This statuette of a bearded and mustachioed Gallic warrior, wearing a torc around his neck (c.

0
BC/AD), was found at St Maur-en-Chaussée, Oise, France.
About the Author
Jean Manco is the author of Ancestral Journeys: The Peopling of Europe
from the First Venturers to the Vikings, published by Thames & Hudson
in 2013. This ground-breaking work made accessible to the general
reader the revolution in archaeology, historical studies and genealogy
now that theories can be directly tested by investigating the DNA not only
of living people, but also of those long past.
 Other titles of interest published by
Thames & Hudson include:

Ancestral Journeys: The Peopling of Europe from the First Venturers to

the Vikings

Bog Bodies Uncovered: Solving Europe’s Ancient Mystery

The Celtic Myths: A Guide to the Ancient Gods and Legends

Exploring the World of the Celts

The Historical Atlas of the Celtic World

See our websites

www.thamesandhudson.com
www.thamesandhudsonusa.com
For my sons Tristan and Simon
Contents

Prologue
Timeline
1 The Voices of the Celts
2 The Gauls and Celtic
3 Bell Beakers and Language
4 The Indo-European Family
5 Stelae to Bell Beaker
6 The Iron Sword
7 On the Move
8 Celts vs Romans
9 Christian Celts
10 Loss and Revival
Appendix: Surnames and DNA

Notes
Bibliography
Sources of Illustrations
Acknowledgments
Index
Prologue

Today Celtic languages cling to precarious life on the northwest fringes of

Europe. [1] Delve into the pre-Roman past and we find Celtic spoken
across the continent. The heritage of the Celts turns up beneath the
trowels of archaeologists from Portugal to Romania, from Scotland to
Spain.
1 Celtic languages were widespread before most Celtic-speakers were enveloped by the Roman
empire. Centuries of Roman rule ensured that the Continental Celts switched to speaking Latin.
Celtic languages survived only in the British Isles. The Celtic language of Britain was taken to
Brittany by British settlers. Another British settlement in northwest Spain did not long retain its
Celtic tongue.
2 The Celtic love of curvilinear design is displayed in this bronze mirror from Desborough,
Northamptonshire (50 BC–AD 50). It is one of the finest examples of a type of Iron Age object that
was exclusively made in Britain. The complex pattern may have been laid out using a compass.

There are many books on the Celts. Their liquid, swirling art fills
lavishly illustrated volumes. Works of deep scholarship explore their
language and literature. For over a century archaeologists have been
triumphantly publishing a wealth of discoveries that can be linked to the
Celts. [2] Indeed the very enthusiasm in the last century for all things
Celtic fed a backlash in the 1990s. A few exasperated archaeologists
produced books assuring us that there was no such thing as a Celt – or
certainly not in the British Isles. Scholars of Celtic studies were unshaken
and even more productive. In the first decade of this century an ambitious
project by the Centre for Advanced Welsh and Celtic Studies at the
University of Wales culminated in the publication of a five-volume
historical encyclopedia of Celtic culture and an atlas for Celtic studies.
Why then is there a need for another book? The fast-moving field of
genetics has opened up new vistas on the past. Ancient DNA is replacing
argument over who the Celts were and where they came from. At the
same time some scholars are taking a fresh look at other kinds of
evidence. The traditional conviction that the Celts arose in Iron Age
Central Europe has been challenged. Historians meanwhile have been
picking apart some of the best-known narratives of that elusive period
from the time of St Patrick to that of Kenneth MacAlpin, and weaving their
threads together again in new patterns. If revisionism has gone too far in
places, it has still ignited fruitful debate. The aim here is to present a new
multidisciplinary synthesis, tracking the Celts from their distant origins to
their modern descendants through genetics, archaeology, history and
linguistics.

Deduced timeline for the prehistory of the Celts

Approximate date Archaeology Language

3400 BC Yamnaya culture on the European steppe Proto-Indo-European
3100–2800 BC Yamnaya movement up the Danube Alteuropäisch/Old
European
2200–1700 BC Late Bell Beaker Early Celtic, Italic and
Ligurian
1200–750 BC Bronze Age Hallstatt (Hallstatt A & B) Celtic
750–600 BC Early Iron Age Hallstatt C Celtic

