Avestan Glossary, gar- means song of praise; to sing (songs of
praise); in Kangas Avesta Dictionary, gar or gar means invocation, and
in the Altiranisches Wrterbuch, garah means hymn, song of praise.
The word vates is said74 to be related to a set of Germanic words that
link the ideas of poetry and possession: Gothic woths possessed, Old
High German wuot frenzied, Old English wod song, Old Norse othr
possessed, inspired; mind, poetry, and to the Norse god Odin, and his
Anglo-Saxon counterpart Woden. N.D. Kazanas speculates75 that Woden/
Odin may be a development of an Indo-European deity appearing as Vedic
Vata ( = wind: a variant of the more common Vayu) who exhibits traits
pertinent to the Germanic god. Thus Vatas swiftness is a standard of comparison for swift motions; his wrath can be roused easily; he blows
down from heaven with rainstorms and roars in the sky thundering,
and has the treasure of immortality in his dwelling whereby he gives life
to his devotees: these traits are found also in Wodan. It may be recalled
that the Taliesin-poet drew on something he called awen (inspiration),
139
Map 11: Signicant sites in Arcadia and Macedonia (c. 700 B.C.), plus Greek
cities in Italy associated with Pythagoras, and the distribution of the cult of
the Thracian Horseman.
which is cognate with Welsh awel breeze obviously the equation breeze
= inspiration is very similar to the equation wind = prophetic fury.
The second piece of evidence that the priests of Wiltshire around 800
B.C. were Druids, or at least came from those distant lands where the oak
was venerated, is to be found in the other burials at Danebury. It was
noted earlier that beside the dismembered horse in the pit at Danebury
was the skeleton of a dog, and we have said little so far about the dog. J.P.
Mallory and Douglas Q. Adams conrm76 that the association of dogs with
death is a common Indo-European theme. Yama, the lord of death in the
ancient Vedas of India, has two four-eyed dogs, who are guardians and
keepers of the path to the afterworld. In Iranian tradition, a four-eyed
dog was brought before a dead body in order for its gaze to expel the
demons. The Daena or inner self of the dead person, in the shape of a
beautiful woman, escorts the souls of the righteous to Paradise accompanied by two dogs whose function is to guide the soul to the proper path
140
by barking. The three-headed Cerberus is the best-known other-worldly

dog, rst attested in the Theogony of the Greek poet Hesiod, written around
700 B.C. Green points out77 the ritual importance of dogs at Danebury.
In one of the pits constructed before 500 B.C. the bodies of two dogs were
found, together with a selection of twenty other bones representing a
range of wild and domestic species. After the animals had been positioned,
chalk blocks were laid over the bodies and then a huge timber structure
was erected over the middle of the whole deposit. Green adds that horses
and dog occur together repeatedly enough for their association to be considered statistically signicant at Danebury.
Another animal overrepresented at Danebury is the raven.78 In
Romano-Celtic iconography, says Green,79 ravens were associated with a
number of benecent deities. The god who presided over the Burgundian
healing spring shrine at Mavilly, in northeastern France, was depicted in
company with a large raven; the goddess Nantosuelta appears on images
at Sarrebourg, near Metz (northeastern France), and Speier (Germany),
accompanied by a raven. In Luxembourg a distinctive type of goddess
was worshipped, a deity whose image consisted of a lady seated within a
house-shaped shrine and accompanied by a raven. Interestingly, one representation of the horse-goddess Epona, at Altrier in Luxembourg, depicts
her with a dog and a raven.80
However, the raven or crow also has ancient connections with Greece
and other Mediterranean lands. In Fragment 89 of the Catalogue of Women,
attributed to Hesiod, the poet tells us that a crow came to Apollo to warn
him that his lover Coronis had wed Ischys. Herodotus also tells that Aristeas, the semi-legendary Greek poet of the 7th century B.C., who came
from Proconnesus (the Turkish island of Marmara), had traveled to the
Issedones (an ancient tribe of Central Asia), transported by Apollo. 81
Herodotus tells us that Aristeas reappeared 200 years later in Metapontum
in southern Italy, telling the people of Metapontus that he had accompanied
Apollo in the shape of a raven.
Metapontum is associated not only with Aristeas, but also with the
Greek philosopher Pythagoras (c. 570 B.C .c. 495 B.C .), who believed in
reincarnation and who has been linked to the Druids (see Chapter 1).
Pythagoras was born on the Greek island of Samos, but in later life moved
to Croton, in southern Italy, and subsequently to Metapontum. Pythagoras
was regarded by his followers as the reincarnation of Hyperborean
Apollo.82 Legend has it that Pythagoras met the legendary sage, healer
and priest of Apollo, Abaris the Hyperboreanwho, according to Herodotus,
141
traveled round the world on an arrow and took away his arrow and made
him his follower.
Apollo is an ancient god, mentioned in the Iliad Book 1, where, angry
that the Greeks have taken the daughter of his priest Chryses, he shoots
his plague-arrows at the Greeks for nine whole days, causing widespread
destruction to animals and men. Morris Silver83 mentions Cooks hypothesis that the cult of Apollo actually made its way along the old amber
route from the land of the Hyperboreans, thought to live north of the
Thracians and Scythians. Support for Cooks theory is found in a tradition
cited by Pausanias84 that Hyperboreans founded the cult of Apollo at Delphi. There is also the report of Apollonius Rhodius85 that Apollo shed
tears of amber when he was sent from heaven to Hyperborea. On the other
hand, Pausanias86 cites another tradition that Delphis second temple,
which replaced a mere hut, was sent to Hyperborea by Apollo. Herodotus
reports87 that in his time the cereal offerings arriving for Apollos summer
festival at Delos were accompanied by Hyperborean maidens and sacred
couriers called Perpherees. Farnell, says Silver, reasons, however, that the
Hyperboreans were not a people at all, but real ministers of the god who
performed certain sacred functions north of Hellas.88
The Doors to the Otherworld

To say that the Druids were oak-seers certainly links them to ancient
religious practices of the Mycenean Greeks or earlier, but it scarcely captures the essence of the priests who must have directed the activities of the
underworld cult of southern England and the ritual deposition of the horse
and dog at Danebury. So I would like to suggest another possible etymology for the word druid, one that takes us back to that mysterious poem,
The Spoils of Annwn. In line 24, the poet refers to the Fortress of FourPeaks, isle of the strong door, and we said (Chapter 3) that the fourpeaked or four-cornered fortress referred to the sky, and the strong door
referred to the door between our world and the otherworld. The Welsh
word used is dor, and this word is universal in Indo-European languages,
with dvar in Old Indian, dvarem in Avestan, thura in Old Greek, duris in
Baltic languages, foris in Latin, and Tor/Tr in German. I would argue
therefore that the rst element of druid is from the Indo-European word
for door, and that a druid is in fact a door-knower, or doorkeeper,
where door can be understood as the door to the Otherworld.
142
I say this because there is some evidence that early in the 1st millennium B.C. the classical world was already seeing doors/gates as entrances
to the Otherworld and linking them to divine gures. Hesiod in his
Theogony (c. 700 B.C.) refers to the gates of strong Hades and awful Persephone,89 which are guarded by Cerberus. The very ancient Roman god
Janus was associated with doors, but is not clearly related to any other
Indo-European deities. Frazer sees Janus as sky-god,90 and derives his name
from the same root *di which gives us Zeus; interestingly, he also links
Janus to the oakwoods of the Janiculum, the hill on the right bank of the
Tiber where Janus is said to have reigned as king. The name Janus may
alternatively be connected to the Etruscan Ani, a sky-god91 (and the Etruscans, as we have seen, could have come from northwest Anatolia, close
to the area where the visitors from the East may have originated).
Although Janus cannot be easily compared to any other European
god, Norman Oliver Brown points out92 that he can be linked to Hermes,
the god of the boundary-stone: Janus, like Hermes, is a trickster and carries
a magic wand; he was said to be the protector of enterprises of all sorts,
and the inventor of various arts, including religious ritual. Hermes, of
course, is another aspect of the rider-god, the Thracian Horseman, who,
like Janus, looks in at least two directions at once, and is sky-god, sungod, and god of the underworld a god who indeed is in charge of many
doors.
Arthur of Arcadia
If the Druids who came to Wiltshire in the 9th century B.C . were
from the Thracian lands to the north or northwest of Greece, then they
must have ofciated over a religion which venerated the oak as sacred, or
was well-versed in doors to the Otherworld, worshipped a horse-god like
Rhesus or the Thracian Horseman, and considered themselves to be
descended from a chthonic deity like the Thracian Hermes ( Julius Caesars Dis Pater). They may also have believed in some form of reincarnation
or transmigration of souls, inspired by early miracle-workers like Aristeas
and Abaris, whose trance-journeys are one probable source of Pythagorass
doctrine of metempsychosis (these trance-journeys were presumably
enhanced by cannabis, which was used by the Scythians and Thracians.93
They must also have been students of astronomy, like the priests of Stonehenge with whom they could well have joined forces, and it is astronomy
143
which provides the most compelling evidence for the early presence of
Druids in Wiltshire.
Arthur was no astronomer, but his name is inextricably bound up
with the stars. There are various ways to interpret his name. One possible
derivation has been popularized by C. Scott Littleton and Linda A. Malcor,94 who contend that the name comes from Artorius (plowman), the
second name of Lucius Artorius Castor, who brought over a group of Sarmatians to northern Britain as Roman cavalry. The Sarmatians, like the
Cimmerians, were Scythians, and Littleton and Malcor see the Arthur stories as variations of the tales of Batraz, hero of the Nart sagas of the Iranian-speaking Ossetians of the North Caucasus.
Another possible derivation, says Thomas Green,95 is from British
Arto-uiros bear-man a derivation supported by the fact that art(h) in
Welsh is used guratively to denote a warrior. Rachel Bromwich has
shown96 that Arcturus, deriving from the Greek word for Keeper of the
Bears and denoting a bright star associated with the Great Bear (Ursa
Major) constellation, was a genuine nonGalfridian variant form of
Arthurs name,
and one for which there is good reason to believe there was traditional
authority. Arcturus, like Arctos ( = Ursa Major or the bear) was often
used to denote the polar region, the far north, and there are references in
Latin literature to the savage and tempestuous weather associated with the
rising and setting of the star Arcturus. By extension, the name of the star
gave rise to the adjective used by Lucan for the Gauls as arctoas gentes
people of the (far) north, Bellum Civile V, 661. To name a hero Arcturus
could therefore be taken to imply that he belonged to the north (i.e. to
north-west Europe), and that he was bear-like in his characteristics.
Graham Anderson takes the identication between Arthur and Arcturus a step further. Anderson points out97 that the star Arcturus was
already referred to in the agricultural poetry of the early archaic period (c.
700 B.C.), when it was used interchangeably with the name of its constellation, Botes98 (see Figure 11). Hesiod says in his Works and Days99:
When in the rosy morn Arcturus shines,
Then pluck the clusters from the parent vines; ...
and again, but for a different season of the year: ...
When from the Tropic, or the winters sun,
Thrice twenty days and nights their course have run;
And when Arcturus leaves the main, to rise
A star bright shining in the evening skies;
Then prune the vine.
144
Anderson conrms100 that the spelling Arcturus is found in medieval texts,

and notes that the form Arthurus was used for the star itself: John of Howdens 13th-century Cythera refers to the giram Arthuri, Arthurs round,
when discussing astronomical movement of the star.
As Bromwich observes, Arcturus was associated with savage and tempestuous weather. Richard Hinckley Allen points out101 that Pliny called
it a horridum sidus (horrible star), and that
Demosthenes, in his action against Lacritus 341 B.C., tells us of a bottomry
bond, made in Athens on a vessel going to the river Borysthenes the
modern Dnieper and to the Tauric Chersonese the Crimea and
back, that stipulated for a rate of 221 /2 percent interest if she arrived
within the Bosporus before Arcturus, i.e. before its heliacal rising about
midSeptember; after which it was to be 30 per cent. Its acronychal rising
xed the date of the husbandmens Lustratio frugum; and Vergil twice
made allusion in his 1st Georgic to its character as unfavorably affecting the
farmers work. Other contemporaneous authors conrmed this stormy reputation, while all classical calendars gave the dates of its risings and settings.
The heliacal rising of a star is when the star makes its rst seasonal appearance on the eastern horizon just before dawn, and the achronycal or acronychal rising is when the star rst becomes visible above the eastern horizon
at nightfall. In antiquity the heliacal rising occurred midSeptember, and
the acronychal rising occurred midFebruary, which Allen says coincided
with the lustratio frugum or purication of the fruits.
The mythology of Arcturus is as ancient as the references to it. In a
fragment attributed to Hesiod and quoted in the Catasterismi, the following
story is told of the nymph Callisto102:
The Great Bear.] Hesiod says she (Callisto) was the daughter of Lycaon
and lived in Arcadia. She chose to occupy herself with wild-beasts in the
mountains together with Artemis, and, when she was seduced by Zeus,
continued some time undetected by the goddess, but afterwards, when she
was already with child, was seen by her bathing and so discovered. Upon
this, the goddess was enraged and changed her into a beast. Thus she
became a bear and gave birth to a son called Arcas. But while she was in
the mountains, she was hunted by some goat-herds and given up with her
babe to Lycaon. Some while after, she thought t to go into the forbidden
precinct of Zeus, not knowing the law, and being pursued by her own son
and the Arcadians, was about to be killed because of the said law; but Zeus
delivered her because of her connection with him and put her among the
stars, giving her the name Bear because of the misfortune which had
befallen her.
145
There is also another story attributed to Hesiod which Anderson quotes103:

It is said that Arkas is the son of Callisto and Zeus ... and Lycaon pretended not to be aware that Zeus had raped his daughter: he entertained
Zeus, according to Hesiod, cut up the child and laid him on the table. At
which Zeus overturned the table from which the city gets its name
Trapezous struck his house with a thunderbolt, and changed Lycaon into
a wolf. He then put Arkas together again and made him whole; and the
child was nourished by a herdsman.
The Latin writer Hyginus, in his Fabula, from the 1st century A.D., says104:
But the sons of Lycaon wanted to test Jupiter to see whether he was a god;
they mixed human esh with other esh and put it in front of him at a
feast. After he became aware of this, he overturned the table in a rage, and
killed Lycaons sons with a thunderbolt. In that place Arkas later fortied a
town which he called Trapezous.
Arcas, who was also changed into a star to protect his mother, was
the founder of the Greek state of Arcadia. Anderson draws a parallel105
between Arcas/Arcturus, who founded a community called Table, and
King Arthur, who built the Round Table. Arthur built his Round Table
to avoid bloodshed that quarrels at a table had caused among his men, just
as Arcas/Arcturus founded the city of Table to take the place of the table
where Zeus had shown his anger at human sacrice.
There is another very distinctive detail about Arcas/Arcturus, says
Anderson.106 He is described by a commentator of the Hellenistic poet
Aratuss astronomical poem the Phaenomena (3rd century B.C.) in the following terms:
And behind Helice the Bear someone like a driver is being carried along,
the so-called Arctophylax (Arcturus). As to his resemblance to a driver, this
is because he is carrying a kalaurops (Aldine edition kalabrops) in his right
hand, as if holding on to the cart, the bear, with his left. For he himself
seems to be the guardian of the cart, called the bear, as Bootes (the herdsman), who is driving the oxen (from it), carries a kalaurops, which is a club.
The kalaurops or kalabrops held by Arcturus sounds very similar to Arthurs

sword Excalibur,107 rst given by Geoffrey of Monmouth as Caliburnus.
But if Excalibur has an Arcadian origin, the Sword in the Stone has
a Scythian origin, according to C. Scott Littleton and Linda A. Malcor.108
They point to a passage from Herodotus109 where he describes Scythian
sacrices to their war-god:
In every district, at the seat of government, there stands a temple of
[Mars], whereof the following is a description. It is a pile of brushwood,
146

made of a vast quantity of fagots, in length and breadth three furlongs; in
height somewhat less, having a square platform upon the top, three sides
of which are precipitous, while the fourth slopes so that men may walk up
it. Each year a hundred and fty wagon-loads of brushwood are added to
the pile, which sinks continually by reason of the rains. An antique iron
sword is planted on the top of every such mound, and serves as the image
of Mars: yearly sacrices of cattle and of horses are made to it, and more
victims are offered thus than to all the rest of their gods. When prisoners
are taken in war, out of every hundred men they sacrice one, not however
with the same rites as the cattle, but with different. Libations of wine are
rst poured upon their heads, after which they are slaughtered over a vessel; the vessel is then carried up to the top of the pile, and the blood
poured upon the scymitar. While this takes place at the top of the mound,
below, by the side of the temple, the right hands and arms of the slaughtered prisoners are cut off, and tossed on high into the air. Then the other
victims are slain, and those who have offered the sacrice depart, leaving
the hands and arms where they may chance to have fallen, and the bodies
also, separate.
Much later, Ammianus Marcellinus, writing around A.D. 390, says this of
the Scythian Alani, who lived northeast of the Black Sea110: Nor is there
any temple or shrine seen in their country, nor even any cabin thatched
with straw, their only idea of religion being to plunge a naked sword into
the ground with barbaric ceremonies, and then they worship that with
great respect, as Mars, the presiding deity of the regions over which they
wander.
However, the aspect of the Arthur story that interests us most is the
cauldron of The Spoils of Annwn, which later became the Holy Grail of
medieval Arthurian romances. As Anderson points out,111 Herodotus tells
a story of the Scythians which is reminiscent of Grail stories112:
According to the account which the Scythians themselves give, they are the
youngest of all nations. Their tradition is as follows. A certain Targitaus
was the rst man who ever lived in their country, which before his time
was a desert without inhabitants. He was a child I do not believe the
tale, but it is told nevertheless of Jove and a daughter of the Borysthenes. Targitaus, thus descended, begat three sons, Leipoxais, Arpoxais,
and Colaxais, who was the youngest born of the three. While they still
ruled the land, there fell from the sky four implements, all of gold a
plough, a yoke, a battle-axe, and a goblet. The eldest of the brothers perceived them rst, and approached to pick them up; when lo! as he came
near, the gold took re, and blazed. He therefore went his way, and the
second coming forward made the attempt, but the same thing happened
again. The gold rejected both the eldest and the second brother. Last of all
the youngest brother approached, and immediately the ames were extin-
147
Figure 11: Stylized representation of Botes the ox-driver or plowman, sometimes also referred to in classical antiquity by the name of its brightest star
Arcturus, here shown on the left knee of the ox-driver/plowman.
guished; so he picked up the gold, and carried it to his home. Then the
two elder agreed together, and made the whole kingdom over to the
youngest born. The sacred gold their kings guard very carefully, and honour it with magnicent sacrices. And if any guardian of the gold falls
asleep in the open air during the festival, the Scythians say he will not last
out the year.
Here the equivalent of the Grail is the golden goblet, which like the other
three golden objects, can only be claimed by someone worthy, in this case
the youngest son. The parallels with the cauldron and Grail are more clear
in the much later Nart sagas, which were only written down in the 19th
century. Three of the Narts lay claim to the Nartamongae, the cauldron of
the Narts they are called Uryssmaeg, Soslan and Sosryquo. There are,
says Anderson,113
148

a number of variants: in one tale, each of the Narts has a little cask in
front of him: if his boast is true it overows three times. In story 16 it is a
magic cup which rises of its own accord and goes to the lip of the bravest
man. Three other cauldron tests are offered in nos. 41ff.: the Nartamongae
will only go to the warrior without fault, and only Batraz qualies, while
Uryssmaeg, Soslan and Sosryquo fail; only Batraz can cause ice in the
cauldron to boil merely by reciting his exploits; or can win the ladle of the
cauldron by emptying the vessel of noxious animals.
The cauldron of renewal is also found in Greek mythology. There is

an early tradition, in a fragment of Hesiod, that Medea, daughter of the
king of Colchis and therefore from the regions north of the Black Sea,
was able to renew Aeson, the father of Jason, in a golden lebes114 (cauldron). The cauldron theme is also taken up indirectly in the foundation
myth of the Macedonians, recounted by Herodotus115:
Now of this Alexander the seventh ancestor was that Perdiccas who rst
became despot of the Macedonians, and that in the manner which here
follows: From Argos there ed to the Illyrians three brothers of the
descendents of Temenos, Gauanes, Aropos, and Perdiccas; and passing
over from the Illyrians into the upper parts of Macedonia they came to the
city of Lebaia. There they became farm-servants for pay in the household
of the king, one pasturing horses, the second oxen, and the youngest of
them, namely Perdiccas, the smaller kinds of cattle; for in ancient times
even those who were rulers over men were poor in money, and not the
common people only; and the wife of the king cooked for them their food
herself. And whenever she baked, the loaf of the boy their servant, namely
Perdiccas, became double as large as by nature it should be. When this
happened constantly in the same manner, she told it to her husband, and
he when he heard it conceived forthwith that this was a portent and
tended to something great. He summoned the farm-servants therefore, and
gave notice to them to depart out of his land; and they said that it was
right that before they went forth they should receive the wages which were
due. Now it chanced that the sun was shining into the house down
through the opening which received the smoke, and the king when he
heard about the wages said, being infatuated by a divine power: I pay you
then this for wages, and it is such as ye deserve, pointing to the sunlight.
So then Gauanes and Aropos the elder brothers stood struck with amazement when they heard this, but the boy, who happened to have in his
hand a knife, said these words: We accept, O king, that which thou dost
give; and he traced a line with his knife round the sunlight on the oor of
the house, and having traced the line round he thrice drew of the sunlight
into his bosom, and after that he departed both himself and his fellows.
Anderson says116 that Gauanes is the equivalent of Gawain in Arthurian

stories, and Lebaia is related to lebes, a cauldron or broad dish, supplying
149
the place-name vesselville or platter-ville (that is, Grail town). Gawain

is traditionally the son of Lot, and a temenos is a lot, in the sense of an
allocation of land. Perdiccas seems to play the same role here as the
youngest son in the Scythian tale the Scythian youngest son was granted
the gold implements, while Perdiccas took possession of the symbolic gold
of the sun.
All this happened around 700 B.C., and Pettazzoni says117 that in this
story we see traces of an ancient devotion to the sun on the part of the
Macedonian royal house. This sun-worship continued in later times, for
the Roman historian Livy tells us that in 181 B.C. Philip V of Macedonia,
accompanied by his son Perseus, climbed a peak of Mt. Haemus, in present-day Bulgaria, and descended again in haste, after having dedicated
two altars, one to Jupiter and the other to the Sun. The fact that Perdiccass
bread swells to double its normal size may be linked to Perdiccass solar
powers in any case, it recalls the cornucopia and the magic hamper or
bag in the First Branch of the Mabinogion and the story of Lludd and Llefelys.
Arthur and Ancient Astronomy

It would clearly be helpful if we could look at the story of Arcturus
in the context of ancient beliefs about the stars, and an insight into such
beliefs is provided by the Canadian physicist and student of ancient astronomy and religion F. Graham Millar, in The Celestial David and Goliath.
His thesis is that the Biblical story of David and Goliath has an astronomical origin, with David representing the constellation of Botes (also known
by the ancients as Arcturus), his sling representing the nearby Corona
Borealis, Goliath representing the constellation of Orion on the other side
of the sky, and Saul, who gives David his armor, representing the constellation Hercules. Linked to the story of David and Goliath is the Irish myth
of Lugh and Balor.118 In Irish legend Balor was a famous warrior who had
one eye in the middle of his forehead; another, his baleful eye, was in the
back of his skull. With it he could strike people dead by looking at them,
but he kept it covered except when he wanted to petrify his enemies. Millar
says that according to Mac Cana,119 when Lugh saw Balor open his eye
against him, Lugh cast one of his father Aeds thunderbolts at him with a
slingshot, driving the thunderbolt through the back of Balors head and
killing him. The version that Millar gives is one of several the Aed he
150
mentions is another name for the Dagda, but in another version it is a

spear made by the weapon-forger of the gods, Goibniu, that kills Balor,
and in yet another version it is, appropriately, a sling-stone.
The investigation of the David and Goliath story as a myth of the
stars, says Millar,120 must begin with noting the effect of precession of the
equinoxes. It is the slow conical swing of the axis of the Earth, acting as
a huge spinning top. The result is that, in a period of 25,800 years, the
pole of the heavens moves as a circle on the celestial sphere on a radius of
23.4.
Precession caused the ancients to fear that the sky was falling.... Vega
was near the celestial pole of 11,000 B.C., and may have been regarded in
ancient times as the supreme god, who supported the apex of the universe.
The Akkadians called it the Life of Heaven, and the Assyrians the Judge
of Heaven.121 By 8000 B.C. they must have noticed that it had fallen from
its place, and it has continued to fall. The fall may be mentioned in the
Bible (Isaiah 14:12):
How you have fallen from heaven,
O morning star, son of the dawn!
You have been cast down to the earth,
you who once laid low the nations!
Today Vega, says Millar, is 51 from the pole.