Timeline for the historical Celts

Approximate date Archaeological Historical event Date

period
600-460 BC Final Hallstatt Inscriptions in Lepontic (Celtic) began
(Hallstatt D) c. 600 BC
Argantonios of Tartessos lived before 540 BC
Hecataeus of Miletus referred to Massalia
(Marseilles) as near Celtica c. 500 BC
460-260 BC Early La Tène Herodotus mentioned Keltoi at the head of
the Danube and in western Iberia c. 450 BC
Gauls ejected the Etruscans from the Po
Valley c. 400 BC
Celtic mercenaries in Greece 369-368 BC
Alexander the Great met a Celtic 335 BC
delegation
Celtic incursion into Thrace 298 BC
Gauls entered Asia Minor, where the
 Gauls entered Asia Minor, where the
Greeks called them Galatoi (Galatae in
Latin) c. 280 BC
Celtic attack on Delphi 279 BC
260-150 BC Middle La Tène Wars between Attalus I of Pergamum and 233-232 BC
the Galatians of Asia Minor
Roman victory over the Boii, Insubres and 225 BC
Gaesatae in Italy
Start of the Roman conquest of Iberia 218 BC
Final defeat of the Cisalpine Boii 191 BC
150-50 BC Final La Tène Founding of the Roman Provincia Gallia 125 BC
Narbonensis in southern Gaul
Caesar conquered the remainder of Gaul 58-51 BC
Caesar made two sorties into Britain 55-54 BC
AD 43-410 Romano-British The Roman conquest of Britain AD 43-84
End of Roman Britain AD 410
AD 410-800 Early medieval Death of St Columba AD 597

As Irish historian Eoin MacNeill sagely said in 1920, there is no Celtic

race, any more than there is a Germanic race or a Latin race, if by ‘race’
we mean some set of physical features that clearly distinguishes one
from another. Roman observer Tacitus was convinced that the Germani
all had ‘wild blue eyes, reddish hair and huge frames’.1 Finding that the
Caledonians of northern Britain fitted the same description, he supposed
that they were of Germanic stock. By contrast the swarthy faces and
curly hair of the Silurians of south Wales he attributed to Iberian
descent.2 Another Roman author describes the Gauls as very tall, with
white skin and blond hair, which is exactly the way other Classical
authors portrayed the Germani. As MacNeill pointed out, what would
most strike a Roman observer in northern Europe would be a higher
percentage of paler colouring than he saw in Italy. Seizing on what is
actually a matter of degree, stereotypes were created. We are still prone
to this today. So it needs to be said that what MacNeill surmised in 1920
we can now prove. He felt that all the present nations of Europe are a
mixture of the same ancestral components in varied proportions. He was
right. As we shall see, there are three main components to the modern
European gene pool. They came from ancient hunter-gatherers, early
farmers and a Copper Age people. The modern Irish have a mixture of all
three, as do the modern Germans and Italians. Any genetic differences
are far too subtle to talk in terms of a Celtic race.
MacNeill’s definition of a Celt was an ancient person known to have
spoken a Celtic language.3 That is the principle followed in this book,
though I see no reason to exclude modern Celtic-speakers. The terms
Insular Celts and Continental Celts are used to distinguish between those
who inhabited the British Isles and those who lived on the continent of
Europe.
What place does genetics have in this? We find correlations between
languages and DNA signatures. The reason is that children usually learn
their first language from their biological parents. Biology and language do
not always go hand in hand. A child could be adopted, or could have
parents of two different language origins and so could grow up speaking
a completely different language from that of one or both biological
parents. Then there are the seismic shudders through society that leave
whole populations speaking a different language. Even so, the link
between language and DNA occurs often enough for us to see patterns
in the data.
This book does not attempt to describe or explain every aspect of the
Celts. The primary focus is on the web of migration that over the ages
crisscrossed a continent and took to the sea. The intertwined strands of
that story show us how complex is the answer to what seem simple
questions of origins and identity.
Since the Celts are here defined as those speaking a Celtic language,
it is appropriate to open with their words, albeit in translation. So Chapter
1 displays the Celts of the British Isles in full voice in the early Middle
Ages. Then in Chapters 2 to 4 we work backwards in time to find the
deepest origins of the Celts and their ancestors, before turning forwards
again to trace them down to their modern-day descendants in the British
Isles and Brittany.
It is a story of sunlight and shadows, as indeed is all of human history.
None of our ancestors lived in a fairy tale, though they might have the wit
to invent one. Prising fact from fiction is one of the tasks of the historian.
CHAPTER ONE