Among those who feared the sky was falling were the Celts: when
Celtic ambassadors discussed a treaty with Alexander the Great in 335
B.C., says Rankin,122 they admitted fear of nothing except, perhaps, that
the sky should fall (Strabo 7.38, Arrian Anabasis 1.4.6). Their form of
words, which may have been close enough to an oath, actually referred to
a substantial fear that the world might sometime end (Livy 40.58.46).
The concept of the sky falling, says Millar, helps us to interpret the
rock carving of the rst millennium B.C . from Camonica in northern
Italy.123 This shows a tall gure, apparently with horns, with arms upraised;
on his right arm is a torc; to his left and at the level of his arms is a ring
or circle, and below this, and just beneath his arms is a snake. Millar
believes that this gure, the Horned One, depicts the combined constellation of Hercules and Ophiucus (Snake-Holder). The combined asterisms possibly constituted the obsolete constellation of Menat.124 Mac Cana,
says Millar, pronounced the tall gure to be the antlered god Cernunnos.
He observed the serpent in his left hand, which Millar sees as the constellation Serpens Caput (to the west of Ophiucus). Mac Cana also commented upon the torc (heavy circlet) on his right arm; Celtic gures have
151
frequently been depicted as wearing the torc as an amulet. The ring or

circle to the left of the gure is, according to Millar, in the right position
for Corona Borealis.
Millar goes on to say125 that the hands of Cernunnos are raised in
the weight lifters pose, often called the orans position, and argues that
the god is holding up the sky. Mircea Eliade has described126 how the Siberian shamans believed that the high branches of the sacred birch reached
to the sphere of the xed stars, and would erect birch pillars to support
the sky. Like the pillars, the antlers of Cernunnos may perhaps be interpreted as supporting the sky.
The pole of the heavens was midway between the upraised arms of
Hercules at 8000 B.C. Hence, the idea that Cernunnos supported the sky
is likely to have originated broadly near that date. When precession continued, the persona of the Horned One moved into the constellation of
Botes, where he was known as Lugh.
Millar contends127 that the Gundestrup Cauldron also provides clues
to ancient astronomical beliefs. Four panels on the Cauldron depict divinities as busts with arms upraised in the orans position. Some or all of them
were probably replications resident in the constellation of Hercules, inheriting from Menat the function of supporting the sky. One of the panels
with a gure having upraised arms on the Gundestrup Cauldron is accompanied by a wheel, which Millar interprets as the Corona Borealis.
The most famous panel on the Gundestrup Cauldron is Plate A, the
one with the horned gure (see Chapter 2), and Millar interprets this in
astronomical terms. The central image, the Horned one, is the constellation
of Botes, and other images are the neighboring constellations:
Subject to artistic license, they are in the right places. Ursa Major is turned
around to face the Horned One to express his vassalship. Clockwise from
the lower right, the following constellations can be recognized: Leo Major
and Minor, Hydra, Botes, Hercules/Ophiuchus (bearing horns), the Ass
(an obsolete constellation), Ursa Minor, Delphinus, and Capricornus. On
comparing the central Horned One with Botes on the star chart, one sees
a compelling likeness: the Buddhic position of the legs, and the torc on his
right arm corresponding in position with Corona Borealis; this was also
Lughs sling.
As Millar points out,128 there is no Balor on the Gundestrup Cauldron.

For him we may look elsewhere for one or more of the motifs a single
eye, blindness, a weapon piercing the eye in the back of his head, the piercing of the head, or beheading. Orion is reputed to be blind129; no other
152
constellation bears this attribute. The head of Orion is represented by only

one star, Meissa.130 It is a double in which the brighter star is pale white
and of magnitude 3.5. To the naked eye this is a dim star, not visible
through hazy cloud, so on occasion Orion is blind or lacks a head. Astrologically, then, Balor can be identied with Orion.
This just leaves the question of the weapon red at Goliath or Balor,
and Millar believes131 that the weapons in the myth
may have been inspired by a meteor shower. A meteor, or shooting star, is
vividly mythologized as a thunderbolt, a thrown spear, a glinting sword,
or a sword half-drawn from the scabbard and reinserted....
From year to year showers may vary in the maximum number visible per
hour. A shower comes from a denite direction in space, and by perspective the meteors seem to radiate from a point on the celestial sphere called
the radiant. A shower takes its name from the constellation in which the
radiant lies.
Of all the constellations, only three bear the name of a mythical hero
and at the same time contain a meteor radiant. Perseus is one, but he can
be identied with Mithra and so is disregarded here. Hercules and Orion
remain, the possessors of the thrown weapons. Never in any age has Orion
been visible at the time of the Herculids, since at that time the Sun is
invariably close to Orion, outblazing him.
Botes, says Millar, is not the radiant for a major meteor shower, and
neither is Corona Borealis. However, the constellation Hercules is the origin of a shower of meteors, and it could have seemed to the ancients that
Hercules (Aed or Goibniu in Irish mythology) was forging weapons and
giving them to Botes, who then ung the bolts with his sling, Corona
Borealis. Around 3000 B.C., Orion was visible from August to October,
and Botes, low in the north on August 4th, was in position to bombard
him until his heliacal setting his last seasonal appearance on the western
horizon just after sunset.
Toward the end of his paper,132 Millar refers to the Welsh story that
comes closest to the story of David and Goliath or Lugh and Balor, an
episode in Culhwch and Olwen. The giant Ysbaddaden has three poisoned
darts: the rst night he throws one at Bedwyr, who catches it and throws
it back at the giant, wounding him in the knee; the second night he throws
one at Menw, who catches it and throws it back at the giant, wounding
him in the center of the breast. They return on the third night, and the
giant says:
Shoot not at me again unless you desire death. Where are my attendants?
Lift up the forks of my eyebrows which have fallen over my eyeballs, that I
153
may see the fashion of my son-in-law. Then they arose, and, as they did
so, Yspaddaden Penkawr took the third poisoned dart and cast it at them.
And Kilhwch caught it and threw it vigorously, and wounded him through
the eyeball, so that the dart came out at the back of his head.
In fact, there is also the episode in the Fourth Branch of the Mabinogion
where Lleu, the Welsh equivalent of Lugh, takes aim, presumably with his
needle, and hits a wren between the sinew and the bone of its leg a very
pale imitation of the Lugh-Balor encounter.
Both of these tales have links to ancient astronomy, if only indirectly.
Lleu can get his name only with the help of his mother Arianrhod ( =
Corona Borealis), and shoots at the wren to achieve this. Culhwch, is associated with pigs and through this to the Otherworld, like Arthur in The
Spoils of Annwn, and perhaps is taking Arthurs place in the Ysbaddaden
episode after all, at one point, his followers urge Arthur to go home and
not bother himself with such small adventures as these, implying that
Arthur may in earlier versions have been involved in adventures like the
Ysbaddaden episode, whose signicance was not appreciated by later storytellers. So it is possible that Arthur/Arcturus was originally the equivalent
of Lugh, originally charged with preventing the skies from falling, and
ghting off celestial enemies with the help of the divine sling of the Corona
Borealis, but reduced in the Mabinogion to ghting giants like Ysbaddaden
with poisoned darts, or hitting wrens with needles to get a name from Arianrhod (the Corona Borealis).
Conclusion
We have attempted to show, through British legendary history,
through archeology, through an exploration of Thracian and Greek religion
and mythology, and through ancient astronomical beliefs, that the MoundPeople of Wiltshire in the 9th century B.C. could have come from somewhere in eastern Europe, possibly by the amber route, from Thrace through
Hyperborea to the Baltic and western Germany, possibly via the Thaco
Cimmerian expansion from Thrace through Hungary, Austria, Switzerland
and eastern France. What language did these people speak? It could have
been one of the extinct languages of eastern or Central Europe like Thracian or Illyrian, or one of the extinct languages of western Europe like
Ligurian, spoken in northwestern Italy and southeastern France, or even
an Indo-Iranian language. There is one tiny and ambiguous clue provided
154
by the name of the river which passes near Stonehenge, the Avon (which
is, admittedly, found in other parts of England). The name is usually said
to be Celtic. Matasovic133 links it to Old Irish ab, aub, Middle Welsh afon,
Gaulish ambe, and ultimately to Hittite hapa river, Old Indian apa
water, Old Greek apeiros coast, Latin amnis stream. In Chapter 3 we
saw that Varuna lost his identity in Zoroastrianism, becoming Apam Napat,
Grandson of the Waters. So the river of Stonehenge could have acquired
its name long before the Celts, as a tribute to a long-forgotten god.
We have also proposed in this chapter that the Druids are not Celtic,
but preceded the Celts into Wiltshire at the same time as the people who
brought the All Cannings Cross pottery. We have linked their veneration
of oak trees to the ancient Illyrians of Dodona, their possible interest in
doors to the earliest Greeks and Romans and possibly to the Etruscans,
and their burial practices to Indian, Iranian and Greek views of the afterlife.
We have seen that their doctrine of reincarnation could have come from
the same sources that Pythagoras drew on. And most importantly, we have
shown that the story of Arthur/Arcturus could have originated in eastern
or central Europe, as part of an astronomical tradition going back thousands of years. But the story of Arthur is not about prehistory, but about
a period of history after the Roman army left Britain, and we will be looking at this Arthur in the nal chapter.
Chapter 7
Brutus of Troy Town

The Truth Lovers
But before we can look at the historical Arthur, we need to review
all the evidence we have assembled for the origins of the Mound-People
and the Druids, and to advance any new evidence which may seem relevant
to the question. The rst piece of evidence comes from an unexpected
source the tale of Culhwch and Olwen, rst discussed in Chapter 1. In
this tale, Arthurs men have to nd Mabon son of Modron, and to do this
they must ask a number of long-lived animals: rst they ask the ousel
(blackbird) of Cilgwri, who does not know where Mabon is, but advises
them to consult an older animal, the stag of Redynvre; the stag cant help,
but takes them to an even older animal, the owl of Cwm Cawlwyd; the
owl cant help either, but leads them to the oldest animal in this world,
the eagle of Gwern Abwy; the eagle doesnt know, but takes them to an
animal even older than he is, the salmon of Llyn Llyw, who nally tells
them that Mabon is a prisoner in Gloucester.
The oldest-animal format of this story is not unique to Culhwch
and Olwen. In a fragment attributed to Hesiod1 it is reckoned that the
crow lives nine human generations, the stag four times as long as a crow,
the raven three times as long as a stag, the date-palm nine times as long
as a raven, and the Nymphs ten times as long as the date-palm. West notes2
that the verses were often quoted in late antiquity and more than once
rendered into Latin, so could have been known to anyone with a classical
education. However, he says,3 neither Classical learning nor any other
form of horizontal transmission can account for the extraordinary parallel
between Culhwch and Olwen and an episode in the Mahabharata, the
Indian epic composed between 800 B.C. and 400 B.C. In the Mahabharata,
it is related that the royal seer Indradyumna fell from heaven because no
one any longer remembered him. He asked the sage Markandeya if he rec155
156
ognized him. Markandeya did not, but told him that there was an owl in
the Himalaya who was older than himself and might know him. They
went there. The owl said that he did not recognize Indradyumna, but that
there was a crane living by a lake who was older than he. The crane was
also unable to identify the seer, but said there was a tortoise in the lake
who was even older than he was and might know more. The tortoise was
summoned and after much reection recognized the seer, who had formerly
built his re altars on the tortoises back.
The link between Britain and ancient India is reinforced by The Spoils
of Annwn. In Chapter 3 we noted that (Caer) Sidi, the Mound-Fortress,
was cognate with Vedic Sanskrit sadas the ritual space belonging to the
gods where the soma sacrice took place, and, more importantly, that (Caer)
Pedryuan, the Four-Peaked or Four-Cornered Fortress, could be best
explained by reference to four-cornered Varena of Zoroastrianism, and
four-faced (caturanika) Varuna (more properly, Indra-Varuna) of the
Rigveda with his four-cornered (caturasrir) thunderbolt. This link to
Varuna is reinforced in another poem attributed to Taliesin, Song Before
the Sons of Llyr, which refers to the Distributor in Indo-Iranian texts,
Varuna was sometimes called Baga, which can be translated as The Distributor. As we noted in Chapter 6, Varuna was called Apam Napat,
Grandson of the Waters, in India and ancient Persia, which gives an
alternative etymology for the river of Wiltshire, the Avon.
Kazanas says4 that Varuna is the
king of the gods, like Odin in Asgard and Zeus on Olympus. He personies more than the sky (space or substance) which encompasses everything.
An ethical god, he lays down laws for every level of creation and rules
through maya, measuring knowledge or unfathomable power. He watches
everything from his golden palace in highest heaven and has spies everywhere. He binds the sinner with fetters but also liberates and grants victory in war. He is also associated with waters and oceans and retains only
this feature in postVedic texts. Varuna is almost invariably lauded with
Mitra and often with Aryaman as well, in a trinity. Both Varuna and Mitra
are called samraja (emperors), and guardians of cosmic order (rta) in highest heaven. In some hymns and later texts Varuna is associated with night
and Mitra with day. Mitra is a daytime aspect of the sun connected with
friendship and contracts.... In the Iranian Avesta, the supreme god is Ahura
Mazda, who resembles Varuna is his ethical aspect and kingship; his power
of light is Mithra.
These connections hint at an Indo-Iranian origin, which might not

be inconsistent with legends that see Britain as founded by a Brutus of
7 Brutus of Troy Town
157
Troy Indo-Iranian peoples may have lived north of the Black Sea and
around the River Danube, which from a British perspective is not far from
northwest Anatolia. At any rate, further weight is given to an Indo-Iranian
origin by Pokornys contention that Pritteni is derived from Indo-Iranian
prtana/partara battle, implying that Pryderi of the Mabinogion was originally a ghter, and Britain was originally Land of the Fighters. If such
warriors did indeed settle in Britain, they did have other more prosaic concerns, if I am correct in assuming that the tors of Dartmoor take their
name from an Indo-Iranian word identied by H.W. Bailey as cognate
with Sanskrit torana arch, gateway and Khotanese ttora top, and
derived from a root meaning place over, upon, or around. Such longrange migrations might seem out of the question, but Beekes argues that
the Etruscans of Italy migrated from northwest Anatolia, in the region
around Troy, in the 12th century B.C., at the same time that their Lydian
brothers headed south for Manisa and Izmir province. If the whole eastern Mediterranean was suffering from famine, war and civil unrest, then
it seems highly likely that mass migrations took place, which could have
led some people as far as Britain.
Further evidence is provided by the story of Elidorus, reported by
Giraldus Cambrensis after his travels through Wales. The mound-people
that Elidorus encountered had no form of public worship but were lovers
and reverers of the truth. This of course may be simply a way of describing
an ideal society, like that envisaged in Platos Republic (4th century B.C.),
or Augustines City of God (5th century A.D.). Plato believed that society
should be ruled by philosophers, who are lovers of the truth; but Platos
works were virtually unknown to western Europeans scholars in the 12th
century. Augustine, in his City of God, touches on Plato, observing that if
a philosopher is a lover of truth, then he must be a lover of God. So it is
more likely that the legend of Elidorus refers to something else. Giraldus
tells us that the mound-people never took an oath, which seems to rule
out a link with the Thracian Hermes.
One possibility is that the mound-peoples love of truth is connected
with Varuna and rta. James Hastings, in discussing rta as moral order,
says5 that the conception of moral order is doubtless Indo-Iranian, that
the conception cannot be more recent than the 15th century B.C., and
that it was developed before the Vedic Aryana entered India. He says
that the gods themselves are not merely born of the rta, but they follow
the rta; they are practicers of the rta and knowers of it. The special guardian
of the rta is Varuna, the great guardian of morality, who moves about dis-
158
cerning the truth and the unrighteousness of mankind. As mentioned earlier the Zoroastrian equivalent of Varuna is Ahura Mazda mazda means
intelligence, wisdom, and is cognate with Sanskrit medha, whose adjectival form medhira wise is used to describe Varuna. It may be signicant
that the name of Pwyll, who becomes Lord of the Otherworld and marries
the horse-goddess Rhiannon, means mind, spirit, reason, suggesting a
possible link to Mazda and to Math, the master magician of the Fourth
Branch and the brother of Don.
In Zoroastrianism the equivalent of rta is asha or arta, and the opposite is druj. John Waterhouse says6 that druj, or Lie, was the enemy who
attacked man from within, and was the foe of Truth. As the opponent of
Asha (Truth) the Druj represented the falsehood of the old gods,7 and
later, in the Acheamenian age (550 B.C.330 B.C .), the Druj became identied with Ahriman, the devil. In an account of the Persians, Herodotus
remarked: They teach the boys, from ve years old to twenty, three things
only to ride, to shoot, and to be truthful.... Most disgraceful of all is
lying accounted.
Lies and Labyrinths

The word druj may possibly be found in England and Wales in an
unexpected place. W.H. Matthews, in Mazes and Labyrinths, discusses the
relationship between Troy and labyrinths. In Historia Regum Britanniae,
Geoffrey of Monmouth said that the founder of Britain, Brutus, was
descended from Aeneas, who left Troy and sailed to Italy. Once in Britain,
Brutus founded a city called New Troy, a name presumably based on the
tribe called the Trinovantes though Geoffrey was Welsh and presumably
knew that Trinovantes had nothing to do with Troy. However, Matthews
provides a new insight into the legendary link between Troy and Britain.
Matthews explores labyrinths of all kinds, but in Chapters 10 and 11
he focuses on turf mazes, which are found all over England with various
names such as Mizmaze, Julians Bower, Troy Town or Shepherds Race.8
Julians Bower is at Alkborough in Lincolnshire; the Mizmaze is at
Breamore, near Fordingbridge in Hampshire9; there is a squarish Mizmaze
at St. Catherines Hill near Winchester in Hampshire10; there was a maze
near Dorchester the memory is still preserved in the name of a farm,
Troy Town, near Puddletown11; there was a Walls of Troy in the marshes
of Rockcliffe in Cumbria12; there is a Troy Town at Somerton, near Banbury
159
in Oxfordshire13; in Surrey a Troy Town was formerly well known in the

neighborhood of Guildford, at Hillbury, between Guildford and Farnham.14 In Chapter 12 Matthews also mentions15 a curious custom formerly
prevalent among Welsh shepherds, which consisted of cutting in the turf
a gure in the form of a labyrinth, which they called Caerdroia, i.e. the
walls, or citadel, of Troy. Finally he points out16 that there are labyrinths
in northern Europe, not of turf but stone-lined, which also have Troy in
their name, for example Trojeborg or Trborg.
Matthews believes that the name Troy Town is quite ancient, and
quotes a French account17 of a voyage to Jerusalem in 1418. The Seigneur
de Caumont stopped off at Crete and wrote of that intricate house made
by Dedalus ... which was named Labyrinth and today is vulgarly called by
many the city of Troy. But the link between Troy and labyrinths is much
older than that, as shown by an Etruscan vase found at Tragliatella. This
was very roughly decorated with incised gures, representing amongst
other things a circular labyrinth of the traditional type and some horsemen
who are thought to be engaged either in the attack on Troy or in the game
known as the Lusus Trojae or Game of Troy. That there can be no doubt
about the artists identication of the labyrinth in some way with the celebrated city in question is clear from the word Truia scratched within it.18
The meaning of the Game of Troy is found in Virgils Aeneid. In
Book 5, Aeneass son Iulus is taking part with his companions in a sport
called Ludus Trojae or Lusus Trojae (Game of Troy), or simply Troja.19
The game consisted of a sort of processional parade or dance, in which
some of the participants appear to have been mounted on horseback. Virgil draws a comparison between the complicated movements of the game
and the convolutions of the Cretan Labyrinth:
As when in lofty Crete (so fame reports)
The Labyrinth of old, in winding walls
A mazy way inclosd, a thousand paths
Ambiguous and perplexed, by which the steps
Should by an error intricate, untracd
Be still deluded.
But Virgil is not the rst to mention a ceremonial dance.20 Homer, in the
Iliad (Book 18), is describing Achilless shield21:
Next on that shield, the celebrated lame god made
an elaborately crafted dancing oor, like the one
Daedalus created long ago in spacious Cnossus,
for Ariadne with the lovely hair. On that oor,
160

young men and women whose bride price would require
many cattle were dancing, holding onto one another
by the wrists.
Sometimes, says Homer, they would dance deftly in a ring with merry
twinkling feet, as it were a potter sitting at his work and making trial of
his wheel to see whether it will run, and sometimes they would go all in
line with one another. The tradition of a ceremonial dance associated
with the Cretan labyrinth continued long after Homer. It was said that
Theseus killed the deadly Minotaur and ed with his Athenian comrades
to Delos, where they celebrated their triumph through song and dance.22
This victory dance was called the Crane dance (Geranos), after the complicated ight patterns of the crane. Plutarch (1st century A.D.) describes
this dance in his Parallel Lives23: Theseus, when he sailed away from Crete,
touched at Delos; here ... he and the youths with him danced a measure
which they say is still practiced by the people of Delos to this day, being
an imitation of the turnings and windings of the Labyrinth expressed by
complicated evolutions performed in regular order. This kind of dance is
called by the Delians the crane dance.
Matthews speculates24 on the possible origins of the word Troy in Troy
Town. It may, he says, have originated not with the name of a town, but
with some ancient root signifying to wind, or turn. He also links Troy
to the Druja or Draogha of Persian legends and the druh demon of the
Rigveda. The rst suggestion derives Troy from the Welsh tro, which is
from the Indo-European root tragh,25 which gives Latin traho pull, Old
Irish traig foot, Welsh troed foot, Old Irish tragud tide, Welsh treio
ow back to the sea, Welsh tro turn, variation, time, Welsh troi to
turn, roll, and English drag. A variation on this theme is provided by
Henning Eichberg26 in discussing northern European stone-lined
labyrinths, he links Troy to German drehen to turn, cognate with English
throw.
But the suggestion that interests us most is the link with Druja and
Druh. Pokorny27 derives these words from the Indo-European root
dhreugh-, which gives Avestan dru aiti lies, cheats, Old Indian dr gha-,
drha- insult, damage, betrayal, Avestan draoga- fallacious, m. lie,
falsity, deception, Old Persian drauga- fallacious, Old Indian drhend, demon, Old High German triogan deceive, Old Norse draugr
m. ghost, Old Norse draumr, Old High German troum, Anglo-Saxon
drm, English dream dream.
This brings us right back to Elidoruss mound-people and their love
161
of truth, and to the Zoroastrian asha/arta and its opposite druj. If a group
of Iranian-speaking people did settle in southern England, what might
they have associated with deception, demons and, possibly, labyrinths?
Dennis Price28 claims that Stonehenge was once a labyrinth, and its association with music and dancing may be reected in the name that Geoffrey
of Monmouth uses for it the Giants Dance. It is certain that any prehistoric visitor to England would have heard of Stonehenge, and might
well have given it a name which reected their belief that it was a place of
deception, or a place made by demons.
The People of the Cauldron