The Voices of the Celts

Emer, daughter of Forgall the tricky, wife of Cú Chulainn, made

speech:—
‘I am the standard of women, in figure, in grace and in wisdom;
None mine equal in beauty, for I am a picture of graces.
Mien full noble and goodly, mine eye like a jewel that flasheth;
Figure, or grace, or beauty, or wisdom, or bounty, or
chasteness,
Joy of sense, or of loving, unto mine has never been likened.
Sighing for me is Ulster,– a nut of the heart I am clearly –
My spouse is the hound of Culann, and not a hound that is
feeble;
Blood from his spear is spurting, with life-blood his sword is
dripping;
Finely his body is fashioned, but his skin is gaping with gashes,
Wounds on his thigh there are many, but nobly his eye looks
westward;
Bright is the dome he supporteth and ever red are his eyes,
Red are the frames of his chariot, and red are also the cushions;
Fighting from ears of horses and over the breaths of men-folk,
Springing in air like a salmon when he springeth the spring of
the heroes,
Rarest of feats he performeth, the leap that is birdlike he
leapeth,
Bounding o’er pools of water, he performeth the feat of nine
men;
Battles of bloody battalions, the world’s proud armies he
heweth,
 Beating down kings in their fury, mowing the hosts of the
foemen.’1

This proud lady is easily the winner in the ‘Ulster women’s war of words’
at The Feast of Bricriu, an Irish tale written down in the Middle Ages. Her
status was high, for her husband Cú Chulainn was the great hero of the
Ulster Cycle of tales. Her contest is linguistic. The wives of other warriors
had proclaimed their own rank and beauty and the deeds of their
husbands. Emer proves her worth with a flow of eloquence to outdo
theirs.
In this one snippet of a tale, we see Celtic-speakers much as ancient
Greek and Roman writers portrayed them – valiant, boastful, fond of
feasting and lovers of language, beauty and wisdom. The Romans left
their most detailed commentary on the Gauls, as we shall see, but here
we have a matching picture of the Irish. No stereotype can truly mirror a
multitude of individuals. Yet the voices of Celts long gone can reveal the
values of the elite for whom songs were sung and pedigrees recited. The
Roman legions brought literacy to many a Celt, but at a price. Celtic-
speakers turned into Latin-speakers within the Roman empire. So the
Celts of Continental Europe, where Roman influence was strongest, left
us nothing that could be called literature in a Celtic tongue. There are
inscriptions, curse tablets and suchlike fragments, but no connected
narrative. We turn then to the poetry and prose of those Celts whose
language was not lost to Latin.

Heroic ideals
Ireland remained outside the Roman empire; the Roman province of
Britannia was on its northernmost fringes. Thus Celtic languages
survived in the British Isles. There we find not only tales in Gaelic, but
also heroic poetry written in Brittonic, once spoken over most of Britain
and the ancestor of Welsh. The collection of death-songs called Y
Gododdin sprang from northern Britain though it was preserved in Wales.
Here is the elegy for a lord of Dumbarton:

He rose early in the morning:—

When the centurions hasten in the mustering of the army
Following from one advanced position to another
 Following from one advanced position to another
At the front of the hundred men he was the first to kill.
As great was his craving for corpses
As for drinking mead or wine.
It was with utter hatred
That the lord of Dumbarton, the laughing fighter,
Used to kill the enemy.2