The antiquity of such settlers seems to be implied in the story from
the Second Branch of the Mabinogion concerning the origin of the cauldron
of regeneration. Bran, it will be recalled, got the cauldron from an Irishman
called Llasar Llaes Gyfewid and his wife Cymidei Kymeinvoll, who had
escaped from the Iron House in Ireland. The story goes that while the
Irish king was hunting, he was on top of a tumulus (gorsedd) above a lake
in Ireland called the Lake of the Cauldron. A large, monstrous man
emerged from the lake with a cauldron on his back, accompanied by an
even bigger woman. Matholwch took them in, but they insulted and
injured people, and made themselves generally hated. However, the two,
plus their children, would not go of their own free will, nor could they be
forced to go. So the Irish king and his people decided to forge a solid iron
chamber, surround it with charcoal, and re it until it was red hot. The
man and woman escaped and ed to Wales with their cauldron.
As Cunliffe points out, 29 the rst bronze cauldrons in Britain are
thought to have emerged at the end of the second millennium in the wake
of contacts with central Europe. Early in the rst millennium Ireland seems
to have taken over as the main production center, which might explain
why the cauldron of regeneration comes from Ireland. It tells us little about
Llasars origins, but a detail in the Third Branch is more revealing. When
Pryderi, Manawydan, and their spouses go to England to look for work,
we are told that Manawydan began to fashion pommels, and they
were coloured in the way he had seen Llassar Llaes Gyfnewid do with
blue azure. This suggests that Llasar or Llassar is derived from Persian
lazhvard, which gives us the word azure and the stone (lapis) lazuli, raising
at least the possibility that the name Llasar/Llassar has an Iranian connec-
162
tion. This may be strengthened by the relationship between Brans

cauldron and that of Medea, with her origin in the Black Sea kingdom of
Colchis.
Much of the argument for eastern or central European origins has
been based on the connection between the horse-sacrice at Danebury and
the horse-goddess Rhiannon on the one hand, and the Homeric Rhesus
and the Thracian rider-god on the other. At this point we should also
mention the Vedic ritual known as the asvamedha or horse-sacrice. This
is described in detail in the Yajurveda (1400 B.C.1000 B.C.). The ceremony
could only be conducted by a king the horse (a stallion) was anointed
with ghee and decorated with golden ornaments; the horse was then
slaughtered, and three queens walked around the horse reciting mantras;
the chief queen then had to mimic copulation with the dead horse, while
the other queens ritually uttered obscenities. The chief queen spent the
night with the horse, and the next day the horse was ritually dissected and
roasted.
Reincarnation and the Cosmic Order

If the Mound-People seem to have Indo-Iranian ancestors, what of
the Druids? Their name is said to be Celtic and to mean oak-seers, but
other origins are possible. In Sanskrit druva means eternal, so they could
be knowers of the eternal; and druh, as we have just seen, means demon
or deception, so they could the demon-knowers or the deceptionknowers, able to see through all falsehood. However, the most compelling
etymology is one that interprets druid as doorkeeper, in the sense of one
who knows the door to the Otherworld, either the underworld that was
such a feature of religion in southern England, or the celestial Otherworld
of the star-god Arcturus. In fact, as The Spoils of Annwn shows, the Otherworld seems to be both underground (Caer Sidi) and in the heavens
(Caer Pedryuan), and the Druids held the key to the door of both these
realms. The association between the Druids and doors also casts new light
on the poem we discussed in Chapter 1, What man is the gatekeeper?,
which could be interpreted as Arthur seeking entry to the Otherworld (at
the very least, it seems that Arthur may be invisible at the beginning of
the poem30).
The Druids are famously said to have believed in reincarnation, and
in Chapter 6 we quoted Jan Bremmer as saying that the Pythagorean doc-
163
trine of reincarnation had Greek rather than Indian origins. However, a

somewhat different point of view is put forward by Thomas McEvilley.
He believes that there are so many similarities between the Greek and
Indian concepts of reincarnation that they must have been drawing on a
common source. For example, on the question of vegetarianism, he notes31
that Empedocles (c. 490430 B.C.), a Greek philosopher born in Sicily,
who was inuenced by the Pythagoreans, preached to meat-eaters: Do
you not see that you are devouring each other in the thoughtlessness of
your minds?; while in India, the mythical progenitor of mankind, Manu,
is supposed to have said: He whose esh I eat in this life, will devour
mine in the next.
McEvilley also believes that Greeks and Indians had a similar attitude
to what he calls32 the knowledge that frees. This knowledge is not the
ordinary discursive kind, but rather a kind by which the unhearable
becomes heard, the unperceivable becomes perceived, the unknowable
becomes known.33 The Orphics, says McEvilley,34 held that release is
obtained through recollection of ones own god-nature.... We may understand the cosmos not by merely beholding it, but by tuning ourselves
to its tuning. As Plato (c. 428c. 348 B.C.) says in Timaeus: The motions
akin to the divine part in us are the thoughts and revolutions of the universe; these, therefore, every man should follow, and correcting those
circuits in the head that were deranged at birth, by learning to know
the harmonies and revolutions of the world, he should bring the intelligent
part, according to its pristine nature, into the likeness of that which intelligence discerns. This doctrine goes back in some degree to Bronze
Age Mesopotamia, says McEvilley, where the trail of the idea of macrocosm/microcosm leads. By about 2000 B.C. in Mesopotamia,35 the universe
was seen as a single huge anthopomorphic being, as shown in a poem
relating the myth of the goddess Inanna. When her father Enki gives her
kingship, Inanna uses the imagery of the cosmic being to describe her
newly expanded state: The heavens he set as a crown on my head, the
earth he set as sandals on my feet. By the Late Bronze Age this mode of
imagery in which the elements are assigned to parts of an anthopomorphic
body had virtually replaced the old narrative mode in which individual
deities or heroes clashed about specic willed projects. The gods and goddesses began merging into a single cosmic being who bore within himself
or herself the old deities as aspects or parts. A Babylonian hymn from
around 1000 B.C. absorbs the pantheon into Marduk as his attributes or
aspects36:
164

Ninurta is Marduk of the hoe,
Nergal is Marduk of the attack,
Zababa Marduk of the hand-to-hand ght,
Enlil is Marduk of lordship and counsel,
Nabium is Marduk of accounting,
Sin is Marduk, the illuminator of the night,
Shamash is Marduk of justice,
Adad is Marduk of rains....
Other hymns identify the universe with the human body37:

O lord, your face is the sun god, your hair, Aya,
your eyes, O lord, are Enlil and Ninlil....
The appearance of your mouth, O lord, is Ishtar of the stars
Anu and Antum are your lips, your command ...
your teeth are the seven gods who lay low the evil ones.
Returning to Greek and Indian concepts of reincarnation, McEvilley

points out38 that in India there is the concept of jivanmukta, a human
who will not be reborn after the death of his present body-habitation.
Empedocles seems to have regarded himself as such a man, saying: I surpass mortal men, who are subject to many deaths.... I walk among men
as a god, no longer mortal I have been set free. It was also said of
Pythagoras that he belonged to a class between gods and humans, meaning perhaps that he would not be born again as a mortal but as an immortal, a released soul. In Indian tradition the jivanmukta, before passing out
of the body forever, would use their last incarnation to teach others how
to achieve the same status.
Special powers are attributed to the released soul. In Greece the lore
about early philosophers called theologoi, and about the Orphic movement
in general, is full of stories of teaching masters who worked in part through
super-powers (e.g. Abaris, Aristeas, the miracle-workers discussed by
Bremmer). Released souls were credited with special abilities, often
including recognition of past incarnations. Porphry says of Pythagoras:
When he strained with all his mind [says Empedocles], he could easily
see everything there is in ten, yes even twenty human lifetimes.
McEvilley reports39 that the philosopher Xenophanes, in a fragment
written about 525 B.C., says of Pythagoras: Once they say that he was
passing by when a puppy was being whipped and he took pity and said:
Stop dont beat that animal; it is the soul of a friend. I recognized its
voice when it cried out. McEvilly notes that this voice-recognition has
resonances in shamanic tradition round the world, as well as in the yogic
165
tradition of India; in both cases the spiritually or magically advanced personality is believed to understand the speech of animals and birds.
Pythagoras is often linked to Orpheus, and it is widely believed, says
McEvilley,40 that Orpheus represents in part Greek contact with a tradition
of ThracoScythian shamanism his famous descent to the underworld
to bring back Eurydice answers to the shamans role as psychopomp.
We know next to nothing, of course, about how the Druids viewed
reincarnation. In the poems attributed to Taliesin, Angar Kyfyndawt (Cruel
Bondage or Hostile Alliance),41 and Kat Godeu (The Battle of the Trees), the
poet appears to be alluding to a series of reincarnations42:
I was a blue salmon,
I was a dog, I was a stag,
I was a buck on the mountain,
I was a trunk, I was a spade (?)
I was a drinking horn in the hand,
I was a peg in forceps
for a year and a half
I was a speckled white rooster
among chickens in Edinburgh.
I
I
I
I
I
I
have been a course, I have been an eagle.

have been a coracle in the seas:
have been compliant in the banquet.
have been a drop in a shower;
have been a sword in the grasp of the hand
have been a shield in battle.
However, the fact that he has been reincarnated not only as living beings
but also as inanimate objects suggests something closer to the Babylonian
hymns, and implies that the poet is identifying himself with the cosmos,
or with aspects of the one cosmic being. A comparison could also be made
with the Bhagavadgita (c. 500 B.C.), in which the god Krishna says to
Arjuna, one of the heroes of the Mahabharata43:
I am the avor in the waters, Kaunteya;
I am the radiance of the moon and sun,
the sacred word in all the Vedas,
the sound in the air, the virility in men,
and the pure fragrance on the earth;
and I am the brilliance in ame,
the life in all beings,
and I am the austerity in ascetics.
Know me as the primeval seed of all beings, Partha.
I am the intuition of the intelligent;
the brilliance of the brilliant am I.
166
The Druids of course conducted their ceremonies in nemeton, which

as Jane Webster says44 is commonly glossed as sacred grove, though it
may also have designated a small shrine or chapel. The word is attested
mainly in place-names, for example Augustonemeton and Nemetodorum
in Gaul, Nemetobriga in Spain, and Medionemeton in Britain. The origin
of the word nemeton is unclear. In Old Irish, says Koch,45 nemed is an
important sociolegal term meaning privileged person, dignitary, professional, sacred place, land owned by a privileged person, sanctuary, privilege; in Early Welsh the cognate word nyfed occurs in the heroic elegies
of the Gododdin in the phrase molut nivet, to be understood as the [poetic]
praise of dignity, rightful privilege, privileged places, persons. Therefore
one possible etymology is the Indo-European root nem- to bend, bow,46
which gives Old Indian namaiti to bend, bow, namas bow, obeisance,
Avestan nemah, nemaiti, bend, bow, Latin nemus grove. Here again we
have a possible link between a Druid term and both Sanskrit and Avestan.
Conclusion
But how does this story of Indo-Iranian ancestry t in with the priests
of Dodona, with the star Arcturus, later Arthur, with Arcas, the legendary
founder of Arcadia, with Lugh and Balor, and Culhwch and Ysbaddaden?
The answer must lie in the civilization of Mycenae, which dates from
around 1600 B.C., reached its peak around 1400 B.C ., and collapsed around
1100 B.C. We know that the Myceneans worshipped a horse-goddess, poti-ni-ya i-que-ya, or Potnia Ikkweia Mistress of Horses, who in classical
Greece became Poseidon Hippios and/or Hera Hippia/Athena Hippia. We
also know that the sacred site of Dodona was in use during the Mycenean
period. It is also clear that by the time of Hesiod, Arcturus had long been
known as a star by which agricultural activities could be regulated, and
the myth of Arcas and Callisto was well established. As for the Iranianspeaking peoples of the Black Sea and Danube area, we have already noted
that Homer refers to the Cimmerians, and there is a passage in the Iliad,
Book 13, where Homer shows further knowledge of the Black Sea area47:
Now when Jove had thus brought Hector and the Trojans to the ships,
he left them to their never-ending toil, and turned his keen eyes away,
looking elsewhither towards the horse-breeders of Thrace, the Mysians,
ghters at close quarters, the noble Hippemolgi who live on milk, and the
167
Abians, justest of mankind. The Hippemolgi or Horse-Milkers who live

on milk are generally thought to be the Scythians, as are the Abians. The
Mysians are said to have fought in the Trojan War as allies of Troy, and to
have settled in Mysia, in northwest Anatolia. They spoke a language which
could have been related to Hittite, or could have been a form of Paionian,
a language spoken in Thrace. Strabo says48 that the Mysians, in accordance
with their religion ... abstain from eating any living thing, and therefore
from their ocks as well; and that they use as food honey and milk and
cheese.
The origin of the Mysians is unclear. Edwards49 notes that the Hittite
Empire collapsed around 1200 B.C., and When the curtain rises again,
central Anatolia is ruled (or at least occupied) by an invading people, a
horse-rearing military aristocracy called the Phrygians. According to the
traditions preserved among the Macedonians, says Herodotus,50 the Phrygians crossed the straits into Anatolia from Macedonia and Thrace, where
they had until then been known as the Bryges or Briges. The Greeks in
general, says Edwards, believed that this event took place before the Trojan
war, enshrining it in legend; though Xanthus, a Lydian historian, held it
took place after that event, in a joint invasion with the Mysians. Excavations in Troy VIIb have revealed the introduction, after the destruction
of the city by re, of a new population using a coarse ware apparently of
central European origin, and this may reasonably be held to mark the passage of the Phrygians and Mysians.51
All the evidence points to the fact that at some time between 1100
B.C. and 800 B.C., a group of people either speaking an Iranian language
or inuenced by Iranian speakers, and living close enough to the Mycenean
Greeks to have assimilated aspects of their culture, migrated to Britain,
landing somewhere in the neighborhood of Totnes and Dartmoor. It is
not possible to say whether these immigrants were responsible for the
ranch boundaries or linears that were constructed as early as 1100 B.C.
and well into the beginning of the 1st millennium, but they would certainly
have been appropriate to people whose lives were focused on horses and
who kept herds of sheep, goats and horses.52 It is likely that these people,
or people related to them, brought the All Cannings Cross Pottery to Wiltshire, and constructed the huge middens at Potterne and East Chisenbury.
Chapter 8
Arthur, King of Wessex?

Ambrosius and Vortigern
The most puzzling thing about Arthur is that he not quite a gure
of myth, but not quite a gure of history. There is no mention of a divine
Arcturus or Arthur in Britain during Roman times, which suggests that
he is a postRoman historical gure. On the other hand, there is no mention of an historical Arthur until A.D. 830, when Nennius has him ghting
battles against the Saxons. In Chapter 1 we linked Arthur via his father
Uther Pendragon to the comet of A.D. 539 and the subsequent plague, and
I will now pursue this theme a little further.
Much of our knowledge of British history immediately after the
Roman army left Britain is due to a British monk, Gildas, who wrote a
long sermon entitled De Excidio et Conquestu Britanniae (On the Ruin
and Conquest of Britain). The dates of Gildas and his work are much
debated Thomas D. OSullivan concludes1 that the sermon was probably
written between A.D. 515 and A.D. 530, while Higham2 dates it to no later
than A.D. 485.
The work itself does seem to provide clues in a number of well-known
passages. In this rst passage, the Britons appeal to Rome for help against
the Picts and Scots3: The wretched remnant, sending to Aetius, a powerful
Roman citizen, address him as follows:To Aetius, now consul for the
third time: the groans of the Britons. And again a little further thus:
The barbarians drive us to the sea; the sea throws us back on the barbarians: thus two modes of death await us, we are either slain or drowned.
The Romans, however, could not assist them. Since the Romans could
not help them, an alternative had to be found: Then all the councillors,
together with that proud tyrant Gurthrigern [Vortigern], the British king,
were so blinded, that, as a protection to their country, they sealed its doom
by inviting in among them (like wolves into the sheep-fold), the erce and
168
8 Arthur, King of Wessex?
169
impious Saxons, a race hateful both to God and men, to repel the invasions
of the northern nations.
Eventually these impious Saxons overran the country, until the
British
took arms under the conduct of Ambrosius Aurelianus, a modest man,
who of all the Roman nation was then alone in the confusion of this troubled period by chance left alive. His parents, who for their merit were
adorned with the purple, had been slain in these same broils, and now his
progeny in these our days, although shamefully degenerated from the worthiness of their ancestors, provoke to battle their cruel conquerors, and by
the goodness of our Lord obtain the victory.
After this, sometimes our countrymen, sometimes the enemy, won the
eld, to the end that our Lord might this land try after his accustomed
manner these his Israelites, whether they loved him or not, until the year
of the siege of Mons Badonicus, when took place also the last almost,
though not the least slaughter of our cruel foes, which was (as I am sure)
forty-four years and one month after the landing of the Saxons, and also
the time of my own nativity. And yet neither to this day are the cities of
our country inhabited as before, but being forsaken and overthrown, still
lie desolate.
Later in his sermon, Gildas castigates ve British whom he considers cruel,

rapacious and living a life of sin: Constantine of Dumnonia, Aurelius Caninus, Vortiporius of the Demetae (Dyfed), Cuneglasus, and Maglocunus,
or Maelgwn.
The rst clue is provided by the appeal to Aetius, a Roman general
who fought a number of campaigns in Gaul against Burgundians, Visigoths
and Alans, and who was consul three times, the third time in A.D. 446.
The second clue is provided by the name Vortigern. It is not clear whether
Gildas actually used the name, but Bede, writing in A.D. 730, did refer to
Vortigern, and he was presumably using Gildas as his source for the period.
The name Vortigern is usually seen as a title rather than a name, meaning
in Celtic something like Over-Lord in fact, Gildas calls him in Latin
superbus tyrannus (proud tyrant), which may be a play on the name Vortigern. According to the genealogies of Powys, he was king of Powys from
around 420 or 425, and was probably based in the Romano-British Viroconium (Wroxeter in Shropshire) (see Map 12): excavations have shown,
says Castleden,4 that after the collapse of Roman control Viroconium was
substantially rebuilt, timber buildings replacing stone: ... big elaborate
buildings of classical design, with symmetrical facades, colonnades and
often a second storey. The central focus of Viroconium was a massive
170
winged building raised on the site of the Roman basilica; it was a pseudoclassical building with a clutch of ancillary buildings and outhouses, and
it seems likely that this was Vortigerns palace.5
The third clue is provided by the name Ambrosius Aurelianus, whose
parents were adorned with purple. Higham says 6 it is possible that
Ambrosius was a distant relative of one of the imperial families sprung
from the west, such as the house of Theodosius (A.D. 379395) which
had branches ourishing in Spain, at least until the campaigns of Constantine III (A.D. 407411). Castleden sees Ambrosius as a Celtic aristocrat,
and argues7 that the Ambros place-names
may represent the stations of units raised and led by Ambrosius and styled
Ambrosiaci. The location of these garrisons is signicant. One group
seems to surround and protect the territory of the Dobunni (known as
Calchvynydd Chalk-Hill in the dark ages). A second group of
Ambrosiaci was positioned in the Lee and Stort valleys, well located to
defend London, then a British enclave, against attack by the East Anglians
along the Roman roads from Cambridge and Colchester.
Myres, says Castleden,8 proposes that Ambrosius owned land at Amesbury,

which was originally Ambresbyrig.
Frank D. Reno9 links Ambrosius to Riothamus, mentioned by the
Roman historian Jordanes in his history of the Goths, written around A.D.
551, as king of the Brittones. According to Jordanes, Riothamus fought
on the side of the Romans against the Visigoths led by Euric, and was
defeated in around A.D. 470. Reno argues that Riothamus then went to
Britain around 475, where he became known as Arthus (from the Welsh
arddus, meaning high10), and that Arthus is an epithet of Ambrosius
Aurelianus.
Ambrosius Aurelianus is a little like Arthur, in that it is difcult to
nd him in the historical record. His name means literally Golden Immortal, and it is tempting to link him to Arthur, though for different reasons
from those proposed by Reno. Arcturus is a prominent star in the constellation Botes, and it would be appropriate to give Arthur/Arcturus the
title of Golden Immortal. The suggested link between Ambrosius and
Amesbury may also be signicant, providing a possible link between
Ambrosius and one of the families that ruled over Salisbury Plain and surrounding areas for so long before the coming of the Romans.
This interpretation of Ambrosius is reinforced by Historia Brittonum,
which seems to be drawing on at least two set of legends about Ambrosius.
In Chapter 31, Nennius tells us11 that while Vortigern was king, the British
171
had cause of dread, not only from the inroads of the Scots and Picts, but
also from the Romans, and their apprehensions of Ambrosius. In Chapter
42 we are told that the boy without a father, who told Vortigern about
the dragons ghting under the site of his citadel (see Chapter 6), was called
Ambrose. In Chapter 48, Nennius reports that Pascent, son of Vortigern,
became ruler of Builth and Gwrtheyrnion, which were granted him by
Ambrosius, who was the great king among the kings of Britain. Finally
in Chapter 66, there is an enigmatic piece of chronology: And from the
reign of Vortigern to the quarrel between Guitolinus and Ambrosius, are
twelve years, which is Guolopum, that is Catgwaloph. Guitolinus (Latin
Vitalinus) is also mentioned in Chapter 48 as the son of Gloui (Gloucester)
and one of Vortigerns ancestors; Guolopum and Catgwaloph are thought
to refer to Wallop and the battle of Wallop, now a village in western Hampshire, 9 miles southwest of Amesbury, and not far from Danebury hillfort
(see Map 13). All these details from Nennius suggest that Ambrosius was
based in Wiltshire or Hampshire, may have fought a battle with Vortigern,
and like Arthur, had become part historical gure, part wonder-worker.
The last clue provided by Gildas is the battle, or siege, of Mons
Badonicus (Mount Badon). It has proved extremely difcult to establish
where this battle or siege took place: candidates include Badbury Hill,
near Faringdon in Oxfordshire; Liddington Castle above the village of
Badbury, near Swindon in Wiltshire; Badbury Hillfort, also known as
Badbury Rings, near Wimborne Minster in Dorset (though this was not
really on the frontline between the Saxons of eastern England and the
British of western England); and Solsbury Hill, near Bath (though it is
unlikely that Gildas would refer to Aquae Sulis by its later Germanic name).
The date is also unknown, with estimates ranging from around A.D. 480
to around A.D. 520 given OSullivans dating of the De Excidio to the
520s, then a date of around 480 seems more likely.
Not surprisingly, perhaps, the Anglo-Saxon Chronicle, written around
A.D. 891, is silent on this battle or siege. The only evidence it provides is
negative evidence: the Chronicle gives a list of Bretwaldas Saxon rulers
who had achieved overlordship over some or all the other Anglo-Saxon
kingdoms and there is a gap between Aelle of Sussex, whose dates are
given as 488514, and Ceawlin of Wessex, who is said to have been Bretwalda from 562 to 593. These dates are highly suspect, but they do show
a gap of almost fty years during which Saxon expansion was apparently
halted.
Lets assume for the sake of argument that the De Excidio was written
172
Map 12: Some signicant British strongholds in England between A.D. 400
and A.D. 500, and possible sites of battles between British and Saxons, showing
modern Wiltshire, Somerset, Dorset, and Oxfordshire.
around A.D. 520 and that the battle/siege of Mons Badonicus took place
around A.D. 480. What was life like then in southwestern Britain? It is
almost certain that most of the Romano-British towns were abandoned by
about A.D. 450 or perhaps a little later, and that some people at least had
moved into hillforts. This is underlined by the history of Cadbury Castle
in south Somerset, 5 miles northeast of Yeovil. It is not far from the
Romano-British town of Lindinis (Ilchester), one of the capitals of the
Durotriges. Cadbury Castle, says Costen, was refortied in the second half
of the 5th century, and continued to be used into the 6th century.12 The
scale of the refortication is such that it cannot be compared with any
other site in the region. It was refortied so that the 3,940-ft perimeter
enclosed 18 acres of land; the new rampart used over 20,000 meters of
timber. The presence of a large hall on the hilltop is good evidence of
occupation by individuals of high status, as is the presence of imported
173
Mediterranean pottery: It seems unlikely that Cadbury Castle would have

been reoccupied and the walls rebuilt so lavishly unless it was used at some
point by military forces. The site looks eastward, and guards access to
Somerset from the southeast and east, acting both as a bulwark against
advance from the east and as a dominant site for the central Somerset
basin. As such it is the successor to Ilchester as the focal point for commerce
and politics in the district.
Cadbury Castle was not the only site in Somerset to be refortied
after the Roman withdrawal. According to Barbara Yorke, Cadbury Congresbury, also known as Cadbury Hill, in north Somerset, near Yatton,
had part of its Iron Age perimeter refurbished in the subRoman period.13
Cadbury Congresbury has produced more imported pottery than any
other English site, apart from Tintagel in Cornwall, as well as evidence of
ne metalworking and substantial feasting. Cannington hillfort, near
Bridgwater in Somerset, also seems to have been reoccupied in the post
Roman period and a substantial cemetery in use at the same time has been
excavated in its vicinity.14
It is possible that the authority responsible for the refurbishment of
Cadbury Castle and other hillforts in the area may have been responsible
for the erection of the earthwork, West Wansdyke, which, says Yorke,
consists of a discontinuous north-facing bank and ditch running parallel
and to the south of the River Avon, from Maes Knoll, south of Bristol, to
just south of Bath.15 It was probably built in the late 5th and 6th centuries,
as a reaction to the extension of Saxon settlement into the Kennet Valley
in the late 5th century.
So Somerset was clearly in British hands at the time of Gildas, but
what of areas adjoining Somerset? Excavations have suggested, says Yorke,
that Bokerley Dyke on the easterly edge of Cranborne Chase in Dorset
was built in the postRoman period when it would have controlled access
to the Dorchester area along the Roman road of Akling Dyke. A substantial
part of the land between Bokerley Dyke and West Wansdyke was occupied
by the formidable natural obstacle of Selwood Forest,16 and Yorke speculates17 that Selwood, West Wansdyke and Bokerley Dyke may represent
a negotiated frontier between Britons and Saxons.
West Wansdyke is not the only earthwork from this period. East
Wansdyke ran for some 15 miles from Morgans Hill near Devizes, across
the Marlborough Downs to just west of Savernake, in east Wiltshire, and
so separated the Kennet Valley from the rest of Wiltshire. East and West
Wansdyke used to be considered as one monument, but there is no trace
174
Map 13: British fortications in Somerset and Wiltshire between A.D. 400
and A.D. 550, and some sites associated with Ambrosius Aurelianus.
of a bank or ditch running between them, although they are linked by the
Roman road from Silchester to Bath.18
Unlike Somerset, there is little evidence of widespread reoccupation
of hillforts in the area to the south of East Wansdyke. Excavation of a hillfort at Whitsbury Castle, near Fordingbridge in East Hampshire, 10 km
south of Salisbury, produced positive evidence for the refurbishment of
its defences.19 There is also evidence for hillforts becoming estate centers
in the early Middle Ages. Margaret Gelling, says Yorke,20 has discussed
the examples of scesbyrig (Ufngton Castle), now in Oxfordshire but formerly in Berkshire, and Blewbury (Berkshire).
Gelling notes21 that there are two types of place-names: habitative,
175
which contain a word for settlement, and topographical, which make

a statement about the setting of the village but do not mention buildings.
Topographical names, says Gelling,
may be specially characteristic of areas where the English are known to
have established themselves at a particularly early date, and in such areas
the more important settlements, the ones which became the centers of
large estates, are more likely to have topographical names than habitative
ones. Lambourn (lamb river), Blewbury (variegated hill-fort), and Faringdon (fern down) are examples of such estate-centres in Berkshire.
This is a county where documentary records from before the Norman
Conquest are particularly rich, and with the help of these it can be shown
that habitative names some times replace topographical ones in the 10th
century or later.
And Gelling gives the example a large estate called scesbyrig (Ash-tree
fortication), which was split into two estates called Ufngton (Uffas
farm/settlement) and Woolstone (Wulfrics farm/settlement).
Yorke believes that Amesbury, Malmesbury and Old Sarum (Searoburh)
are probably similar examples from Wiltshire. Old Sarum in Roman times
was the fort of Sorviodunum, and there is archeological evidence of a
Romano-British village in and around Stratford Sub Castle, next to Old
Sarum,22 so it is possible that the hillfort was occupied in the 5th century.
Amesbury may have been named after Ambrosus Aurelianus, who
could have occupied Vespasians Camp in the 5th century. Certainly there
was a Romano-British village at Buttereld Down on the outskirts of
Amesbury, where a hoard of late Roman gold coins was found, and a cemetery nearby, including at least one high-status grave.23 Moreover, as
McOmish et al. have shown,24 during Roman times there were at least
eleven ourishing villages on what is now Salisbury Plain Training Area,
one of which, on Charlton Down, covered over 25 ha. The range of artefacts from the sites, they say, points to established sedentary agricultural
communities.
What does all this tell us about Ambrosius and Vortigern? The evidence and that is exceedingly slim is that Vortigern was an early 5thcentury ruler, possibly ruling from Viroconium, who enlisted the Saxons
as allies against the Irish and Picts. Ambrosius was a later ruler from the
period A.D. 450 to A.D. 500 who in some sense succeeded Vortigern and,
from a base in Wiltshire, rallied the British against the Saxons. As Reno
points out,25 it is possible that Ambrosius is the gure who later became
known as Arthur, but it is equally possible that the story of Arcturus
involves another much more surprising gure.
176
Caratacus and Cunorix

As we saw in Chapter 1, Europe witnessed a spectacular comet in
A . D. 539, and this was followed a few years later by an outbreak of
the plague. At this time, large areas of western England, including Somerset, Dorset, Devon and Cornwall, and north of the Thames, Gloucestershire, Herefordshire, Shropshire and northwest England were still
controlled by the British, and eastern England, including Essex, East
Anglia, Sussex and the Isle of Wight were under the control of Angles,
Saxons or Jutes.
Throughout this book we have focused on Salisbury Plain and its
surroundings, in particular the Marlborough Downs and west Hampshire,
and it is plain that from A.D. 500, if not earlier, this area was on the front
line between the British and the Saxons, as the construction of East Wansdyke demonstrates. But what actually happened there, in the struggle for
the ancient heart of southern England? The best place to start is the AngloSaxon Chronicle, which reports the following events26:
A .D. 495. This year came two leaders (aldormen) into Britain, Cerdic
and Cynric his son, with ve ships, at a place that is called Cerdics-ore.
And they fought with the Welsh the same day.
A .D. 508. This year Cerdic and Cynric slew a British king, whose name
was Natanleod, and ve thousand men with him. After this was the land
named Netley, from him, as far as Charford.
A .D. 514. This year came the West-Saxons into Britain, with three ships,
at the place that is called Cerdics-ore. And Stuff and Wihtgar fought with
the Britons, and put them to ight.
A .D. 519. This year Cerdic and Cynric undertook the government of the
West-Saxons; the same year they fought with the Britons at a place now
called Charford. From that day have reigned the children of the WestSaxon kings.
A .D. 527. This year Cerdic and Cynric fought with the Britons in the
place that is called Cerdics-ley.
A .D. 530. This year Cerdic and Cynric took the Isle of Wight, and slew
many men in Carisbrook.
A .D. 534. This year died Cerdic, the rst king of the West-Saxons. Cynric his son succeeded to the government, and reigned afterwards twenty-six
winters. And they gave to their two nephews, Stuff and Wihtgar, the
whole of the Isle of Wight.
A .D. 552. This year Cynric fought with the Britons on the spot that is
called Sarum, and put them to ight.
A .D. 556. This year Cynric and Ceawlin fought with the Britons at
Beranbury.
177
Netley is on the south coast of Hampshire, near Southampton, while Charford is in Hampshire, near Fordingbridge and south of Salisbury. Cerdic
and Cynric are said to be Saxons but, as Yorke points out,27 both the Isle
of Wight and the Hampshire coast were settled by Jutes. Bede says that
the Hampshire coast was still known in his time as the Iutarum natio
nation of the Jutes, and place-names survive identifying various southern
Hampshire locations as Jutish, including the New Forest (Ytene), Bishopstoke (Ytingstoc) on the River Itchen, a few miles south from Winchester,
and a valley near East Meon (Ytedene).
In that case we can discount the early details, except the assertion
that in A.D. 519 Cerdic and Cynric undertook the government of the West
Saxons, and that their descendants had ruled the West Saxons ever since.
But who were Cerdic and Cynric, and if they did not rule in south Hampshire where did they rule? The Anglo-Saxon Chronicle gives a clue in its
entry for A.D. 635, when it tells us that one of the descendants of Cerdic
and Cynric, Cynegils, was baptized by Bishop Birinus at Dorchester [see
Map 14]; and Oswald, king of the Northumbrians, was his sponsor. Bede
gives us another clue when he says (Book 3, Chapter 7) that the West Saxons in early days were called the Gewisse.28 And there is another clue:
Cerdic and Cynric are, or could be interpreted as, British names.
The simplest of these clues is the name, Gewisse. This Germanic word
is usually translated as the the sure ones, the reliable ones, implying
they were people the Anglo-Saxons could trust. However, Hoops
and Beck29 link the word to Gothic gawidan, join, and Old High German
(gi-)wetan to bind, suggesting that Gewisse meant confederation
or confederates, and implying something similar to the foederati of the
late Roman Empire entire tribes paid, in cash or in kind, to ght for
Rome.
The matter of Dorchester, that is Dorchester-on-Thames, Oxfordshire, is also fairly straightforward. Dorchester was originally a Roman
town called Durocina. In the late Roman period, says Daniel G. Russo,30
Dorchester was not a civitas-capital but rather a small town or vicus (c.
14 acres) which had likely grown from minor civilian settlements (canabae)
attached to an earlier legionary fortress. Nonetheless it was strategically
important enough to receive stonework defenses during the third century,
and it seems to have had some local administrative and commercial prominence as well. Excavations have shown that life continued in Dorchester
into the early 5th century. Evidence of this subRoman occupation
includes two-storey timber buildings on stone foundations and quantities
of Theodosian-period coinage. One of the towns late Roman cemeteries
178

to the north was found to have been in use in the early 5th century but
not beyond.
Recent research suggests the Romano-British town was virtually deserted

by A.D. 450.
As postRoman Dorchester declined, Saxon inuence grew. In fact,
as Yorke points out,31 Dorchester provides some of the earliest evidence of
Anglo-Saxon settlement:
Three early Germanic burials have been found near Dorchester-onThames, a man and a woman at Dyke Hills and a woman at Minchin
recreation ground. The man was buried with the most complete late
Roman ofcial belt-set so far discovered in England [and] the belt-set is
probably of early 5th century date, thus making the Dorchester man a
candidate for a Germanic soldier in the employ of a British rather than
Roman authority....
In the vicinity of Dorchester a number of cremation and inhumation

cemeteries seem to have been established on both banks of the upper
Thames and its tributaries in the rst half of the fth century. By the end
of the fth century settlement may have spread to the Vale of the White
Horse and the Kennet Valley, and it could have been this expansion which
provoked the building of one or both of the Wansdykes.
According to Russo,32 concentrated settlement in and around Dorchester
did not commence, however, until the late sixth century. The context for
this intensication seems to have been the campaigns of the West Saxon
Bretwalda Ceawlin ... against the subRoman enclaves in the Chiltern
and Severn valleys.... Indeed, there are strong indications that Dorchester
had become the headquarters of some prominent Anglo-Saxon ruler or
leading family at this time. Excavations within the town disclosed Byzantine and early Anglo-Saxon gold coinage, along with gold ornaments and
personal jewelry dated c. A.D. 600625. Another princely or royal site
of the early Thames Valley Saxons has been identied at Cuddesdon,
some 6 miles north of Dorchester. The unusually rich nds there were
contemporary with those within Dorchester and probably represent a
rural villa regalis (royal estate centre) of the same family. It has also been
observed correctly that some place in this region, probably Dorchester,
was serving as a central, perhaps royally controlled redistribution or production centre for early Anglo-Saxon prestige items and raw materials.
The sum of the archaeological evidence thus strongly points to Dorchester
as an important multifunctional central place of the Thames Valley Saxons
and a likely headquarters of their kings by the opening of the seventh century.
179
Map 14: Sites associated with the West Saxon Gewisse between A.D. 550 and
A.D. 660, in Wiltshire, Hampshire, Somerset, Dorset, Devon, Gloucestershire
and Oxfordshire.
The most perplexing aspect of Cerdic and Cynric is their apparently

British names. Cerdic appears to be the Anglo-Saxon rendering of the
Welsh Ceredig, itself a later version of Caratacus, the leader of the Catuvellauni who resisted the Roman invasion of Britain. There were several
people of this name in Britain at the time: Matthews33 mentions the British
king Coroticus addressed by Saint Patrick in an open letter (St. Patrick
excommunicated some of Coroticuss soldiers for taking some of St.
Patricks converts into slavery), and Ceredig, the last king of Elmet, a
British kingdom in west Yorkshire; Reno34 mentions Ceretic, the fth son
of Cunedda, the founder of the Welsh kingdom of Gwynedd; and Ceretic,
described in Historia Brittonum Chapter 37 as the interpreter for Hengist
and Vortigern in their negotiations over Hengists beautiful daughter. Reno
thinks35 that Cerdic and the interpreter Ceretic are the same individual,
though their dates appear to be different.
180
Cynric could be a Saxon name, meaning Kin-Ruler, but it is more

likely to be British and mean Hound-King, like the god of Nettleton
Shrub Roman temple, Cunomaglus or Hound-Lord. Matthews links it
to an Irish name Cunorix found at Wroxeter, but given the long association
between southern England (especially Wiltshire) and otherworldly dogs,
I see Cunorix as more probably a native Briton. The possibility that Cerdic
and Cynric were British may be subtly hinted at in the Chronicle when it
describes them as aldormen (high-ranking royal ofcials) perhaps an
admission that they were high-born British working to further the Saxon
cause.
A Tale of Two Hillforts

The Anglo-Saxon Chronicle has Cerdic and Cynric doing nothing
much in southern Hampshire until A.D. 534, and Cynric doing the same
until A.D. 552. The Chronicle says that Cerdic assumed the kingship in
A.D. 519, tough Dumville36 has calculated that Cerdic became king somewhat later, around A.D. 538. This period of apparent inactivity may be
the Chronicles way of dealing with the long pause in Anglo-Saxon expansion westward after the battle of Mons Badonicus. Interestingly, Cerdic
may have become ruler around the time of the comet, and the end of the
pause may have come after the outbreak of plague that followed the comet.
It is difcult to reach any conclusions about Cerdic because the only
activities attributed to him are ones which seem historically implausible.
However, he must have enjoyed great prestige among the Gewisse/West
Saxons who succeeded him, since they were all keen to claim him as their
ancestor. The most likely explanation is that he was a British aristocrat
from somewhere in the region of Dorchester-on-Thames who spoke both
Celtic and Saxon. He may have been a descendant of the Ceretic who was
Vortigerns translator, or he may have had some connection with the court
of Powys, which by this time is thought to have abandoned Viroconium/
Wroxeter and moved to Pengwern (possibly Shrewsbury).
More can be said about Cynric, however, because he is credited with
some very specic deeds. In A.D. 552 he is said to have captured Old Sarum
(see Map 14), though as Yorke points out,37 the area had already been
settled by Saxons: the Anglo-Saxon cemetery at Harnham Hill, Salisbury,
has burials going back to around A.D. 500; the graves at Petersnger, just
to the south of Salisbury, date from the 5th century; the cemetery at Win-
181
terbourne Gunner, northeast of Salisbury, has graves from the 5th and 6th
centuries; and the Anglo-Saxon graves at Collingbourne Ducis, on Salisbury Plain, date from a round A.D. 500.
Then in A.D. 556 he is said to have captured Beranbury (Barbury
Castle) along with Ceawlin, of which more later. Barbury Castle is an Iron
Age hillfort near Wroughton in northeast Wiltshire, just south of Swindon.
It is on the Ridgeway, an ancient trackway from Buckinhamshire to the
Kennet Valley, which passed near Dorchester-on-Thames control of Barbury Castle would have secured the route from Oxfordshire to Wiltshire
for the Gewisse.
This implies that Cynric and the Wiltshire Gewisse had a base somewhere in north Wiltshire, and one possibility for this base is Ramsbury,
in northeast Wiltshire, near the border with Berkshire, 7 miles south of
Barbury Castle. Ramsbury was made a bishopric in 909, and this suggests,
says Haslam,38 that Ramsbury was at the time a villa regalis. Ramsbury is
only 4 miles from the Roman fortied vicus of Cunetio, which was the
successor to the Forest Hill Farm hillfort. Cunetio appears to have been
refortied in A.D. 367. The concentration of Roman villas around Cunetio
marks its immediate environs as being a highly organized agricultural
region, comparable to the environs of Bath and to Cirencester.39
Haslam argues,40 following the lead of Cunliffe, that during the late
Roman period, some villa estates increased in size at the expense of others,
and resources became concentrated at a few estate centers. What followed
after the withdrawal of Roman forces is suggested by a model for the transition of Roman to Anglo-Saxon Winchester put forward by Biddle,41 who
has suggested that
a ruling element which emerged from the mercenary presence in Winchester in the late 4th century assumed power and territorial control from the
last remnants of the Romano-British administration, supplanting the social
order which it had been their rst duty to defend. The nd of a military
belt buckle of Hawkess type IIA at Cunetio might suggest that this town
could also have survived through the support of mercenaries (whatever
their precise origins) as some sort of political focus after the general collapse of the Roman industrial economy.
Haslam believes that the large estate centers survived into the 6th
century, and would have been the natural focus for anybody wishing to
establish a military presence in the area. He argues42 that
the proximity of Ramsbury to Cunetio, the presence there of probable
Roman villa and presumably a late Roman estate centre, and its position in
182

the probable avenue of Saxon penetration up the Kennet valley, all suggest
that it could well have become the focus of the area in succession to Cunetio, and could have taken on some of the administrative functions of the
former late Roman and subRoman town, subsequently becoming a villa
regalis on the consolidation of the West Saxon kingdom.
The probability is strengthened, says Haslam,43 by the name Ramsbury,

where the burh- element means fortied dwelling rather than hillfort
(the nearest hillfort is Membury, 5 km to the northwest), and the whole
name means Fortied Dwelling of the Raven (which has intriguing associations with the Wiltshire past and British mythology).
North and South

Let us now return to the Anglo-Saxon Chronicle and continue the
story of the Gewisse:
A .D. 560. This year Ceawlin undertook the government of the WestSaxons.
A .D. 568. This year Ceawlin, and Cutha the brother of Ceawlin, fought
with Ethelbert, and pursued him into Kent. And they slew two aldermen
at Wibbandun, Oslake and Cnebba.
A .D. 571. This year Cuthulf fought with the Britons at Bedford, and
took four towns, Lenbury, Aylesbury, Benson, and Ensham. And this same
year he died.
A .D. 577. This year Cuthwin and Ceawlin fought with the Britons, and
slew three kings, Commail, and Condida, and Farinmail, on the spot that
is called Derham, and took from them three cities, Gloucester, Cirencester, and Bath.
A .D. 584. This year Ceawlin and Cutha fought with the Britons on the
spot that is called Fethan leag. There Cutha was slain. And Ceawlin took
many towns, as well as immense booty and wealth. He then retreated to
his own people.
A .D. 591. This year there was a great slaughter of Britons at Wodens
Barrow; Ceawlin was driven from his kingdom, and Ceol reigned six years.
Ceawlin is said to have become king of the Gewisse and Bretwalda in 560,
though the actual date is more likely to be around 580. Ceawlin is likely
to have been another Briton, but like Cerdic and Cynric, his origins are
obscure. One possibility is that the name is related to the British name
Coel, as in the Welsh Coel Hen, the Old King Cole of nursery-rhyme
fame. Koch44 notes that Coel is said to be the ancestor of many early
medieval north British rulers, known collectively as the Coeling. They are
183
associated with Rheged (Cumbria) and Elmet (Yorkshire) though one of

the Coeling, Llywarch Hen, is said to have ed south to Powys, in Shropshire. Welsh coel as a common noun means belief or omen; Old Irish cel
means auspicious.
The chances are, then, that Ceawlin, like Cerdic, originated in or
had links to the Midlands or the north of England. It is perhaps signicant
that, unlike Cynric, his battles all occurred north of London, and were all
fought alongside a supposed relative or relatives known variously as Cutha,
Cuthwine or Cuthwulf. Their names suggest that they were Saxons, derived
from cu well-known, wine friend, wulf wolf, and therefore allies of
the Dorchester-on-Thames rulers.
Of the battles he fought, the rst, against Aethelberht, king of Kent,
was presumably a border dispute; the location of Wibbandun is unclear
(the supposition that it was Wimbledon is now rejected). The battle at
Bedford is puzzling, since this area is thought to have already been in the
hands of the Angles of East Anglia. The most interesting of the battles is
the one at Dyrham in which Cuthwin and Ceawlin captured Cirencester,
Gloucester and Bath and killed three kings, Conmail, Condida, and Farinmail. Patrick Sims-Williams45 believes these names represent Welsh Cynfael, Cynddylan, and Ffernfael or Ff yrnfael two of these names are
unknown, but Cynddylan is the same as the name of a 7th century king
killed in Shropshire ghting with Penda of Mercia against Oswald of
Northumbria. However, Gloucestershire did not remain long in the hands
of the Gewisse, for by 603 it was part of the kingdom of the Hwicce.
The last battle that Ceawlin fought was at Fethan leag, thought to
be Stoke Lyne in Oxfordshire. In Swantons translation of the entry for
584, the last sentence is given as and in anger he turned back to his own
[territory].46 This suggests strongly that he lost the battle, and may
account for the entry of 591, in which Ceawlin was driven from his kingdom. This battle is said to have taken place at Wodens Barrow, now
called Adams Grave, near Alton Priors, east of Devizes in Wiltshire. The
fact that it took place in Wiltshire suggests conict between the Wiltshire
Gewisse and the Oxfordshire Gewisse.
Ceawlins successor as ruler of the Gewisse was Ceol, the son of Cutha,
who is said to have become king in 591 or A.D. 592. The name is likely to
be British, but its origins are obscure. In Old Irish there are related words
giall captive, promise, giall to serve, obey, gell/gill to pledge, promise.47 Perhaps his name indicates that he was a sub-king who had pledged
allegiance to the Dorchester-on-Thames kings. He reigned for only ve
184
or six years, and was succeeded in A.D. 597 by his brother Ceolwulf. The
second element wulf sounds Germanic, but may be British there is an
Old Irish word olc, genitive uilc mad, wicked, evil,48 which is cognate
with wulf. His name, like that of Ceol, may indicate that he also was a
sub-king who had pledged allegiance to the Dorchester-on-Thames kings.
This casts some doubt on whether Cutha, Cuthwine and Cuthwulf were
indeed Saxons it is possible that the name Cutha is somehow connected
with that of the goddess Cuda (see Chapter 2), known from a relief found
at Cirencester, whose name is said to denote prosperity.49
Very little is known of the reign of Ceol and Ceolwulf, who were followed in A.D. 611 by Cynegils, whose relationship to Ceolwulf is uncertain.
The name Cynegils is also British, being the equivalent of Cuneglas or
Cuneglasus, also the name of one of the tyrants denounced by Gildas in
his De Excidio Britanniae. The rst element of Cuneglas is presumably the
word for dog, and the second element glas is a color word that means
something like blue-grey. Cuneglas (Blue/Grey-Hound) is obviously
similar to Cunorix (Hound-King), so it seems likely that Cynegils
belonged to the same Wiltshire branch of the Gewisse as Cynric. During
his reign, he is said to have fought alongside Cwichelm (presumably a
Saxon ally) at Beandun in A.D. 614, and killed 2046 of the Welsh (Beandun
is though to be Bindon in Devon). He and Cwichelm are also said to have
fought Penda of Mercia at Cirencester, and it is likely that Penda was victorious. Two points emerge from this list of battles: the Gewisse were
under pressure from Mercia, and they were already beginning to turn their
attention to lands in the south and west of England. The pressure from
Mercia may explain Cynegils and Cwichelms conversion to Christianity
in the 630s, and the establishment of a bishopric at Dorchester-on-Thames
(King Oswald of Bernicia was their godfather, and presumably became a
useful ally).
Apparently Cynegilss decision to convert to Christianity did not meet
with the approval of his successor Cenwalh. He is said to be the son of
Cynegils, and his name is also British, possibly Cunovalus (HoundStrong). Bede says that Cenwalh refused to be baptized, and abandoned
his wife, the sister of Penda, king of Mercia, whereupon Penda forced him
into exile. During his time in exile with the Christian king Anna of
East Anglia, however, Cenwalh did convert to Christianity and became
king of the Gewisse around A.D. 650. After this, the push toward the south
and west began in earnest. In A.D. 652 Cenwalh defeated the British at
Bradford-upon-Avon, and in A.D. 658 at Penselwood in east Somerset,
185
giving him control of the eastern part of Dumnonia. It is also likely

that around this time Cenwalh gained control of parts of Dorset, since
he is named as the rst Saxon patron of Sherborne Abbey in Dorset. In
around A.D. 660, Cenwalh established a bishopric at Winchester, probably
because the Gewisse no longer had effective control of Dorchester-onThames.50
Winchester became the second bishopric of the Gewisse for a number
of reasons. Dorchester-on-Thames was no longer viable because of pressure
from the Mercians; Winchester was some distance from the Mercians, but
close to the Jutish and British kingdoms that the Gewisse presumably coveted; it was relatively close to Salisbury Plain, the traditional center of
southern England; and it had a settled population of Saxons.
Winchester, of course, was originally the Roman civitas of Venta Belgarum, the capital of the Belgae, whose territory in Roman times may
have included Salisbury Plain. Russo says51 that by the end of the fourth
century, new public and private construction had virtually ceased in Venta,
the cemeteries had gone out of use by A.D. 420, and by A.D. 450 Ventas
subRoman town life had virtually ended. The rst sign of an AngloSaxon presence dates from the late fth century, with burials in the large
pagan cemetery at Kings Worthy, some 2 miles north of Venta. Inside the
town, the earliest actual evidence of Anglo-Saxon settlement are some
pottery sherds, from around A.D. 480.
The rst references to Winchester in Anglo-Saxon documents occur
in connection with the foundation of its episcopal see toward the middle
of the seventh century. The Anglo-Saxon Chronicle records that in A.D.
648 Cenwalh ordered the church of Winchester to be built and identies
it as St. Peters, evidently the Old Minster. So by this time Winchester
was under royal West Saxon control and termed a ceaster (castrum) fortied place.52 Archeological evidence
does indicate limited concentrations of seventh century intramural occupation in the central and northeast town quadrants.... Indeed, the presence
of an early Anglo-Saxon royal residence adjacent to or within the old
Roman forum-basilica complex is considered very likely.... In addition,
excavations have shown that late Roman Winchesters main south gate
was blocked during the late fth and sixth centuries, rst by a ditch and
then by a mortared wall. Trafc from the south thereafter had to enter
the town via more easterly gate (known signicantly enough as Kingsgate)
leading precisely to the forum and proposed early Anglo-Saxon palace
site.... The Lower Brook Street excavations of 1971 ... disclosed an early
Anglo-Saxon residential complex with a seventh century burial that
186

included an elaborate necklace of gold and garnet pendants and a collar of
silver rings.53
It is clear from this that Winchester, or at least the areas surrounding

Winchester, had been in Saxon hands for some time before Cenwalh established the bishopric there. The fact that the bishopric was established there
rather than in north Wiltshire may tell us something about the Wiltshire
Gewisse. Cynric rst entered the historical record when he captured Old
Sarum, and was next mentioned when he fought at Barbury Castle with
Ceawlin. This suggests that Cynric may have come from Venta Belgarum
or the surrounding area, or even from Amesbury or from one of the villages
on Salisbury Plain.
Why did the Wiltshire Gewisse not adopt Calleva Atrebatum (Silchester) as their capital? Yorke notes54 that some form of life did continue in
Silchester between 400 and 600, as shown by fragments of three Celtic
brooches which would be otherwise more at home in western Wessex.
It may be that Silchester was too close to Mercia and Essex to be a secure
capital, but I suspect there may be a historical reason. The Atrebates and
the powerful families of Salisbury Plain and surrounding areas must have
come into conict indeed, the Atrebates probably controlled territory
that had previously been in the hands of the Wiltshire elite. Cunliffe points
out55 that the abandoned hillfort of Bury Hill, in northwest Hampshire,
was reinforced around 100 B.C., presumably by the Atrebates, just at the
time that Danebury went out of use, and speculates that the polity that
constructed Bury Hill was in some way challenging the authority of the
nearby, long-established, Danebury, presumably under the control of the
ancient rulers of Salisbury Plain.
So what became of the displaced rulers of Wiltshire and Hampshire?
Those from the north of the region may have moved to Gloucestershire
in the territory of the Dobunni, but returned to north Wiltshire after the
Roman invasion both Gloucestershire and north Wiltshire seem to have
been dominated by a wealthy and sophisticated elite during Roman times.
Those from the south may have moved to hillforts like Cadbury Castle in
Somerset, and moved back later in Roman times, when the territory of
the Atrebates was divided among the Atrebates, the Regnenses and the
Belgae, with their capital at Venta Belgarum/Winchester. Given the history
between the old ruling families of Salisbury Plain and its surroundings
and the Atrebates, the abandonment of Calleva Atrebatum may have been
a conscious decision on the part of the Gewisse to get revenge on their
erstwhile enemies.
187
The Druid Arthur

What does all this tell us about southern England between A.D. 400
and A.D. 600? It seems that between A.D. 400 and A.D. 500 there was
conict between two groups: the Wiltshire group led by Ambrosius, which
was opposed to seeking help from Germanic mercenaries, and the Viroconium/Powys group led by Vortigern, which was happy to use Saxons in
their ght against the Irish and Picts. Ambrosius was most likely a Christian, and supported by the ancient families that had for so long dominated
Wiltshire and east Hampshire, but in spite of that, he seems to have become
celebrated in later legend as Arcturus/Arthur.
However, it is probable, as we saw in Chapter 1, that Arcturus/Arthur
only came to prominence after the comet of A.D. 539 and the ensuing
plague. Arcturus was a gure of the Wiltshire Druids, perhaps another
name for Rhesus, the Thracian Horseman, or for the Thracian Hermes,
the ancestor god of the Thracians, or for Lugus, who held up the heavens.
Arcturus, or Botes, or Lugus, was the giant-killer who used the shower
of meteors to slay Goliath or Balor or Ysbaddaden. The Druids, or those
who survived the centuries of Roman persecution, were keen astronomers,
and when they witnessed the comet, then experienced the ravages of the
plague, they must have seen these events as a sign, just as the aftermath of
the volcanic eruptions in 1628 B.C. and 1159 B.C. must have been interpreted
as omens. Their response to these events was to rally the British with the
tales of Arcturus/Arthur, a celestial gure well able to deal with celestial
events. But Arcturus/Arthur could not stop the Saxons, and at least one
descendant of the Wiltshire/east Hampshire families who had built and
controlled Stonehenge, parceled up the land for agriculture, organized the
middens and supervised the construction of the hillforts, decided that to
maintain any semblance of the power they had exercised for centuries or
even millennia, they must join forces with the Saxons and create a kingdom
that preserved the interests of the British living in the ancient heart of
southern England.
They seem to have done this rather successfully, since the capital of
Wessex was Winchester, not far from their former seats of power on Salisbury Plain and Danebury. While Arthur/Arcturus was ghting mythical
battles, the Wiltshire Gewisse were ghting real ones against Saxons who
threatened them, but also against British who did not share their view of
the future of southern England. Eventually, of course, after the Danes had
destroyed Northumbria and east Anglia and cut Mercia in half in the 890s,
188
Wessex under King Alfred became in effect the only English kingdom, and
the center of a unied England from the 10th century until the Norman
invasion in 1066.
There is an intriguing piece of evidence, however, that the kings of
Wessex never forgot their roots. In A.D. 979 the Dowager Queen lfthryth
founded a Benedictine monastery at Amesbury, the Abbey of St. Mary
and St. Melor. The abbey was named after St. Melor, because the saints
relics were kept there. Melor was Breton saint, whose story is very revealing.
His uncle Riwal killed Melors father St. Miliau, and wished to kill Melor,
who was only seven at the time. A council of bishops dissuaded him, so
he decided instead to maim the boy, cutting off his right hand (later
replaced by a silver prosthesis) and left foot (replaced with one of bronze).
The silver hand links Melor directly to Lludd (Nodens, the god of Lydney
Park), the king who experienced the three oppressions, and whose full
name was Lludd Llaw Ereint (Lludd of the Silver Hand) which suggests
that the Wessex royal family had long memories.
The length of their memories may explain another curiosity in the
Wessex story. In Henry of Huntingdons account of the Battle of Burford
in A.D. 752, the forces of Cuthred of Wessex were preceded into battle
against Aethelbald of Mercia by the aldorman Edelhun carrying the royal
emblem of a golden dragon, and in the scene of the Battle of Hastings on
the Bayeux Tapestry, Harold is seen ghting under a pennant-style standard
which is clearly a dragon.56 Was this the Wessex answer to the red dragon
of the Welsh, or does it preserve distant memories of the comet which perhaps gave birth to their kingdom?
Ambrosius, Arthur and Glastonbury

The descendants of Cerdic and Cynric ruled Wessex and then England
until 1066. In that year Edward the Confessor died childless, and William
of Normandy claimed the throne, on the basis that he was the great-nephew
of Emma of Normandy, Edwards mother. To enforce this claim, he
invaded England in 1066 and killed the English king Harold at the Battle
of Hastings. The Anglo-Saxon Chronicle records in the entry for 1066 that
this year appeared a comet on the fourteenth before the kalends of May
suggesting that the kingdom of Wessex came to an end very much as it
had come into being.57
From 1066 England became a Norman kingdom, and the Anglo-
189
Saxon nobles were displaced: some became prisoners in Normandy, or in

monastic communities; some ed to Scotland, including the last of the
Anglo-Saxon royal line, Edgar the Aetheling, who was named king after
the death of Harold but was never crowned, together with his sisters Margaret and Christina; and others, like Harolds mother Gytha and
Gospatrick Earl of Northumbria, ed to Flanders.58 William seized the
lands of the displaced Anglo-Saxon nobility and handed it over to Norman
aristocrats; by the time of the Domesday Book in 1086, says Daniell,59
the dispossession of the Anglo-Saxon nobles was practically complete.
Daniell says60 that the period from 1080 to 1120 was the lowest point
of the fortunes of the Anglo-Saxon aristocracy. By the reign of Henry I
their fortunes began to recover, especially when the Norman king Henry
I married Matilda, the daughter of Margaret and niece of the last AngloSaxon king Edgar the Aetheling. Matilda displayed an interest in her West
Saxon roots, received a genealogical table from the monks of Malmesbury
and was eventually buried in Westminster Abbey near Edward the Confessor.
Following the Conquest, Anglo-Saxon culture was under threat, but
was kept alive by the monasteries, which preserved the traditions of their
own communities against the changing religious and political world.61
Daniell points out62 that one of the dening works of the 12th century was
Geoffrey of Monmouths Historia Regum Britanniae. An important element
of the political power of the stories was that Arthur had not died but
waited to return to overthrow the oppressors. Arthur could therefore be
used by rebels trying to replace the king. The name Arthur was a powerful
symbol in its own right for example, Richard Is nephew and Henry IIs
grandson was called Arthur of Brittany.
Interest in Arthur was further stimulated by excavations at Glastonbury, which discovered Arthurs tomb and body, complete with an inscription on the cofn. Henry II showed great interest in the excavations, and
later used Arthurs reputed conquest of Ireland to justify his own conquest
of the island.
It might seem from this that the ancient families who had dominated
Salisbury Plain and its surroundings for so long had completely lost their
hold on power. However, the Wessex line was not completely eliminated.
Henry I had married Matilda, niece of the last Anglo-Saxon king, and
their daughter, also Matilda, was later to marry Geoffrey of Anjou and
become the mother of Henry II, the rst king of the House of Plantagenet.
Oddly, the name Plantagenet means broom (the plant), and recalls one
190
of the ingredients used by Math and Gwydion in the Fourth Branch of

the Mabinogion to make Lleus treacherous wife Blodeuedd.
The Wessex capital had been Winchester, and that initially remained
a center of power Daniell notes63 that the royal treasury was still at Winchester when Henry I became king. However, in 1141, during the civil war
that followed the death of Henry I, when Matilda fought the usurper king
Stephen of Blois, Winchester was partially destroyed by re, and was not
really rebuilt until the 14th century. In Wiltshire itself, Wilton, the administrative capital, founded in the 8th century, continued to prosper, and
the Benedictine abbey of Wilton was, at the time of the Domesday survey,
the richest of the womens monasteries.64 Another important town was
Malmesbury, in northwest Wiltshire, not far from the old Roman temple
at Nettleton Shrub. Malmesbury was the site of an Iron Age hillfort, and
a Benedictine monastery was established there in 656 by Adhelm, a nephew
of King Ine of Wessex. In 941, King Athelstan was buried in the Abbey.
By the 11th century it contained the second largest library in Europe and
was considered one of the leading European seats of learning.
But perhaps the fate of the Wiltshire families after the Conquest can
be summed up symbolically in the story of two religious establishments
in Wessex. As we said earlier, in 979 a Benedictine monastery had been
established at Amesbury, the Abbey of St. Mary and St. Melor. This continued for over a century after the Conquest, but in 1177 Henry II, as part
of his penance for instigating the murder of Thomas Becket, invited the
French Fontevrault order to set up a double priory at Amesbury.65 Beatrice,
the abbess of Amesbury, who had deed Henry and Theobald, archbishop
of Canterbury, in 1160 by refusing to accept their nominee in her church
at Froyle (Hampshire), was deposed, and her English nuns dispersed
apparently none chose to remain under the new regime.
The other religious establishment is Glastonbury Abbey, in Somerset.
Glastonbury has a very long history of human habitation, dating to the
Neolithic. Several timber and brushwood trackways have been discovered,
mostly running southward across the wetlands of the Somerset Levels.66
The oldest and best known of these is the Sweet Track, which was constructed around 3800 B.C. As Koch says, the vast amount of work required
to construct such a trackway demonstrates that the Glastonbury area must
have been home to a relatively large and well-organized community.
Another trackway, that at Meare Heath, was constructed in the second
millennium B.C., showing that Glastonbury still had a relatively large population in the Bronze Age. Occupation continued into the Iron Age, and
191
Glastonbury Lake Village was home to about 100 people between around
300 B.C. and A.D. 100. There is little sign of Roman occupation, though
it has been suggested that at least one of the wells in the Abbey St.
Josephs well in the crypt of the Lady Chapel may be of Roman date.67
The legends of Glastonbury suggest, says Gathercole,68 that it was
an important place in the shadowy years between the collapse of imperial
government in Britain and the establishment of Anglo-Saxon power in the
west. Glastonburys associations with postRoman British resistance to the
Anglo-Saxon expansion and in particular with Arthur may have been
exaggerated by the medieval monks, but are not necessarily wholly without
foundation. High status dark age occupation on the Tor, perhaps a chieftains stronghold (though perhaps a monastic site), has been conrmed by
archeological excavations.
Finds from these excavations include reused Roman pottery and building
materials and fragments of high quality imported Mediterranean wares,
the latter comparable with sherds found at other high status sites both
secular and religious in the South West. Many of these are the remains
of amphorae believed to have been used to trade wine and oil.69
William of Malmesbury, the 12th-century historian and monk at
Malmesbury Abbey, who had a Norman father and an English mother,
provides some interesting details on Glastonbury.70 He says that the
rst church at Glastonbury, which the Anglo-Saxons called the Old
Church, was a wattle-and-daub structure, built by the disciples of Christ.
Gathercole says71 that there is evidence of wattle-and-daub structures,
but whether they are British, as William implies, or Saxon is unclear.
William also says that some of the earliest abbots were British, including
a certain Worgrez, who was granted land by a king of Dumnonia, when
Glastonbury was known as Ineswitrin (Ynys Witrin, or Glass Island in
Welsh).
It is possible, says Gathercole,72 that
the traditions of a British origin for the Abbey may be based in truth,
though this remains more problematical. The rst reliable charters for the
estates of the Abbey, the driving force of Glastonburys medieval history,
are late 7th century, and there is no proof archeological or documentary that the Abbey existed before this. However, an earlier charter of
601, though not considered authentic in its present form, may record an
actual grant of land by an unnamed British king of Dumnonia to the Old
Church. The Saxon charters also show that some (though not the earliest
recorded) of the early Abbots had British names, which may support the
idea of an existing British tradition, later taken over by the Saxons.
192
There are signs that Glastonbury Abbey did not prosper in the time
of Alfred the Great (871899), possibly because of Viking incursions into
the south of England.73 However, after Dunstan became abbot in 946 and
refounded it as a Benedictine abbey, Glastonbury ourished. Kings
Edmund, Edgar and Edmund Ironside were buried there; and for a short
time the English treasury was held at Glastonbury. The Abbey became
one of the richest, and at times the richest, of all the great Benedictine
houses in England. The Domesday Book reveals Glastonbury Abbey to
have been the wealthiest in England in the second half of the 11th century,
even though the Conquest had disrupted its economic, and its spiritual,
life. In 1184, however, a serious re destroyed the Old Church and many
books, vestments and relics which had been accumulated. It is in this context that in 1191 the monks of Glastonbury apparently discovered the body
and tomb of Arthur.
There is, it seems to me, a symbolic connection between the removal
of the English nuns from Amesbury and the discovery of Arthurs grave
at Glastonbury. The nuns at Amesbury had been custodians of the relics
of St. Melor, whose silver prosthetic hand recalls the ancient god of Lydney
Park, Nodens, and his successor Lludd Llaw Ereint. In honoring the relics
of St. Melor, the nuns were paying tribute to a preChristian ancestor,
just as the Christians of Uley honored their preChristian ancestor by taking good care of the head of Mercury/Lugus in the century after the end
of Roman rule. When the monks discovered the grave of Arthur at Glastonbury, they were seeking to pay their respects to an ancestor in the same
way. Arthur may have begun life as the Druids Arcturus, a powerful stargod, and the real Arthur may have been, as Reno proposes, Ambrosius
Aurelianus, but by the 12th century the monks could not separate fact
from legend, and honored this important gure by putting his relics on
display. The Druids had long since disappeared, but their monster-slaying
hero managed to live on at Glastonbury and that may be the Druids
most enduring legacy.
Chapter Notes
Chapter 1
ical Encyclopedia (Santa Barbara, CA: ABCCLIO, 2006), 612.

14. John F. Healy, Natural History: A Selection (London: Penguin), 216.
15. John Bostock and H.T. Riley, trans.,
The Natural History of Pliny, vol. 5 (London:
Henry G. Bohn, 1857), 412.
16. Ammianus Marcellinus, Roman History, trans. J.C. Rolfe (Cambridge, MA: Harvard University Press, 1950), Book 15, http://
penelope.uchicago.edu.
1. Geoffrey of Monmouth, History of the

Kings of Britain, trans. J.A. Giles (London:
Henry G. Bohn, 1848), Book 8, Chapters 14
17, http://www.lib.rochester.edu.
2. Edward Gibbon, The History of the Decline and Fall of the Roman Empire, ed. David
Womersley (London: Penguin), 577. Chinese
records mention the appearance of Halleys
Comet in 530 A.D., and record another notable comet in 539 A.D.
3. Nennius, History of the Britons, trans.
J.A. Giles (London: Henry G. Bohn, 1848),
Chapter 56, http://www.fordham.edu.
4. Ibid., Chapter 73.
5. Sarah Higley, trans., The Spoils of
Annwn, http://www.lib.rochester.edu.
6. What man is the gatekeeper/porter?
Mark Adderley, trans., http://www.markadde
rley.net.
7. Culhwch and Olwen, in The Mabinogion, trans. Lady Charlotte Guest (London:
J.M. Dent, 1910), http://www.maryjones.us.
8. Julius Caesar, Gallic Wars, trans. W.A.
McDevitte and W.S. Bohn (New York: Harper
and Brothers, 1869), Book 6, Chapter 13,
http://classics.mit.edu.
9. Diodorus Siculus, Historical Library,
trans. C.H. Oldfather (Cambridge, MA: Harvard University Press, 1935), Book 5, Chapter
31, http://penelope.uchicago.edu.
10. Strabo, Geography, trans. H.L. Jones
(Cambridge, MA: Harvard University Press,
1932), Book 4, Chapter 4.4, http://penelope.
uchicago.edu.
11. Hippolytus of Rome, Philosophumena,
trans. F. Legge (London: Society for Promoting Christian Knowledge, Macmillan, 1921)
Book 1, Chapter 22, http://www.archive.org.
12. Clement of Alexandria, Stromata, trans.
William Wilson (Buffalo, NY: Christian Literature Publishing, 1885), Book 1, Chapter 5,
http://earlychristianwritings.com.
13. John T. Koch, Celtic Culture: A Histor-
Chapter 2
1. Tacitus, Annals, trans. Alfred John
Church and William Jackson Broadribb (New
York: Random House, 1942), Book 14, Chapter 30, http://www.perseus.tufts.edu.
2. Ammianus Marcellinus, Epitome of
Book 32.
3. Curse Tablets from Roman Britain,
at the website of the Centre for the Study of
Ancient Documents, http://curses.csad.ox.ac.
uk, provides excellent information on a number of Roman temples, including Uley, Lydney
and Pagans Hill.
4. Alexei Kondratiev, Lugus: the ManyGifted Lord, An Trbhs Mhr: The IMBAS
Journal of Celtic Reconstructionism 1 (1997),
http://www.imbas.org.
5. See www.roman-britain.org/nettleton.
6. Ibid.
7. Wiltshire and Swindon Sites and Monument Record Information: Nettleton Shrub,
http://history.wiltshire.gov.uk/smr/getsmr.ph
p?id=2096.
8. David Blamires, Herzog Ernst and the
Otherworld Voyage (Manchester: Manchester
University Press, 1979), 389.
9. Miranda Green, Animals in Celtic Life
and Myth (London: Routledge, 1998), 126.
10. The results of this investigation were
televised in Time Team, Series 8, Episode 8,
193
194
Notes Chapter 3
The Bone Cave, Alveston, Gloucestershire, summarized at http://www.channel4.com/prog

rammes/time-team/episode-guide/series-8/
episode-8.
11. Jodie Lewis, Upwards at 45 Degrees:
The Use of Vertical Caves during the Neolithic and Early Bronze Age on Mendip,
Somerset, Capra 2 (2000), http://capra.gro
up.shef.ac.uk.
12. Wiltshire Council, Sites and Monuments Record: Wilsford Shaft, Normanton
Down, http://history.wiltshire.gov.uk.
13. Wessex Archaeology, Remarkably preserved Roman remains from grave, http://
www.wessexarch.co.uk/projects/wiltshire/bosc
ombe/preserved-roman-remains/index.html.
14. Graham Thomas, The Romans at Woodchester, http://grahamthomas.com.
15. Giles Clarke, The Roman Villa at
Woodchester, Britannia 13 (1982): 221222.
16. Ibid., 213.
17. Katherine Dunbabin, Mosaics of the
Greek and Roman World (Cambridge, UK :
Cambridge University Press, 1999), 923.
18. Martin Henig, Religion in Roman Britain
(London: Routledge, 1984), 220.
19. Ibid., 201.
20. Jocelyn Toynbee, Death and Burial in
the Roman World (Baltimore, MD: Johns
Hopkins University Press, 1996), 276.
21. Miranda Green, Symbol and Image in
Celtic Religious Art (London: Routledge, 1992),
58.
22. Ibid., 57.
23. Ibid., 37.
24. Ibid.
25. Ibid., 185.
26. Ibid.
27. Ibid., 186.
28. Ibid.
29. Ibid., 187.
30. Ibid.
31. Stephen Yeates, The Tribe of Witches:
The Religion of the Dobunni and the Hwicce
(Oxford, UK: Oxbow Books, 2009).
32. Hilda Roderick Ellis Davidson, Myths
and Symbols in Pagan Europe (Manchester,
UK: Manchester University Press, 1988), 47.
33. Miranda Green, Celtic Myths (Austin:
University of Texas Press, 1998), 62.
34. Miranda Green, An Archaeolog y of Images (London: Routledge, 2004), 118.
35. Henig, 49.
36. Ibid., 48.
37. Rachel Bromwich, Trioedd Ynys Prydein. The Welsh Triads (Cardiff, UK: University of Wales Press, 1978).
38. Henri Hubert, Le mythe dEpona, in
Mlanges linguistiques offerts M.J. Vendryes,
ed. Jules Bloch (Paris: Librairie Ancienne

Edouard Champion, 1925), 187198.
39. Miranda Green, The Gods of Roman
Britain (Princes Risborough, Buckinghamshire,
UK: Shire Publications, 1983), 56.
40. Ibid.
Chapter 3
1. Koch, Celtic Culture, 1456.
2. Nennius, Chapter 62.
3. Higley, The Spoils of Annwn.
4. Julius Pokorny, Indogermanisches Etymologisches Wrterbuch (Bern, Switzerland:
Francke, 1959), 534, http://www.indoeurope
an.nl.
6. Alexander Falileyev, Dictionary of Continental Celtic Place-Names (Aberystwyth,
UK : CMCS Publications, 2010), entry for
sedo- seat, location, http://cadair.aber.ac.uk.
7. Roger D. Woodard, Indo-European Sacred Space (Champaign: University of Illinois
Press, 2006), 78. Here as elsewhere in the
book, I have taken my insights into Sanskrit
from the Cologne Digital Sanskrit Lexicon,
which is based on Monier-Williamss SanskritEnglish dictionary, and can be accessed at
http://webapps.uni-koeln.de.
8. Mary Boyce and Frantz Grenet, Zoroastrianism under Macedonian and Roman Rule,
Part 1 (Leiden, Holland: Brill, 1991), 477.
9. James Darmesteter, trans., Vendidad,
in Sacred Books of the East (New York: Oxford
University Press, 1880), 1.17, http://www.av
esta.org.
10. Darmesteter, 10.14.
11. Raffaele Pettazzoni, The All-Knowing
God (New York: Arno Press, 1978), 119.
12. Rigveda, trans. Ralph Grifth (Benares,
India: E.J. Lazarus, 1897), 1.152.2, http://
www.sacred-texts.com.
13. Pokorny, Indogermanisches Etymologisches Wrterbuch, 11601162.
14. Frank E. Romer, trans., Pomponius
Melas Description of the World (Ann Arbor:
University of Michigan Press, 1998); Daithi
O Hogain, The Celts: A History, Part 70 (Woodbridge, Suffolk, UK : Boydell Press, 2003),
25.
15. Thomas Taylor, trans., Life of St. Samson of Dol (London: Society for Promoting
Christian Knowledge, Macmillan, 1925),
Book 1, Chapter 27, http://www.lamp.ac.uk.
16. Geoffrey of Monmouth, The Life of
Merlin, trans. John Jay Parry (Urbana: University of Illinois, 1925), http://www.sacredtexts.com.
Notes Chapter 4
17. Peredur, Son of Efrawg. In The
Mabinogion, trans. Lady Charlotte Guest.
18. Roger Sherman Loomis, The Grail:
From Celtic Myth to Christian Symbol (Princeton: Princeton University Press, 1991), 33.
19. Joseph Clancy, trans., Y Gododdin,
in Earliest Welsh Poetry (London & New York:
Macmillan, 1970), http://www.maryjones.us.
The lines Ive quoted come from stanzas VIII,
XI, XXXII, and LXI.
20. Song Before the Sons of Llyr, in The
Four Ancient Books of Wales, trans. W.S. Skene
(Edinburgh: Edmonston and Douglas), http://
www.maryjones.us.
21. Will Parker, Mabinogi: Branwen,
Daughter of Llyr, http://www.mabinogi.net.
22. Ernest Brehaut, An Encyclopedist of the
Dark Ages: Isidore of Seville (New York: Columbia University, 1912), http://bestiary.ca.
23. Pokorny, Indogermanisches etymologisches Wrterbuch, 855.
24. Giraldus Cambrensis, De principis instructione, in The Journey through Wales; and,
The Description of Wales, trans. Lewis Thorpe
(London: Penguin, 1978), http://www.brita
nnia.com.
25. Giraldus Cambrensis, Speculum Ecclesiae, in The Journey through Wales; and,
The Description of Wales, trans. Lewis Thorpe
(London: Penguin, 1978), http://www.britan
nia.com.
27. The Sick Bed (Wasting Sickness) of
Cuchulain, in Cuchulain of Muirthemne,
trans. Lady Augusta Gregory (London: John
Murray, 1902), http://www.sacred-texts.com.
28. Hostile Confederacy, in Skene, Four
Ancient Books.
29. The Chair of the Sovereign, in
Skene, Four Ancient Books.
30. The Dialogue of Gwyddno Garanhir
and Gwynn ap Nudd, in Skene, Four Ancient
Books.
31. Chinua Achebe, Things Fall Apart
(London: Heinemann, 1996), 108.
32. Lewis Spence, The Mysteries of Britain
(Pomeroy, WA : Health Research Books,
1996), 123.
34. John R. Hinnells, Mithraic Studies
(Manchester, UK : Manchester University
Press, 1975), 9.
35. Carl Huffman, Philolaus, in Stanford
Encyclopedia of Philosophy, ed. Edward N.
Zalta, http://plato.stanford.edu.
36. Huffman, Pythagoreanism, in Stanford Encyclopedia of Philosophy.
37. Richard G. Geldard, The Travelers Key
195
to Ancient Greece: A Guide to Sacred Places

(Wheaton, IL: Quest Books, 2000), 60.
39. W.J. Gruffydd, Mabon ab Modron,
Revue celtique 33 (1912): 459460.
40. Dan Merkur, Fruit of the Terrestrial
Paradise: The Psychedelic Sacrament in St.
Ephrem the Syrian and Celtic Christianity, http://
www.danmerkur.com.
42. Giraldus Cambrensis, The Topography
of Ireland, trans. Thomas Forester (London:
Henry G. Bohn, 1863), 967.
43. Dorothy Watts, Christians and Pagans
in Roman Britain (London: Taylor & Francis,
1991), 1423.
Chapter 4
1. Patrick K. Ford, The Mabinogi and
Other Medieval Welsh Tales (Berkeley: University of California Press, 2008), 3.
3. Parker, http://www.mabinogi.net.
5. Parker, Mabinogi: Pwyll, Prince of
Dyfed.
6. Ibid.
7. Ibid.
8. Ibid.
9. Ifor Williams, Pedeir Keinc y Mabinogi
(Cardiff, UK: Gwasg Prifysgol Cymru, 1974).
10. Joseph Baudis, Mabinogion, FolkLore 37 (1916): 51.
11. Parker, Mabinogi: Branwen, Daughter
of Llyr.
12. Ranko Matasovic, Etymological Lexicon
of Proto-Celtic, entry for liro-, http://www.
indo-european.nl.
13. J.G. Oosten, The War of the Gods: The
Social Code in Indo-European Mytholog y (London: Routledge, 1985), 869.
15. Pausanias, Description of Greece, trans.
W.H.S. Jones (Cambridge, MA: Harvard University Press, 1918), 10.23, http://www.theoi.
com.
16. Diodorus Siculus, 22.9.
17. Strabo, Book 4, Chapter 4.1.
18. John T. Koch, A Welsh Window on
the Iron Age: Manawydan, Mandubracios,
CMCS 14 (1987): 1752.
22. Parker, Mabinogi: Manawydan Son of
Llyr.
196
Notes Chapter 5
23. Ibid.
24. The Metrical Dindshenchas, ed. Edward
Gwynn (Dublin: Dublin Institute for
Advanced Studies, 1991), http://www.ucc.
ie.
27. John Carey, A British myth of origins? History of Religions 31 (1991).
28. Parker, Mabinogi: Math, Son of
Mathonwy.
29. Ibid.
30. Ibid.
31. Ibid.
32. Ibid.
33. Ibid.
34. Henri dArbois de Jubainville, Les
Druides et les dieux celtiques forme danimaux
(Boston, MA: Adamant Media Corporation,
2005), 160162.
35. Dimitri Nikolai Boekhoorn, Mythical, Legendary and Supernatural Bestiary in
Celtic Tradition: From Oral to Written Literature (PhD diss., Universit Rennes 2/University College Cork, 2008), 82.
37. Ibid., 83.
38. John Rhys, Celtic Folklore: Welsh and
Manx (Boston, MA: Adamant media Corporation, 2004), 645.
40. Elizabeth Archibald, Incest and the Medieval Imagination (Oxford, UK: Oxford University Press, 2001), 5455.
41. Christopher R. Fee and David Leeming, Gods, Heroes and Kings: The Battle for
Mythic Britain (Oxford, UK: Oxford University Press, 2004), 47.
42. Sarah Higley, Between Languages: The
Uncooperative Text in Early Welsh and Old English Nature Poetry (University Park : Penn
State Press, 1993), 2912.
43. The Battle of the Trees, Skene, Four
Ancient Books.
44. Mary Jones, Notes to The Battle of
the Trees, http://www.maryjones.us.
Chapter 5
Kings of Britain, Book 6, Chapter 15, and
Book 8, Chapter 10.
2. Julian Thomas, Understanding the Neolithic (London: Routledge, 1999), 204.
3. Ibid.
4. Ibid., 206.
5. Barry Cunliffe, The Oxford Illustrated

History of Prehistoric Europe (Oxford: Oxford
6. David Field, The Investigation and Analytical Survey of Silbury Hill (Swindon, Wiltshire, UK: English Heritage, 2002), 12.
7. David Field, Great sites: Silbury Hill,
British Archaeolog y 70 (2003), http://www.bri
tarch.ac.uk.
8. Field, Great sites: Silbury Hill.
9. Field, Investigation, 74.
10. Field, Great sites: Silbury Hill.
11. Mike Parker-Pearson, Stonehenge Riverside Project, http://www.shef.ac.uk/archaeol
ogy/research/stonehenge.
12. Brian John, The Bluestone Enigma
(Newport, Pembrokeshire, UK : Greencroft
Books, 2008); O. Williams-Thorpe, P.J.
Potts, M.C. Jones, and P.C. Webb, Preseli
Spotted Dolerite Bluestones: Axe-heads,
Stonehenge Monoliths and Outcrop Sources,
Oxford Journal of Archaeolog y 25 (2006):
2964.
13. The Amesbury Archer, http://www.
wessexarch.co.uk.
14. The Boscombe Bowmen, http://www.
wessexarch.co.uk.
15. Cunliffe, Oxford Illustrated History,
251.
16. Ibid., 253.
17. Ibid., 254.
18. See Durrington Walls: a Time Team
Special, summarized at http://www.channel4.
com/history/microsites/T/timeteam/2005_du
rr_t.html.
19. Parker-Pearson.
20. Ibid.
21. Stuart Needham, Andrew Lawson, and
Ann Woodward, Rethinking Bush Barrow,
British Archaeolog y 104 (2009), http://www.
britarch.ac.uk.
22. Ibid.
23. James Dyer, Discovering Prehistoric England (Princes Risborough, Buckinghamshire:
Shire Publications, 2001), 192.
24. Mike Pitts, Hengeworld (London:
Arrow Books, 2000), 2958.
25. A.F. Harding, European Societies in the
Bronze Age (Cambridge, UK: Cambridge University Press, 2000), 190.
26. Lidia D. Sciama and Joanne Bubolz
Eicher, Beads and Bead Makers (Oxford, UK:
Berg Publishers, 1998), 254.
27. Mike Baillie, Irish Tree Rings and an
Event in 1628 B.C., in Thera and the Aegean
World III, ed. David A. Hardy (London:
Thera Foundation, 1990), 160166.
28. Barry Cunliffe, Landscape with People, in Culture, Landscape and the Environ-
Notes Chapter 5
ment, ed. Kate Flint and Howard Morphy
(Oxford, UK: Oxford University Press, 2000),
120. In Iron Age Communities, Cunliffe
calls these elds coaxial eld systems. They
are popularly and erroneously known as
Celtic elds.
29. Cunliffe, Landscape with People,
121.
30. Ibid.
31. Ibid.
32. Ibid.
33. Ibid.
34. Ibid.
35. Ibid.
36. Cunliffe, Landscape with People, 122.
In Iron Age Communities, Cunliffe implies that
some ranch boundaries were constructed late
in the 2nd millennium, but that others were
constructed some time after 900 B.C.
37. Ibid.
38. Ibid.
39. Ibid., 124.
40. Mike Baillie, A Slice Through Time:
Dendrochronolog y and Precision Dating (London: Routledge, 1995), 778.
1245.
42. David McOmish, Landscapes Preserved by the Men of War, British Archaeolog y
34 (1998), http://www.britarch.ac.uk.
43. Rodney Castleden, The Making of
Stonehenge (London: Routledge, 1993), 227.
44. English Heritage, National Monument
Report, SU14SW67, SU14SW68.
45. Kurt Hunter-Mann, Excavations at
Vespasians Camp Iron Age Hillfort, 1987,
The Wiltshire Archaeological and Natural History Magazine 92 (1999): 3952.
46. Ian Plimer, Heaven and Earth (Ballan,
Victoria, Australia: Connor Court, 2009), 57;
Charles A. Perry and Kenneth Hsu, Geophysical, Archaeological, and Historical Evidence Support a Solar-Output Model for
Climate Change, PNAS 97.23 (2000): 12435.
47. Trevor Bryce, The Kingdom of the Hittites (Oxford, UK: Oxford University Press,
2005), 356.
48. Herodotus, The Histories, trans. A.D.
Godley (Cambridge, MA: Harvard University
Press, 1920), 7.170171, http://perseus.tufts.
edu.
49. Michael Wood, In Search of the Trojan
War (Berkeley: University of California Press,
1998), 2178.
50. Herodotus, 1.94.
51. Robert Beekes, The Origins of the
Etruscans, http://www.knaw.nl.
52. Cunliffe, Iron Age Communities in
Britain (London: Routledge, 2005), 2478.
197
53. Ibid., 92.

54. Ibid., 90.
55. Ibid., 92.
56. John Barrett and David McOmish, All
Cannings Cross, http://www.allcannings.org.
57. David McOmish, East Chisenbury:
Ritual and Rubbish at the British Bronze-AgeIron Age Transition, Antiquity 70.267 (1996):
6876.
1256.
59. Ibid., 126.
60. Cunliffe, Iron Age Communities, 65.
61. Ibid.
62. Ibid., 66.
126.
64. Ibid.
65. Ibid.
66. Ibid.
67. Ibid.
68. Ibid.
69. Timothy Darvill, Prehistoric Britain
(London: Routledge, 1997), 135; Kristian
Kristiansen, Europe Before History (Cambridge, UK : Cambridge University Press,
2000), 300.
70. Kristiansen, ibid.
71. Darvill, 135.
72. Ibid.
73. Ibid.
74. Ibid., 136.
75. Ibid.
76. Ibid.
77. Ibid., 1367.
126.
79. Ibid., 128.
80. Anne Ross, Ritual and the Druids, in
The Celtic World, ed. Miranda Green (London: Routledge, 1996), 439.
81. Ibid.
83. Ibid., 98.
84. Ibid., 989.
85. Darvill, 137.
86. Ibid.
87. Ibid.
89. Ibid.
90. Ibid.
91. Ibid., 178.
92. Ibid., 189.
93. Ibid.
94. Ibid., 127.
95. Dennis Price, The Missing Years of Jesus:
The Greatest Story Never Told (London: Hay
House, 2009), 2278.
96. Diodorus Siculus, 2.47.
198
Notes Chapter 6
Chapter 6
Kings of Britain, Book 1, Chapters 1516.
3. The Parish and Priory Church of St.
Mary, Totnes, Devon (Diocese of Exeter),
http://www.churchcare.co.uk.
4. Prehistoric Dartmoor, http://www.dart
moor-npa.gov.uk.
5. Ibid.
6. Herodotus, 1.94.
7. Beekes, Origins of the Etruscans.
8. H.W. Bailey, Arya II, BSOAS 23.1
(1960).
9. Ibid., 34.
11. Julius Pokorny, Zur keltischen Namenkunde und Etymologie, Vox Romanica 10
(1949): 232.
13. Old Iranian Online, www.utexas.edu.
14. Giraldus Cambrensis, Itinerary Through
Wales, trans. Sir Richard Colt Hoare (London:
W. Miller, 1806), 390392, http://www.arch
ive.org.
16. Lludd and Llefelys, in The Mabinogion, trans. Lady Charlotte Guest.
17. Bromwich, Welsh Triads. The Triads
themselves can be accessed at http://norin77.
50megs.com/triads.htm.
18. Nennius, chapters 4042.
19. Athenaeus, Deipnosophists, trans. C.D.
Yonge (London: H.G. Bohn, 1854), Book 6,
Chapter 49, http://www.attalus.org.
20. Paul Dunbavin, Picts and Ancient
Britons (Long Eaton: Nottingham: Third Millennium Publishing, 2001), 1.
21. Dunbavin, 1.
22. Ibid., 93.
23. Ibid.
24. Ibid., 94.
25. Ibid., 99.
27. Asdis Magnusdottir, La Voix du cor
(Amsterdam: Rodopi, 1998), 1212.
28. Jan Bouzek, Cimmerians and Early
Scythians: The Transition from Geometric to
Orientalising Style in the Pontic Area, in
North Pontic Archaeolog y: Recent Discoveries
and Studies, ed. Gocha S. Tsetskhladze (Leiden: Brill, 2001), 38.
29. Herodotus, Book 4.
30. Carola Metzner-Nebelsick, Early Iron
Age Pastoral Nomadism in the Great Hungarian Plain Migration or Assimilation? The
ThracoCimmerian Problem Revisited,

in Kurgans, Ritual Sites and Settlements, ed.
Jeannine Davis-Kimball, http://www.csen.
org.
31. Kristiansen, 194.
32. Ibid., 193.
33. Bouzek, 368.
34. Kristiansen, 205.
35. Ibid., 206.
36. Ibid., 2067.
37. Hermann Sauter, Denition der
thrako-kimmerischen Bronzen, http://www.
kimmerier.de.
38. Nancy K. Sandars, Bronze Age Cultures
in France (Cambridge, UK: Cambridge University Press, 1957), 215, 225.
39. Barry Cunliffe, Understanding Hillforts: Have We Progressed?, in The Wessex
Hillforts Project, ed. Andrew Payne, Mark
Corney and Barry Cunliffe (Swindon, Wiltshire, UK: English Heritage, 2006).
40. Cunliffe, Understanding Hillforts,
1556.
41. Ibid., 156.
42. Ibid.
43. Ibid.
44. Ibid.
46. Robert Beekes, Greek Etymological Dictionary, http://www.indo-european.nl.
48. Cunliffe, Understanding Hillforts, 160.
49. Ibid., 161.
51. Martin Litcheld West, Indo-European
Poetry and Myth (Oxford, UK: Oxford University Press, 2007), 1456.
52. Ian Johnston, The Iliad (Arlington,
VA : Richer Resources Publications, 2006),
2189.
53. Robert Turcan, The Cults of the Roman
Empire (Oxford, UK: Blackwell, 1996), 248.
54. Turcan, 249.
55. Ibid.
56. Pettazzoni, 179180.
57. Ibid., 181.
58. Herodotus, Book 5, Chapter 7.
60. Ibid., 1845.
61. Ibid.
62. Ibid.
63. Ibid., 186.
64. David Rankin, Celts and the Classical
World (London: Routledge, 1996), 30.
65. John OMeara, The History and Topography of Ireland (London: Penguin, 1982),
110.
Notes Chapter 7
66. Daniel Bray, Sacral Elements of Irish
Kingship, http://escholarship.usyd.edu.au.
67. West, 28.
68. Susan Guettel Cole, Greek Religion,
in A Handbook of Ancient Religions, ed. John
R. Hinnells (Cambridge, UK: Cambridge University Press, 2007), 2989.
69. Philipp Vandenberg, Mysteries of the
Oracles (London: Tauris Parke Paperbacks,
2007), 27.
70. Vandenberg, 29.
71. Pausanias, 8.38.2.
73. Raimo Anttila, Greek and Indo-European Etymolog y in Action (Amsterdam: John
Benjamins, 2000), 72.
74. West, 28.
75. N.D. Kazanas, Indo-European Deities
and the Rigveda, Journal of Indo-European
Studies 29 (2001).
76. J.P. Mallory and Douglas Q. Adams,
Encyclopedia of Indo-European culture (London: Taylor & Francis, 1997), 265.
77. Green, Animals in Celtic Life, 112.
78. Ibid., 52.
79. Ibid., 211212.
80. Green, Symbol and Image, 18.
81. Jan Bremmer, The Rise and Fall of the
Afterlife (London: Routledge, 2002), 38.
82. Fritz Graf, Apollo (London: Taylor &
Francis, 2008), 3940.
83. Morris Silver, Taking Ancient Myths
Economically (Leiden, Holland: Brill, 1992),
237.
84. Pausanias, Book 10.5.7.
85. Apollonius Rhodius, Argonautica,
trans. R.C. Seaton (Cambridge, MA: Harvard
University Press, 1912), 4.603618, http://
www.omacl.org.
86. Pausanias, Book 10.5.710.
87. Herodotus, Book 4.33.
88. Silver, 238.
89. Hesiod, Theogony, trans. Hugh G. Evelyn-White (Cambridge, MA: Harvard University Press, 1914), lines 767774, http://
90. Sir James George Frazer, The Golden
Bough: Studies in Magic and Religion (London:
Penguin, 1996), 1989.
91. Manfred Lurker, The Routledge Dictionary of Gods and Goddesses, Devils and
Demons (London: Routledge, 2004), 1314.
92. Norman Oliver Brown, Hermes the
Thief (Great Barrington, MA: Steiner Books,
1990), 367.
93. Bremmer, 301.
94. C. Scott Littleton and Linda Malcor,
From Scythia to Camelot (London, Taylor &
Francis, 2000).
199
95. Thomas Green, The Historicity and

Historicisation of Arthur, http://www.arth
uriana.co.uk.
96. Bromwich, Welsh Triads, 5445.
97. Graham Anderson, King Arthur in
Antiquity (London: Routledge, 2004), 28.
98. Richard Hinckley Allen, Star Names:
Their Lore and Meaning (Mineola, NY: Dover
Publications, 1963), 924.
99. Allen, 95.
100. Anderson, 29.
101. Allen, 99.
102. Hugh G. Evelyn-White, Hesiod the
Homeric Hymns and Homerica (Charleston,
SC: BiblioBazaar, 2008), 75.
103. Anderson, 32.
104. Ibid., 323.
105. Ibid., 33.
106. Ibid., 334.
107. Ibid., 34.
108. C. Scott Littleton and Linda Malcor,
The Germanic Sword in the Tree, The Heroic
Age 11 (2008).
110. Ammianus Marcellinus, 31.2.
111. Anderson, 96.
113. Anderson, 16.
114. Ibid., 104.
116. Anderson, 99.
118. F. Graham Millar, The Celestial
David and Goliath, Journal of the Royal Astronomical Society of Canada 89.4 (1995): 142.
119. Proinsias Mac Cana, Celtic Mytholog y
(New York: Hamlyn, 1970).
120. Millar, 144.
121. Allen, 285.
122. Rankin, 59.
123. Millar, 1456.
124. Allen, 20.
125. Ibid., 147.
126. Mircea Eliade, Shamanism: Archaic
Techniques of Ecstasy (London: Routledge &
Kegan, 1964).
127. Millar, 1489.
128. Ibid., 149.
129. Allen, 319.
130. Ibid., 318.
131. Millar, 1501.
132. Ibid., 151.
133. Matasovic, entry for abon-.
Chapter 7
1. West, 287.
2. Ibid., 378.
200
Notes Chapter 8
3. West, 3789.
4. Kazanas, 257293.
5. James Hastings, Encyclopedia of Religion
and Ethics (Whitesh, MT: Kessinger, 2003),
805.
6. John Waterhouse, Zoroastrianism (San
Diego, CA: Book Tree, 2006), 63.
7. Waterhouse, 65.
8. W.H. Matthews, Mazes and Labyrinths
(Charleston, SC: Forgotten Books, 2008), 71.
9. W.H. Matthews, 74.
10. Ibid., 79.
11. Ibid., 812.
12. Ibid., 87.
13. Ibid., 88.
14. Ibid., 90.
15. Ibid., 92.
16. Ibid., 156.
17. Ibid.
18. Ibid., 52.
19. Ibid., 158.
20. Ibid., 160.
21. Johnston, 416.
22. Craig M. Wright, The Maze and the
Warrior (Cambridge, MA: Harvard University
Press, 2001), 130.
23. Plutarch, Life of Theseus, in Parallel
Lives, trans. Aubrey Stewart and George Long
(London: Henry G. Bohn, 1881), http://www.
gutenberg.org.
24. W.H. Matthews, 162.
26. Henning Eichberg, The Labyrinth of
the City Fractal Movement and Identity,
in Nature and Identity: Essays on the Culture
of Nature, ed. Kirsti Pedersen and Arvid
Viken. Bloomington: Indiana University
Press, 2003.
28. Price, 1046.
30. Green, Historicisation of Arthur, 10.
31. Thomas McEvilley, The Shape of Ancient Thought (New York : Allworth Press,
2002), 99.
32. McEvilley, 100.
33. Ibid.
34. Ibid., 101.
35. Ibid., 25.
36. Ibid., 256.
37. Ibid., 26.
38. Ibid., 102.
39. Ibid., 104.
40. Ibid., 105.
41. Higley, Between Languages, 2912.
42. The Battle of the Trees, in Skene,
Four Ancient Books.
43. Bhagavadgita, translated by Sanderson

Beck, http://www.san.beck.org.
44. Jane Webster, Sanctuaries and Sacred
Places, in The Celtic World, edited by Miranda Green (London: Routledge, 1996), 448.
47. Homer, The Iliad, trans. Samuel Butler, http://classics.mit.edu.
48. Strabo, Book 7.3.
49. Iowerth Eiddon Stephen Edwards, The
Cambridge Ancient History (Cambridge, UK:
Cambridge University Press, 1970), 417.
51. Edwards, 418.
52. Andrew Bell-Fialkoff, The Role of Migration in the History of the Eurasian Steppe
(New York: Palgrave Macmillan, 2000), 195.
Chapter 8
1. Thomas OSullivan, The De Excidio of
Gildas: Its Authenticity and Date (Leiden: Brill,
1978), 180.
2. N.J. Higham, The English Conquest:
Gildas and Britain in the Fifth Century (Manchester: Manchester University Press, 1994),
141.
3. Gildas, De Excidio et Conquestu Britanniae, trans. J.A. Giles (London: Henry G.
Bohn, 1848), http://www.fordham.edu.
4. Rodney Castleden, King Arthur: The
Truth Behind the Legend (London: Routledge,
2000), 456.
5. Castleden, King Arthur, 46.
6. Higham, 456.
7. Castleden, King Arthur, 81.
8. Ibid., 82.
9. Frank D. Reno, Historic Figures of the
Arthurian Era ( Jefferson, NC : McFarland,
2000), 59.
10. Reno, 69.
11. Nennius.
12. Michael D. Costen, The Origins of
Somerset (Manchester, UK: Manchester University Press, 1992), 67.
13. Barbara Yorke, Wessex in the Early
Middle Ages (London: Continuum, 1995), 19
20.
14. Yorke, 20.
15. Ibid., 22.
16. Ibid., 23.
17. Ibid., 24.
18. Ibid., 267.
19. Ibid., 26.
20. Ibid., 249.
21. Gelling, The Effect of Man on the
Notes Chapter 8
Landscape : The Place-name Evidence in
Berkshire, in The Effect of Man on the Landscape: The Lowland Zone, ed. Susan Limbrey
and John G. Evans (London: Council for
British Archaeology, 1978), 124.
22. David James, Sorviodunum: A Review of the Archaeological Evidence, The
Wiltshire Archaeological and Natural History
Magazine 95 (2002).
23. Wessex Archaeology.
24. David McOmish, David Field, and
Graham Brown, The Field Archaeolog y of Salisbury Plain Training Area (Swindon, Wiltshire: English Heritage, 2002).
25. Reno, 59.
26. Anglo-Saxon Chronicle, trans. James
Henry Ingram (London: Everyman Press, 1912),
http://www.gutenberg.org.
27. Yorke, 369.
28. Judith McClure and Roger Collins,
The Ecclesiastical History of the English People
(Oxford, UK: Oxford University Press, 1999),
119.
29. Johannes Hoops and Heinrich Beck,
Reallexikon der germanischen Altertumskunde,
Volume 12 (BerlinNew York : Walter de
Gruyter, 1998), 51.
30. Daniel G. Russo, Town Origins and Development in Early England, c. 400 950 A.D.
(Westport, CT: Greenwood, 1998), 109.
31. Yorke, 2930.
32. Russo, 109110.
33. Keith Matthews, Whats in a Name?
Britons, Angles, Ethnicity and Material Culture from the Fourth to the Seventh Centuries. Heroic Age 4.1 (2001).
34. Reno, 164.
35. Reno, 64.
36. David Dumville, The West Saxon
Genealogical Regnal List and the Chronology
of Early Wessex, Peritia 4 (1985).
37. Yorke, 32.
38. Jeremy Haslam, A Middle Saxon Iron
Smelting Site at Ramsbury, Wiltshire, Mediaeval Archaeolog y 24 (1980): 58.
39. Haslam, 5960.
40. Haslam, 60.
41. Martin Biddle, Winchester, the Development of an Early Capital, in Vor- und
Frh-formen der europaischen Stadt im Mittelalter, ed. H. Jankuhn, W. Schlesinger and H.
Steuer (Gttingen: Vandenhoeck und
Ruprecht, 1973).
42. Haslam, 61.
201
43. Ibid., 63.

45. Patrick Sims-Williams, The Settlement of England in Bede and the Chronicle,
in Anglo-Saxon England, ed. Peter Clemoes,
Simon Keynes and Michael Lapidge (Cambridge: Cambridge University Press, 2007),
334.
46. Michael Swanton, Anglo-Saxon Chronicle (London: Routledge, 1998), 20.
48. Ibid., 327.
49. Green, Symbol and Image, 186.
50. Yorke, 58.
51. Russo, 112.
52. Ibid., 1123.
53. Russo, 113.
54. Yorke, 27.
55. Barry Cunliffe, Understanding Hillforts, 162.
56. William A. Chaney, The Cult of Kingship in Anglo-Saxon England: The Transition
from Paganism to Christianity (Manchester,
UK: Manchester University Press, 1970), 127
8.
57. This is thought, like the comet of 530,
to have been Halleys Comet.
58. Christopher Daniell, From Norman
Conquest to Magna Carta (London: Routledge,
2003), 13.
59. Daniell, 17.
60. Ibid., 28.
61. Ibid., 29.
62. Ibid., 30.
63. Ibid., 37.
64. Berenice M. Kerr, Religious Life for
Women: c. 1100 c. 1350 (Oxford, UK: Oxford
65. Kerr, 702.
67. Clare Gathercole, Glastonbury Archaeological Assessment (Swindon, Wiltshire, UK:
English Heritage), 5.
68. Gathercole, 6.
69. Ibid., 23.
70. William of Malmesbury, Chronicle of
the Kings of England, trans. Rev. John Sharpe
(London: George Bell and Sons, 1904), Chapter 2, http://www.fordham.edu.
71. Gathercole, 289.
72. Ibid., 6.
73. Ibid., 67.
Bibliography
buch. Strassburg: K.J. Trbner, 1904.
http://www.archive.org.
Baudis, Joseph. Mabinogion. Folk-Lore
37 (1916): 3168.
Beekes, Robert. Greek Etymological Dictionary. http://www.indo-european.nl.
_____. The Origins of the Etruscans. http://
www.knaw.nl.
Bell-Fialkoff, Andrew. The Role of Migration
in the History of the Eurasian Steppe. New
York: Palgrave Macmillan, 2000.
Bhagavadgita. Translated by Sanderson
Beck. http://www.san.beck.org.
Biddle, Martin. Winchester, the Development of an Early Capital. In Vor- und
Friih-formen der europaischen Stadt im
Mittelalter. Edited by H. Jankuhn, W.
Schlesinger and H. Steuer. Gttingen,
Germany: Vandenhoeck und Ruprecht,
1973.
Blamires, David. Herzog Ernst and the Otherworld Voyage. Manchester, UK: Manchester University Press, 1979.
Boekhoorn, Dimitri Nikolai. Mythical,
Legendary and Supernatural Bestiary in
Celtic Tradition: From Oral to Written
Literature. PhD diss., Universit Rennes
2/University College Cork, 2008.
Bostock, John, and H.T. Riley, trans. The
Natural History of Pliny. Vol. 5. London:
Henry G. Bohn, 1857.
Bouzek, Jan. Cimmerians and Early Scythians: the Transition from Geometric to
Orientalising Style in the Pontic Area. In
North Pontic Archaeology: Recent Discoveries
and Studies. Edited by Gocha S. Tsetskhladze. Leiden, Holland: Brill, 2001.
Boyce, Mary, and Frantz Grenet. Zoroastrianism under Macedonian and Roman
Rule. Part 1. Leiden, Holland: Brill, 1991.
Achebe, Chinua. Things Fall Apart. London: Heinemann, 1996.

Allen, Richard Hinckley. Star Names: Their
Lore and Meaning. Mineola, NY: Dover
Publications, 1963.
Ammianus Marcellinus. Roman History.
Translated by J.C. Rolfe. Cambridge,
MA: Harvard University Press, 1950.
http://penelope.uchicago.edu.
Anderson, Graham. King Arthur in Antiquity. London: Routledge, 2004.
Anglo-Saxon Chronicle. Translated by James
Henry Ingram. London: Everyman
Press, 1912. http://www.gutenberg.org.
Anttila, Raimo. Greek and Indo-European
Etymolog y in Action. Amsterdam: John
Benjamins, 2000.
Apollonius Rhodius. Argonautica. Translated by R.C. Seaton. Cambridge, MA:
Harvard University Press, 1912. http://
www.omacl.org.
Archibald, Elizabeth. Incest and the Medieval
Imagination. Oxford, UK: Oxford University Press, 2001.
Athenaeus. Deipnosophists. Translated by
C.D. Yonge. London: H.G. Bohn, 1854.
http://www.attalus.org.
Bailey, H.W. Arya II. BSOAS 23.1 (1960):
1339.
Baillie, M.G.L. Irish Tree Rings and an
Event in 1628 B.C. In Thera and the
Aegean World III. Edited by David A.
Hardy. London: Thera Foundation, 1990.
_____. A Slice Through Time: Dendrochronology and Precision Dating. London:
Routledge, 1995.
Barrett, John, and David McOmish. All
Cannings Cross. http://www.allcannin
gs.org.
Bartholomae, Christian. Altiranisches Wrter-
203
204
Bibliography
Bray, Daniel. Sacral Elements of Irish Kingship. http://escholarship.usyd.edu.au.

Brehaut, Ernest. An Encyclopedist of the
Dark Ages: Isidore of Seville. New York:
Columbia University, 1912. http://besti
ary.ca.
Bremmer, Jan. The Rise and Fall of the Afterlife. London: Routledge, 2002.
Bromwich, Rachel. Trioedd Ynys Prydein.
The Welsh Triads. Cardiff: University of
Wales Press, 1978.
Brown, Norman Oliver. Hermes the Thief:
The Evolution of a Myth. Great Barrington, MA: Steiner Books, 1990.
Bryce, Trevor. The Kingdom of the Hittites.
Oxford, UK: Oxford University Press,
2005.
Carey, John. A British myth of origins?
History of Religions 31 (1991): 2438.
Cassius Dio. Roman History. Translated by
Earnest Cary. Cambridge, MA: Harvard
University Press, 1927. http://penelope.
uchicago.edu.
Castleden, Rodney. King Arthur: The Truth
Behind the Legend. London: Routledge,
2000.
_____. The Making of Stonehenge. London:
Routledge, 1993.
Centre for the Study of Ancient Documents.
Curse Tablets from Roman Britain: Archaeological Sites. http://curses.csad.
ox.ac.uk.
Chaney, William A. The Cult of Kingship in
Anglo-Saxon England: The Transition from
Paganism to Christianity. Manchester,
UK: Manchester University Press, 1970.
Clancy, Joseph, trans. Y Gododdin. In
Earliest Welsh Poetry. London & New
York: Macmillan, 1970. http://www.ma
ryjones.us.
Clarke, Giles. The Roman Villa at Woodchester. Britannia 13 (1982): 197228.
Clement of Alexandria. Stromata. Translated by William Wilson. Buffalo, NY:
Christian Literature Publishing, 1885.
http://earlychristianwritings.com.
Cole, Susan Guettel. Greek Religion. In
A Handbook of Ancient Religions. Edited
by John R. Hinnells. Cambridge, UK:
Cambridge University Press, 2007.
Coles, John M., and A.F. Harding. The
Bronze Age in Europe. London: Taylor &
Francis, 1979.
Costen, Michael D. The Origins of Somerset.

Manchester, UK: Manchester University
Press, 1992.
Cunliffe, Barry. Iron Age Communities in
Britain. 4th ed. London: Routledge,
2005.
_____. Landscape with People. In Culture, Landscape and the Environment. Edited by Kate Flint and Howard Morphy.
Oxford, UK: Oxford University Press,
2000.
_____. The Oxford Illustrated History of Prehistoric Europe. Oxford, UK: Oxford
University Press, 2001.
_____. Understanding Hillforts: Have We
Progressed? In The Wessex Hillforts Project. Edited by Andrew Payne, Mark Corney and Barry Cunliffe. Swindon, Wiltshire, UK: English Heritage, 2006.
Daniell, Christopher. From Norman Conquest to Magna Carta. London: Routledge, 2003.
DArbois de Jubainville, Henri. Les Druides
et les dieux celtiques forme danimaux.
Boston, MA: Adamant Media Corporation, 2005.
Darmesteter, James, trans. Vendidad. In
Sacred Books of the East. New York: Oxford University Press, 1880. http://www.
avesta.org.
Darvill, Timothy. Prehistoric Britain. London: Routledge, 1997.
Davidson, Hilda Roderick Ellis. Myths and
Symbols in Pagan Europe. Manchester,
UK: Manchester University Press, 1988.
Diodorus Siculus. Historical Library. Translated by C.H. Oldfather. Cambridge,
MA: Harvard University Press, 1935.
http://penelope.uchicago.edu.
Dumville, David. The West Saxon Genealogical Regnal Lists and the Chronology of Early Wessex. Peritia 4 (1985):
2166.
Dunbabin, Katherine. Mosaics of the Greek
and Roman World. Cambridge, UK:
Cambridge University Press, 1999.
Dunbavin, Paul. Picts and Ancient Britons.
Long Eaton, Nottingham, UK: Third
Millennium Publishing, 2001.
Dyer, James. Discovering Prehistoric England. Princes Risborough, Buckinghamshire, UK: Shire Publications, 2001.
Edwards, Iorwerth Eiddon Stephen. The
Bibliography
Cambridge Ancient History. Cambridge,
UK: Cambridge University Press, 1970.
Eichberg, Henning. The Labyrinth of the
City Fractal Movement and Identity.
In Nature and Identity: Essays on the Culture of Nature. Edited by Kirsti Pedersen
and Arvid Viken. Bloomington: Indiana
Eliade, Mircea. Shamanism: Archaic Techniques of Ecstasy. London: Routledge &
Kegan, 1964.
English Heritage. National Monument Report. SU14SW67, SU14SW68.
Evelyn-White, Hugh G. Hesiod: The Homeric Hymns and Homerica. Charleston,
SC: BiblioBazaar, 2008.
Falileyev, Alexander. Dictionary of Continental Celtic Place-Names. Aberystwyth,
UK: CMCS Publications, 2010. http://
cadair.aber.ac.uk.
Fee, Christopher R., and David A. Leeming. Gods, Heroes and Kings: The Battle
for Mythic Britain. Oxford, UK: Oxford
Field, David. Great sites: Silbury Hill.
British Archaeolog y 70 (2003). http://
www.britarch.ac.uk.
_____. The Investigation and Analytical Survey of Silbury Hill. Swindon, Wiltshire,
UK: English Heritage, 2002.
Ford, Patrick K. The Mabinogi and Other
Medieval Welsh Tales. Berkeley: University of California Press, 2008.
Frazer, Sir James George. The Golden Bough:
Studies in Magic and Religion. London:
Penguin, 1996.
Gathercole, Clare. Glastonbury Archaeological Assessment. Swindon, Wiltshire, UK:
English Heritage, 2003.
Geldard, Richard G. The Travelers Key to
Ancient Greece: A Guide to Sacred Places.
Wheaton, IL: Quest Books, 2000.
Gelling, Margaret. The Effect of Man on
the Landscape: The Place-name Evidence in Berkshire. In The Effect of Man
on the Landscape: The Lowland Zone. Edited by Susan Limbrey and John G.
Evans. London: Council for British Archaeology, 1978.
Geoffrey of Monmouth. History of the Kings
of Britain. Translated by J.A. Giles. London: Henry G. Bohn, 1848. http://www.
lib.rochester.edu.
205
_____. The Life of Merlin. Translated by

John Jay Parry. Urbana: University of
Illinois, 1925. http://www.sacred-texts.
com.
Gibbon, Edward. The History of the Decline
and Fall of the Roman Empire. Edited by
David Womersley. London: Penguin,
2000.
Gildas. De Excidio et Conquestu Britanniae.
Translated by J.A. Giles. London: Henry
G. Bohn, 1848. http://www.fordham.edu.
Giraldus Cambrensis. Itinerary Through
Wales. Translated by Sir Richard Colt
Hoare. London: W. Miller, 1806. http://
www.archive.org.
_____. The Journey through Wales; and, The
Description of Wales. Translated by Lewis
Thorpe. London: Penguin, 1978. http://
www.britannia.com.
_____. Topography of Ireland. Translated by
Thomas Forester. London: Henry G.
Bohn, 1863.
Graf, Fritz. Apollo. London: Taylor & Francis, 2008.
Green, Miranda. Animals in Celtic Life and
Myth. London: Routledge, 1998.
_____. An Archaeolog y of Images. London:
Routledge, 2004.
_____. Celtic Myths. Austin: University of
Texas Press, 1998.
_____. The Gods of Roman Britain. Princes
Risborough, Buckinghamshire, UK: Shire
Publications, 1983.
_____. Symbol and Image in Celtic Religious
Art. London: Routledge, 1992.
Green, Thomas. A Bibliographic Guide to
Welsh Arthurian Literature. http://www.
arthuriana.co.uk.
_____. Concepts of Arthur. Stroud: Tempus,
2007.
_____. The Historicity and Historicisation
of Arthur. http://www.arthuriana.co.uk.
Gruffydd, W.J. Mabon ab Modron.
Revue Celtique 33 (1912): 459460.
Guest, Lady Charlotte, trans. The Mabinogion. London: J.M. Dent, 1910. http://
www.maryjones.us.
Harding, A.F. European Societies in the
Bronze Age. Cambridge, UK: Cambridge
Haslam, Jeremy. A Middle Saxon Iron
Smelting Site at Ramsbury, Wiltshire.
Mediaeval Archaeolog y 24 (1980): 168.
206
Bibliography
Hastings, James. Encyclopedia of Religion and

Ethics. Whitesh, MT: Kessinger, 2003.
Healy, John F. Natural History: A Selection.
London: Penguin, 1991.
Henig, Martin. Religion in Roman Britain.
London: Routledge, 1984.
Herodotus. The Histories. Translated by
A.D. Godley. Cambridge, MA: Harvard
University Press, 1920. http://old.pers
eus.tufts.edu.
_____. The Histories. Translated by George
Rawlinson. New York: Dutton, 1862.
http://www.fordham.edu.
Hesiod. Theogony. Translated by Hugh G.
Evelyn-White. Cambridge, MA: Harvard University Press, 1912. http://www.
sacred-texts.com.
Higham, N.J. The English Conquest: Gildas
and Britain in the Fifth Century. Manchester, UK: Manchester University Press,
1994.
Higley, Sarah. Between Languages: The Uncooperative Text in Early Welsh and Old
English Nature Poetry. University Park,
PA: Penn State Press, 1993.
_____, trans. The Spoils of Annwn. http://
www.lib.rochester.edu.
Hinnells, John R. Mithraic Studies. Manchester, UK: Manchester University Press, 1975.
Hippolytus of Rome. Philosophumena.
Translated by F. Legge. London: Society
for Promoting Christian Knowledge,
Macmillan, 1921. http://www.archive.org.
Homer. The Iliad. Trans. Samuel Butler.
http://classics.mit.edu.
Hoops, Johannes, and Heinrich Beck. Reallexikon der Germanischen Altertumskunde. Vol. 12. BerlinNew York: Walter
de Gruyter, 1998.
Hubert, Henri. Le mythe dEpona. In
Mlanges linguistiques offerts M.J.
Vendryes. Edited by Jules Bloch. Paris:
Librairie Ancienne Edouard Champion,
1925.
Hunter-Mann, Kurt. Excavations at Vespasians Camp Iron Age Hillfort, 1987.
The Wiltshire Archaeological and Natural
History Magazine 92 (1999): 3952.
James, David. Sorviodunum: A Review of
the Archaeological Evidence. The Wiltshire Archaeological and Natural History
Magazine 95 (2002): 126.
John, Brian. The Bluestone Enigma. New-
port, Pembrokeshire, UK: Greencroft

Books, 2008.
Johnston, Ian. The Iliad. Arlington, VA:
Richer Resources Publications, 2006.
Jones, Mary. Notes to The Battle of the
Trees. http://www.maryjones.us.
Julius Caesar. Gallic Wars. Translated by
W.A. McDevitte and W.S. Bohn. New
York: Harper & Brothers, 1869. http://
classics.mit.edu.
Kanga, K.E. Avesta Dictionary. Bombay:
Education Society, 1900. http://www.
avesta.org.
Kazanas, N.D. Indo-European Deities
and the Rigveda. Journal of Indo-European Studies 29 (2001): 257293.
Kerr, Berenice M. Religious Life for Women:
c. 1100 c. 1350. Oxford, UK: Oxford
Koch, John T. Celtic Culture: A Historical
Encyclopedia. Santa Barbara, CA: ABCCLIO, 2006.
_____. A Welsh Window on the Iron Age:
Manawydan, Mandubracios. CMCS 14
(1987): 1752.
Kondratiev, Alexei. Lugus: The ManyGifted Lord. An Trbhs Mhr: The
IMBAS Journal of Celtic Reconstructionism
1 (1997). http://www.imbas.org.
Kristiansen, Kristian. Europe Before History.
Cambridge, UK: Cambridge University
Press, 2000.
Lewis, Jodie. Upwards at 45 Degrees: The
Use of Vertical Caves during the Neolithic and Early Bronze Age on Mendip,
Somerset. Capra 2 (2000). http://capra.
group.shef.ac.uk.
Littleton, C. Scott, and Linda A. Malcor.
From Scythia to Camelot. London: Taylor
& Francis, 2000.
_____. The Germanic Sword in the Tree.
The Heroic Age 11 (2008).
Loomis, Roger Sherman. The Grail: From
Celtic Myth to Christian Symbol. Princeton: Princeton University Press, 1991.
Lucan. Pharsalia. Translated by Sir Edward
Ridley. London: Longmans, Green, 1896.
http://www.omacl.org.
Lurker, Manfred. The Routledge Dictionary
of Gods and Goddesses, Devils and Demons.
Mac Cana, Proinsias. Celtic Mytholog y.
New York: Hamlyn, 1970.
Bibliography
Magnusdottir, Asdis R. La Voix du cor.
Amsterdam: Rodopi, 1998.
Mallory, J.P., and Douglas Q. Adams. Encyclopedia of Indo-European Culture.
London: Taylor & Francis, 1997.
Matasovic, Ranko. Etymological Lexicon of
Proto-Celtic. http://www.indo-european.
nl.
Matthews, Keith. Whats in a Name?
Britons, Angles, Ethnicity and Material
Culture from the Fourth to the Seventh
Centuries. Heroic Age 4.1 (2001).
Matthews, W.H. Mazes and Labyrinths.
Charleston, SC: Forgotten Books, 2008.
McClure, Judith, and Roger Collins. The
Ecclesiastical History of the English People.
Oxford: Oxford University Press, 1999.
McEvilley, Thomas. The Shape of Ancient
Thought. New York: Allworth Press, 2002.
McOmish, David. East Chisenbury: Ritual and Rubbish at the British BronzeAge-Iron Age Transition. Antiquity 70.
267 (1996): 6876.
_____. Landscapes Preserved by the Men
of War. British Archaeolog y 34 (1998),
http://www.britarch.ac.uk.
McOmish, David, David Field, and Graham Brown. The Field Archaeolog y of the
Salisbury Plain Training Area. Swindon,
Wiltshire, UK: English Heritage, 2002.
Merkur, Dan. Fruit of the Terrestrial Paradise: The Psychedelic Sacrament in St.
Ephrem the Syrian and Celtic Christianity.
http://www.danmerkur.com.
The Metrical Dindshenchas. Edited by Edward Gwynn. Dublin, Ireland: Dublin
Institute for Advanced Studies, 1991.
http://www.ucc.ie.
Metzner-Nebelsick, Carola. Early Iron
Age Pastoral Nomadism in the Great
Hungarian Plain Migration or Assimilation? The ThracoCimmerian Problem Revisited. In Kurgans, Ritual Sites
and Settlements. Edited by Jeannine
Davis-Kimball. http://www.csen.org.
Millar, F. Graham. The Celestial David
and Goliath. Journal of the Royal Astronomical Society of Canada 89.4 (1995):
141154.
Needham, Stuart, Andrew Lawson, and
Ann Woodward. Rethinking Bush Barrow. British Archaeolog y 104 (2009).
http://www.britarch.ac.uk.
207
Nennius. History of the Britons. Translated

by J.A. Giles. London: Henry G. Bohn,
1848. http://www.fordham.edu.
O Hogain, Daithi. The Celts: A History.
Part 70. Woodbridge, Suffolk: Boydell
Press, 2003.
Old Iranian Online, www.utexas.edu.
OMeara, John. The History and Topography
of Ireland. London: Penguin, 1982.
Oosten, J.G. The War of the Gods: The
Social Code in Indo-European Mytholog y.
OSullivan, Thomas D. The De Excidio of
Gildas: Its Authenticity and Date. Leiden,
Holland: Brill, 1978.
The Parish and Priory Church of St Mary,
Totnes, Devon (Diocese of Exeter),
http://www.churchcare.co.uk.
Parker, Will. Mabinogi. http://www.mabin
ogi.net.
Parker-Pearson, Mike. Stonehenge Riverside
Project. http://www.shef.ac.uk/archaeol
ogy/research/stonehenge.
Pausanias. Description of Greece. Translated
by W.H.S. Jones. Cambridge, MA: Harvard University Press, 1918. http://www.
theoi.com.
Perry, Charles, A., and Kenneth Hsu.
Geophysical, Archaeological, and Historical Evidence Support a Solar-Output
Model for Climate Change. PNAS
97.23 (2000): 1243312438.
Pettazzoni, Raffaele. The All-Knowing God.
New York: Arno Press, 1978.
Pitts, Mike. Hengeworld. London: Arrow
Books, 2000.
Plimer, Ian. Heaven and Earth: Global
Warming: The Missing Science. Ballan,
Victoria, Australia: Connor Court, 2009.
Plutarch. Life of Theseus. In Parallel
Lives. Translated by Aubrey Stewart and
George Long. London: Henry G. Bohn,
1881. http://www.gutenberg.org.
Pokorny, Julius. Indogermanisches Etymologisches Wrterbuch. Bern, Switzerland:
Francke, 1959. http://www.indoeuropean.nl.
_____. Zur keltischen Namenkunde und
Etymologie. Vox Romanica 10 (1949):
220267.
Prehistoric Dartmoor, http://www.dart
moor-npa.gov.uk.
Price, Dennis. The Missing Years of Jesus:
208
Bibliography
The Greatest Story Never Told. London:

Hay House, 2009.
Rankin, David. Celts and the Classical World.
Reno, Frank D. Historic Figures of the
Arthurian Era. Jefferson, NC: McFarland, 2000.
Rhys, John. Celtic Folklore: Welsh and Manx.
Boston, MA: Adamant Media Corporation, 2004.
Rigveda. Translated by Ralph Grifth.
Benares, India: E.J. Lazarus, 1897.
http://www.sacred-texts.com.
Romer, Frank E. Pomponius Melas Description of the World. Ann Arbor: University
of Michigan Press, 1998.
Ross, Anne. Ritual and the Druids. In
The Celtic World, edited by Miranda
Green. London: Routledge, 1996.
Russo, Daniel G. Town Origins and Development in Early England, c. 400 950
A.D. Westport, CT: Greenwood, 1998.
Sandars, Nancy K. Bronze Age Cultures in
France. Cambridge, UK: Cambridge
Sauter, Hermann. Denition der thrakokimmerischen Bronzen. http://www.ki
mmerier.de.
Sciama, Lidia D., and Joanne Bubolz
Eicher. Beads and Bead Makers. Oxford,
UK: Berg Publishers, 1998.
The Sick-Bed (Wasting Sickness) of
Cuchulain. In Cuchulain of Muirthemne.
Translated by Lady Augusta Gregory.
London: John Murray, 1902. http://
Silver, Morris. Taking Ancient Myths Economically. Leiden, Holland: Brill, 1992.
Sims-Williams, Patrick. The Settlement
of England in Bede and the Chronicle.
In Anglo-Saxon England. Edited by Peter
Clemoes, Simon Keynes, and Michael
Lapidge. Cambridge, UK: Cambridge
Skene, W.S., trans. The Four Ancient Books
of Wales. Edinburgh: Edmonston and
Douglas, 1868. http://www.maryjones.
us.
Skjrv, Prods Oktor. Old Avestan Glossary.
http://www.fas.harvard.edu.
Spence, Lewis. The Mysteries of Britain.
Pomeroy, WA: Health Research Books,
1996.
Strabo. Geography. Translated by H.L. Jones.

Cambridge, MA: Harvard University
Press, 1932. http://penelope.uchicago.edu.
Swanton, Michael. Anglo-Saxon Chronicle.
Tacitus. Annals. Translated by Alfred John
Church and William Jackson Brodribb.
New York: Random House, 1942. http://
www.perseus.tufts.edu.
Taylor, Thomas, trans. Life of St. Samson of
Dol. London: Society for Promoting
Christian Knowledge, Macmillan, 1925.
http://www.lamp.ac.uk.
Thomas, Graham. The Romans at Woodchester. http://grahamthomas.com.
Thomas, Julian. Understanding the Neolithic.
Time Team. The Bone Cave, Alveston,
Gloucestershire. http://www.channel4.
com/ programmes/time-team/episodeguide/series-8/episode-8.
_____. Durrington Walls: A Time Team
Special. http://www.channel4.com/his
tory/microsites/T/timeteam/2005_durr_
t.html.
Toynbee, Jocelyn. Death and Burial in the
Roman World. Baltimore, MD: JHU
Press, 1996.
Turcan, Robert. The Cults of the Roman
Empire. Oxford, UK: Blackwell, 1996.
Vandenberg, Philipp. Mysteries of the Oracles. London: Tauris Parke Paperbacks,
2007.
Wagner, Heinrich. Studies in the Origins of
the Celts and of Early Celtic Civilization.
Tbingen, Germany: Niemeyer, 1971.
Waterhouse, John. Zoroastrianism. San
Diego, CA: Book Tree, 2006.
Watts, Dorothy. Christians and Pagans in
Roman Britain. London: Taylor & Francis, 1991.
Webster, Jane. Sanctuaries and Sacred
Places. In The Celtic World. Edited by
Miranda Green. London: Routledge,
1996.
Wessex Archaeology. Remarkably preserved Roman remains from grave.
http://www.wessexarch.co.uk/projects/w
iltshire/boscombe/preserved-roman-re
mains/index.html.
West, Martin Litcheld. Indo-European Poetry and Myth. Oxford, UK: Oxford
Bibliography
What man is the gatekeeper/porter?
Translated by Mark Adderley. http://
www.markadderley.net.
William of Malmesbury. Chronicle of the
Kings of England. Translated by the Rev.
John Sharpe. London: George Bell and
Sons, 1904. http://www.fordham.edu.
Williams, Ifor. Pedeir Keinc y Mabinogi.
Cardiff, UK: Gwasg Prifysgol Cymru,
1974.
Williams-Thorpe, O., P.J. Potts, M.C.
Jones, and P.C. Webb. Preseli Spotted
Dolerite Bluestones: Axe-heads, Stonehenge Monoliths and Outcrop Sources.
Oxford Journal of Archaeolog y 25 (2006):
2964.
Wiltshire and Swindon Sites and Monument Record Information: Nettleton
Shrub, http://history.wiltshire.gov.uk/
smr/getsmr.php?id=2096.
209
Wiltshire Council. Sites and Monuments

Record: Wilsford Shaft, Normanton
Down. http://history.wiltshire.gov.uk.
Wood, Michael. In Search of the Trojan
War. Berkeley: University of California
Press, 1998.
Woodard, Roger D. Indo-European Sacred
Space. Champaign: University of Illinois
Press, 2006.
Wright, Craig M. The Maze and the Warrior. Cambridge, MA: Harvard University Press, 2001.
Yeates, Stephen James. The Tribe of Witches:
The Religion of the Dobunni and the
Hwicce. Oxford, UK: Oxbow Books,
2009.
Yorke, Barbara. Wessex in the Early Middle
Ages. London: Continuum, 1995.
Zalta, Edward N., ed. Stanford Encyclopedia
of Philosophy. http://plato.stanford.edu.
Index
Aeneas 114
Ahura-Mazda 156, 158
All Cannings Cross (pottery) 3, 104, 108,
113, 128, 130131, 154, 167
Allen, Richard Hinckley 2, 144
Alton Priors 183
Alveston Cave 3031
amber 9798, 141, 153
Ambrosius Aurelianus 169, 170175, 187,
188
Amesbury 171, 172, 174, 175, 179, 186,
188, 190, 192
Amesbury Archer 9293, 95
Ammianus Marcellinus 1617, 146
Anderson, Graham 2, 143144, 145, 147,
149
Anglo-Saxon Chronicle 64, 171, 176, 180,
188
Annwn 7, 4263, 67, 77
Apam Napat 46, 154, 156
Arawn 67, 77
Arcadia 139, 142, 145
Arcas/Arkas 1, 145, 166
Arcturus (Botes) 1, 143145, 147, 149,
151, 142, 153, 154, 162, 166, 168, 170,
175, 187, 192
Arianrhod 7784
Aristotle 60
Artorius 143
Atrebates 111, 186
Avebury Stone Circle 86
Avesta 46, 156
Bhagavadgita 165
Blodeuedd 80, 82, 94, 190
Bokerley Dyke 173
Bolgios/Belgius 74, 85
Boscombe Bowmen 93, 95
Boudica (Boadicea) 1920
Bran 47, 52, 55, 58, 6974, 162
Branwen 66, 6974
Brennos/Brennus 7273, 85
Britain (etymology) 69, 118119
broom (plant) 80, 190
Brutus of Troy 114117, 118, 156
bull, cult 39, 5556, 109
Bush Barrow 9697
Buttereld Down 31
Cabal (Arthurs dog) 7, 9
Cadbury Castle 172173, 179, 186
Cadbury Congresbury 172, 173
Caer Sidi 4344, 52, 59, 61, 63, 162
Cai (Sir Kay) 89, 49
Calleva Atrebatum 186
Caratacus 19, 176, 179
Cassiopeia 82
Cassius Dio 20, 125
Cassivellaunus 18, 72, 73, 74, 81, 85
cauldron 9, 36, 41, 47, 49, 50, 51, 6162,
64, 7071, 72, 75, 106, 126, 146147,
148, 161162
Ceawlin 181, 182183
Celts 1, 2, 14, 16, 3738, 47, 53, 5758,
73, 74, 109110, 112, 124, 150, 154
Cenwalh 184185
Cerdic 176, 179, 180, 188
Cernunnos 25, 3738, 150151
chthonic deities 108, 131
Cimmerians 126130
Cirencester/Corinium 22, 27, 3334, 40,
64, 183, 184
Clement of Alexandria 14
comet 6, 176, 180, 187, 188
Coranians/Coraniaid 121123, 125
Barbury Castle 181

bards 12, 78, 124, 138
Bath 3536, 64, 183
Beaker culture 87, 9394, 9596
Bede 55, 64, 124, 169, 176, 177
Bedwyr (Sir Bedivere) 89
Belenus 61
Belgae 18, 74, 111112, 185
Beli Mawr 7374, 81, 121
211
212
Index
Corona Borealis 82, 151, 152, 153

Crete 98, 102, 159160
Culhwch and Olwen 810, 36, 45, 47, 51,
56, 66, 152, 155
Cunetio 34, 181182
Cunliffe, Barry 2, 3, 87, 9394, 100101,
104, 105106, 108111, 112, 129, 130
131, 161, 186
Cunobelinus 19, 74
Cunomaglus 27, 35, 180
Cynegils 184
Cynric 176, 179, 180, 184, 188
Dagda 76, 135, 150
Danebury 2, 100, 107110, 128, 130131,
134, 139, 186
Dartmoor 101, 116117, 118, 157, 167
Delphi 7273
Diodorus Siculus 1213, 7273, 112, 129
Dionysus 34, 134
Dis (Pater) 12, 16
Dobunni 30, 3536, 49, 64, 111, 186
Dodona 1, 135, 154, 166
dog, burials 3, 3031, 109
dog, cult 27, 28, 133, 139140
Don (Mabinogion) 66, 77, 85, 113
door (to Otherworld) 5253, 62, 72,
141142, 162
Dorchester-on-Thames 172, 177178,
179, 181, 183, 184, 185
dragon 5, 46, 121122, 171, 188
Durotriges 111112
Durrington Walls 9495
eagle 9, 80, 155
East Chisenbury 104105, 107, 128, 129,
167
Egypt, ancient 103
Elidorus 119120, 157, 160
Epona 41, 45, 6869, 140
Etruscans 103, 117118, 122, 142, 154, 157,
159
Excalibur 145
eld-systems 100102, 103, 116, 167
Four-Cornered (Fortress) 4647, 52, 61
63, 141, 156, 162
Gawain 8, 148149
genii cucullati 3536, 38, 41, 49, 65
Geoffrey of Monmouth 2, 5, 10, 48, 86,
114, 115117, 120, 158, 161, 189
Gewisse 177, 179, 180, 181, 183, 184, 186
Giants Dance 86, 161
Gibbons Decline and Fall 3, 6
Gildas 64, 168
Giraldus Cambrensis 54, 6162, 114, 119

120, 135, 157
Glass (Fortress) 5355, 62
Glastonbury 54, 189, 190192
Gloucester/Glevum 9, 22, 35, 4849,
64, 69, 183
Greeks, ancient 2
Green, Miranda 2, 35, 39, 41
Green, Thomas 2, 7, 143
Guinevere 8
Gundestrup Cauldron 36, 47, 55, 129,
151
Gwales 52, 72
Gweir 4345, 61, 63, 76
Gwydion 7784, 190
Hengist 86, 179
Hermes 12, 134, 135, 142, 157, 187
Herodotus 102, 103, 117, 126, 134, 140,
141, 145, 146, 148, 167
Hesiod 2, 140, 142, 143, 144, 148, 155
Higley, Sarah 43, 46, 51, 53, 55, 5758,
82
hillforts 101, 106111, 115, 128, 133, 136,
172174
Hippolytus 14, 59
Hittites 98, 102, 117
Holy Grail 4950, 146147, 149
Homer 2
horse, burials 3, 109
horse, cult 41, 6869, 70, 131132, 135,
166
Iliad 114, 126, 132, 137, 141, 159160,
166
Illyrian 118, 136137, 153
Indo-European 3, 44, 47, 56, 59, 77,
118
initiation 5863
Iranian languages 118, 125, 126, 143, 153,
157, 161, 167
Ireland 7071, 161
Iron House 71, 75, 126, 161
Janus 142
Jordan Hill 2930, 72
Julius Caesar 2, 1012, 18, 24, 38, 57,
125
Koch, John 2, 3, 14, 55, 67, 7374, 77,
125, 166, 182, 190
labyrinths 158161
Life of St Samson 4748
Littlecote 34
Livy 149
Index
Lleu (Mabinogion) 7982, 121, 153
Llud Llaw Eraint 8, 56, 188, 192
Lludd and Llefelys 66, 74, 114, 121126,
149, 188
Llyr 8, 52, 59, 66, 6974, 85, 156
Lucan 38
Lugus/Lugh 2326, 2930, 51, 72, 82,
135, 149, 152, 153, 187, 192
Lydia/Lydians 117118
Lydney 2627
Mabinogion 1, 2, 8, 18, 36, 45, 46, 52,
55, 58, 6685, 94, 121, 149, 153, 161
Mabon/Maponos 89, 26, 39, 44, 66,
69, 155
Macedonians, ancient 2, 139, 148149
Mahabharata 155156, 165
Maiden Castle 2829
Manawydan 8, 66, 6974, 7477, 126,
161
Mandubracius 18, 73, 85
Matasovic, Ranko 3, 67, 70, 154
Math (Mabinogion) 66, 7784, 125, 158,
190
McEvilley, Thomas 2, 163165
mead 5152, 62, 122
meadowsweet 80, 94
Medea 148, 162
Mercury 12, 2226, 27, 30, 35, 37
Merlin 5, 48, 192
middens 103105, 107, 113, 128, 129, 136,
167
Milky Way 82
Millar, F. Graham 2, 149152
Modron/Matrona 9, 3941, 64
Mona/Anglesey 19
Mons Badonicus 6, 171, 172, 180
Mound-People 6769, 7477, 85, 112
113, 117, 153, 157, 160, 162
Mycenae/Mycenean 97, 102, 136, 141,
166, 167
nemeton 166
Nennius 2, 6, 42, 54, 114, 120, 123, 170
171, 179
Nettleton Shrub 2728, 35, 180
nine maidens 8, 4749, 61, 65
Nodens 8, 26, 56, 121, 188, 192
Normanton Down 9596, 97, 102
oak 15, 79, 80, 82, 137138, 142, 154,
162
Old Sarum 175, 179, 180
Orpheus/Orphism 3235, 6061, 163,
164165
Ossetia/Ossetic 118, 143, 147148
213
Pa gur 78, 47, 162

Pagans Hill 28
Parker, Will 66, 76
Pausanias 73, 138, 141
Perceval 48, 4950
Peredur, Son of Efrawg 48, 66
Pettazzoni, Raffaele 2, 46, 133, 149
Picts 124125
pits, ritual 108109, 130131
Plato 157, 163
Pliny the Elder 2, 15, 55, 71
Plutarch 129, 160
Pokorny, Julius 3, 44, 53, 59, 118119,
157, 160
Pomponius Mela 14, 16, 47, 49
Potterne 104104, 107, 128, 129, 167
precession 150151
Price, Dennis 112113, 161
Pryderi 43, 45, 61, 66, 6769, 7477,
7778, 81, 118, 157
Prydwen 9, 45
Pwyll 43, 45, 66, 6769, 124, 158
Pythagoras/Pythagorean 14, 5960, 139,
140141, 142, 162163, 164165
Pytheas of Massilia 112
Ramsbury 34, 181182
raven, cult 29, 72, 140
reincarnation 14, 82, 84, 142, 154, 162
166
Reno, Frank D. 2, 170, 175, 179, 192
Rhesus 132, 134, 162, 187
Rhiannon 66, 6769, 7477, 124, 135,
162
Rig-Veda (Vedas) 44, 46, 131, 135, 138,
139, 156, 160, 162
Rosmerta 25, 27, 35, 37
Round Table 145
Salisbury Plain 100, 101, 102, 104, 116,
175, 176, 185
Santorini (volcano) 9899
Sarmatians 143
Scythians 2, 118, 124, 126, 134, 142, 143,
145146, 149, 167
Sea Peoples 103
shafts, ritual 29, 3132, 72, 96, 99
100
shoemakers 24, 7576, 81, 82
sidi/sidh 44, 67, 76, 86, 117, 156
Silbury Hill 86, 8890
Silures 19, 64
Skene, W.F. 43
sovereignty goddess 52, 68
Spoils of Annwn 1, 2, 7, 36, 4265, 67,
141, 146, 153, 156, 162
214
Index
Stonehenge 1, 86, 9097, 98, 99, 100,

102, 112113, 142, 154, 161
Strabo 1314, 73, 129, 150
Sword in the Stone 145
underworld cult 31
Uther Pendragon 3, 56, 17
Varuna 46, 59, 154, 156, 157
vates 14, 138139
Vespasians Camp 102, 112113, 175
Viroconium 20, 169, 175, 180, 187
volcanic eruptions 9899, 101, 187
Vortigern 123, 168, 169, 170175, 179,
187
Tacitus 2, 19
Taliesin 7, 4243, 47, 52, 56, 59, 6365,
71, 8284, 156, 165
Taranis 3839
Tarvus Trigaranus 29, 38, 55
La Tne culture 109110
Things Fall Apart 58
Thracian 2, 14, 36, 126130, 142, 153
Thracian Horseman 1, 68, 130135, 137,
139, 162, 187
Timagenes 16, 73
tor 117118, 157
Totnes 115116, 167
Troy (city) 1617, 102, 117, 120, 158,
167
Troy (dance) 159160
Troy Town (maze) 158159
Troynt (boar) 7, 9
Tuatha de Danann 8, 77, 84
Twrch Trwyth (boar) 910
Y Gododdin 7, 42, 5152, 166

Yorke, Barbara 2, 173, 174, 177, 178, 180,
186
Ysbaddaden 89, 152153, 187
Uley 2223, 3536, 64, 86, 192
Zoroastrianism 14, 46, 154, 156, 158, 161
Wansdyke 173174, 176

Welsh Triads 4041, 44, 74, 81, 114, 123
Wessex, kingdom of 1, 3, 168
West Kennet Long Barrow 8688
Whitsbury Castle 174
William of Malmesbury 191
Winchester 5, 40, 112, 177, 179, 185186,
187, 190
Woodchester 33

The Druids and King Arthur

Uploaded by

Copyright:

Available Formats

The Druids and King Arthur

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Druids and King Arthur

Uploaded by

Copyright:

Available Formats

The Druids

and King Arthur

This page intentionally left blank

The Druids and

McFarland & Company, Inc., Publishers

Illustrations by Diana Gardner

CONGRESS CATALOGUING-IN-PUBLICATION DATA

McFarland & Company, Inc., Publishers

Chapter 1 The Dragon Star

Chapter 2 The Severed Head and the Bone Cave:

Chapter 3 Arthurs Voyage: The Spoils of Annwn

Chapter 4 Magic Mounds, Sea People and Shape-Shifters:

Chapter 5 Mounds, Mounds, Mounds: Rubbish Heaps,

Chapter 6 Visitors from the East

Chapter 7 Brutus of Troy Town

Chapter 8 Arthur, King of Wessex?

This page intentionally left blank

In researching this book, I have drawn on the works of a large number

This page intentionally left blank

The Dragon Star