Once again we see admiration of courage. A lord was expected to lead

his men into battle, not plan tactics from the rear. We can picture this
‘laughing fighter’ charging exhilarated at the enemy. He was the type of
man who would become a leader, for he could inspire others to follow.
This was not a society of nation states, with taxation supporting standing
armies.
Through the lens of Y Gododdin we can glimpse a tribal structure
similar to that which the Romans had encountered among the Celts.
Indeed the tribal name Gododdin is the Middle Welsh version of Votadini,
recorded in the Roman period. The tribe lived between the Firth of Forth
and the River Wear. [see 87] Today their former territory is partitioned
between northeast England and southeast Scotland.3
One of the strongholds of the Gododdin lies beneath the present castle
on Edinburgh Rock.4 Under the name Eidyn it features in Y Gododdin.
Here a war-band feasted before the major battle celebrated in these
verses. A chief who lavished meat and mead on his followers could
expect feats of heroism in repayment. As one verse declares: ‘The trees
of battle were trampled – vengeance in payment for mead.’5 Among the
feasting warriors were men both of Gododdin and from further afield.
War-bands were often of mixed origin. The lord of Dumbarton had come
to Eidyn from the neighbouring Brittonic kingdom of Alt Clud, with its
stronghold at Dumbarton overlooking the River Clyde.
The lord of Eidyn, leader of the war-band, was the son of a man with
the English name Wolstan.6 The Angles of Deira, in what is now east
Yorkshire, were his enemies.7 This enmity can be explained by the power
struggles between the various bands of Angles and Saxons who had
poured into Britain after it lost the protection of the Roman empire. Some
Angles became neighbours of the Gododdin when they settled north of
Hadrian’s Wall, founding the kingdom of Bernicia. They then annexed the
fellow-Anglian kingdom of Deira to the south, but in AD 616 the exiled heir
of Deira successfully fought back, taking both kingdoms. The royal heirs
of Bernicia sought refuge with the Picts or Irish.8 So it would be no
surprise to find one of their retinue among the Gododdin, thirsting for
vengeance. The main battle commemorated in Y Gododdin is generally
dated about AD 600, but a date during the period of Deiran control of
Bernicia, AD 616–33, would explain why the enemy across the border
from the Gododdin are the Deirans in the earliest form of the elegies.9
Another of the fallen commemorated in these verses was Heini son of
Neithon, renowned for killing 100 gold-torced chieftains before he joined
the band at Eidyn.10 Here we have a man with a Celtic name
slaughtering men wearing the Celtic symbol of the noble warrior. That
tells us that warfare among Celts was not unusual either. War is a
common theme in early Irish literature. Here are just the first few stanzas
of a poem attributed to Laidcenn mac Bairceda:

It ill beseems me to forget the affairs of every famous king, the

careers of the kings of Tara, mustered tribes on the warpath.

A noble battle-hero, fair and tall was Moen, Labraid Longsech; a

cruel lion, a lover of praise, a mighty lover of battle.

A fair warrior was Ailill in battles against the frontiers of

Crothomun; Abratchaín shook the ranks of the field of
Ethonmun.

Dreaded master of Ireland was glorious Oengus Amlongaid. He

dwelt upon the slopes of Tara: with his own will alone he
conquered it.11

The style and archaic language of this poem places it among the
earliest surviving Irish verses, from the late 6th century or early 7th
century. The tradition of preserving knowledge orally in verse had come
into contact with Christian literacy. Poetry began to be written down.12
Celtic poetry was not all blood-soaked. Bards retained by a lord to sing
his praises might be limited in their official repertoire, but some poetry
could be composed simply for pleasure. This pair of quatrains was written
by a scribe in Old Irish in the margin of a Latin grammar that he was
copying c. 845. It captures the delight in writing in the open air, amid
foliage and birdsong.

A hedge of trees surrounds me,

A blackbird’s lay sings to me;
Above my lined booklet
The trilling birds chant to me.
In a grey mantle from the top of bushes
The cuckoo sings:
Verily—may the Lord shield me!—
Well do I write under the greenwood.13

The Book of Taliesin preserves some of the oldest poems in Brittonic.

Many of them are attributed to Taliesin, court poet to the 6th-century King
Urien of Rheged in what is now the north of England. One poem from the
collection, ‘The Fold of the Bards’, alludes in a round-about way to poetic
contests, like the ‘Ulster women’s war of words’ that opened this chapter.
This extract boasts of the poet’s skills:

I am a harmonious one; I am a clear singer.

I am steel; I am a druid.
I am an artificer; I am a scientific one.
I am a serpent; I am love; I will indulge in feasting.
I am not a confused bard drivelling.14

Druids
Caesar linked the memorizing of verses particularly to a privileged class
among the Celts, the druids, who acted as judges, priests and
teachers.15 How we would love to have an authentic account by a druid
of his activities. Instead we have a view of druids from Classical sources,
which may be biased. On the other hand we have modern fantasy woven
around a long-bearded figure with an aura of power.
The wickedly entertaining Terry Pratchett had a druid flying a megalith
through the clouds to patch up an astronomical stone circle which had
developed a hardware glitch.16 Pratchett managed thereby to satirize
targets ancient and modern. One old story was that the great wizard
Merlin had transported the stones of Stonehenge from Ireland, though on
ships rather than by magical levitation.17 Archaeologists may protest that
the shadowy Merlin (if he lived at all) dates from a time long after the
construction of Stonehenge, and so did druids. Yet nothing seems to
shake the popular perception of a link between druids and megaliths. [3]
The druids in early Irish literature are rather different. One is
persistently praised as good and marvellous in the Ulster Cycle of tales:

At that time a certain féinnid [a warrior living apart from his tribe]
came from the south of Ulster, performing féinnidecht [war-like
deeds] across Ireland, with a band of three times nine men: his
name was Cathbad, the wondrous druid. He had great
knowledge, and magic [druídecht], and manly strength; he was
of the Ulaid by birth, but had run off from them.18

3 Antiquary John Aubrey surveyed Stonehenge in the late 17th century and considered it the work
of natives of Britain, rather than the Romans. He knew from Classical texts of pre-Roman British
priests called druids and so attributed Stonehenge to them. This 18th-century engraving of a druid
in a grove by Stonehenge, holding a sickle and mistletoe, is in this antiquarian tradition.
 The tale goes on to relate how Cathbad took a princess of the Ulaid to
wife by force and was granted territory by her father, the king of Ulster. A
son was born to the princess, Conchobar mac Nessa, who became a
king of Ulster in his turn, while Cathbad became the chief druid at his
court – a position of the highest status. The warriors of Ulster fell silent
when King Conchobar rose to speak; yet the king waited for Cathbad to
open the proceedings.19 So Cathbad was far from venerable when we
first meet him as a fiery young warrior. His hands were stained with
blood. The Ulster Cycle is full of marvels, but Cathbad usually has no part
in them. He stands out in one way though. He is portrayed as having the
gift of prophecy.20
Druids as prophets appear in a 7th-century Life of Patrick, where they
foretell his coming as no glad tidings for them.21 Indeed the gradual rise
of Christianity in Ireland rendered druids redundant as priests and
teachers.22 We find Celtic Christian saints credited with the gift of
prophecy. Since the Church had taken over the religious functions of the
druids, prophecy would also logically fall within its sphere. The Christian
tradition enveloped it easily, with its roots in Old Testament prophets. The
far-famed St Columba or Colum Cille (died 597) was credited with many
prophetic revelations.23 ‘He was a sage, a prophet and a poet’ says an
elegy written shortly after his death.24
Other powers associated with druids, such as control over nature, also
appear in the early hagiographies of Irish saints. A fascinating passage in
the Life of the 6th-century saint Mochuda presents a magical contest
between a druid and the saint, in which the saint triumphed by causing an
apple tree to blossom instantly and then fruit.25 St Columba too contested
with druids, we are told. In his mission to convert the pagan Picts of
northern Britain, he faced particular opposition from Briochan, foster-
father and tutor to King Brude. On one occasion Briochan aimed to
prevent the saint from sailing away on Loch Ness by making the wind
unfavourable. Columba foiled him by sailing anyway, and the wind swung
round to support him.26
Though the powers of druids are presented as magical, prediction is
what we expect from modern scientists. The understanding of natural
forces enables a meteorologist to predict the weather, for example. The
knowledge gleaned by living through many a season, stored up and
passed down from druid to druid would have been invaluable.

Literature and archaeology

Of all the rich legacy of early Irish literature, the Táin Bó Cúailnge (Cattle
Raid of Cooley) stands out for its length and fame. It forms the core of
what is known either as the Ulster Cycle of tales, after the home of its
hero, or the Red Branch Cycle, from the name of the banqueting hall at
Emain Macha, the royal centre of Ulster. It tells of an heroic clash of
arms. Medb, queen of Connacht, and her husband Ailill take an army to
Ulster to steal a great brown bull, the Donn Cúailnge. The men of Ulster
being struck by a debilitating curse, it is left to the young Cú Chulainn to
defend the province. [4] He takes his stand at a ford, engaging one
Connachtman after another in single combat until reinforcements
arrive.27
Medieval Irish monks treated Cú Chulainn as an historical personage,
attempting to work out dates for him long before their own time.28 Modern
scholars generally see the story as pure fiction. Indeed, statistical
analysis of its network of social relationships reveals a pattern akin to that
of the superheroes of the Marvel comics.29 Could the Táin have any
basis in fact? Conflict between the peoples of Ulster and their southern
neighbours is suggested by the remains of protective earthworks, of
which the Black Pig’s Dyke is the best known. [see 67] This is not one
long frontier, but discontinuous stretches that probably protected points of
easy access for cattle raiders. These fortifications included a massive
timber palisade, probably made of oak tree trunks. Radiocarbon dating of
remains of the palisade near Scotshouse in Co. Monaghan could be no
more precise than somewhere between 500 BC and 25 BC.30
Fortunately, tree-ring dating can pinpoint felling years with some
precision. Close study of the Dorsey earthwork in South Armagh reveals
two successive lines of defence.31 The southern one can be dated to 95
BC and the northern one to 140 BC.32 So it is most likely to be part of the
same works as Black Pig’s Dyke. Navan Fort in Armagh can be identified
as Emain Macha in the Ulster Cycle. Its occupants around 100 BC felt the
need for a bank and ditch around the top of the hill. Within, a large
circular building was erected. The massive central post has been dated
by tree rings to 95 BC.33 So there was a burst of defensive activity in the
year 95 BC. This is not to say that the events of the Táin Bó Cúailnge,
complete with their magic and marvellous cast of characters, are fully
historical. The difference between the surviving versions of the tale show
that it was elaborated over centuries.34 While enthusiasts see the Táin
Bó Cúailnge as a window on the Iron Age,35 sceptics feel that the lifestyle
it portrays would fit as comfortably just before the arrival of Christianity in
Ireland in the 5th century AD.36
Tree-ring dating has sprung another surprise. Here is a snippet of a
tale thought to be entirely mythological:

These are the four things

which Eochaid Airem chose
from the many manly-seeming companies,
with abundant shields and swords:
A causeway across the bog of the Lámraige,
a wood across Bréifne, without difficulty,
the fair removal of the stones of great Mide,
and rushes across Tethna.37

Eochaid Airem is described as king of Tara. The story itself is set in a

supernatural world. It forms part of the Mythological Cycle of Irish tales
about pre-Christian kings. So the appearance of an Eochaidh Aireamh in
the Annals of the Four Masters as a king of Ireland reigning between
Anno Mundi 5070 (c. 142 BC) and Anno Mundi 5084 (c. 116 BC) was
dismissed as non-historical.
4 Cú Chulainn, the hero of the Ulster Cycle of tales, is pictured riding into battle in his chariot, with
his charioteer beside him.

True annals are contemporary with the events they record. Monastic
annals might begin as jotted notes in the table used to calculate the date
of Easter, as reminders of memorable years, such as that in which a king
died. Transferred later to separate manuscripts, they could be continued
by one scribe after another. The Irish annals provide a contemporary
record of Irish history for over a millennium, from the middle of the 6th to
the early 17th centuries AD. Annalists could make mistakes, but the fact
that they were recording events as they happened makes annals one of
the most reliable sources for historians. However, the Annals of the Four
Masters is a compilation made in the 17th century. It aimed to gather
together all the Irish annals surviving at that time. Since it includes annals
now otherwise lost, it is an immensely valuable source.38 Yet its
prehistoric section is not based on any annal. In common with many
another chronicle composed retrospectively, it attempted to date events
from miscellaneous sources.
No wonder that Eochaid Airem’s causeway was seen as imaginary –
until, that is, an impressive timber causeway was uncovered at Corlea.
[5] It was built in 148 BC across the boglands of Longford, close to the
River Shannon.39 Richard Warner, then archaeologist at the Ulster
Museum in Belfast, argued that the name Lámraige in the tale is
preserved in the townland name Laragh, close to Corlea. So the Corlea
trackway could be identified as the work of Eochaid Airem. In fact Warner
defended the whole prehistoric section of the Annals of the Four Masters
by attempting to fit reported events to tree-ring dating.40 This met with a
shower of cold water from J. P. Mallory, who was teaching archaeology
at Queen’s University Belfast. How plausible is the concept of oral
memories passed down over thousands of years? The Corlea trackway,
being much closer in time to literacy, could be the one diamond in a heap
of broken glass.41
5 The massive Iron Age causeway at Corlea, Kenagh, Co. Longford, Ireland, crosses boglands
close to the River Shannon. It was built from split planks laid on top of raised rails and was solid
enough to carry wheeled traffic.

Pseudo-history
The tradition of a learned class who memorized verse has encouraged
the optimistic idea that authentic history could have been passed down
orally for thousands of years, finally to emerge in the work of medieval
historians. Yet it is rare for such histories to make any reference to
drawing on oral sources. A common theme instead is a frustrated search
for native written records. It begins in the Dark Ages with Gildas
complaining that if there ever were any British records of Britannia as a
Roman province, they must have perished in the towns burnt by the
Saxon enemy or accompanied those Britons who fled to distant lands.42
Gildas was primarily a religious writer, whose sermonizing Ruin of Britain
was preserved among Christian literature. So it survived as a precious
resource for historians from a period of British history otherwise almost
mute.
In the 8th or early 9th century the first attempt at a history of Britain, the
Historia Brittonum, was composed. In a prologue in one manuscript copy
of it, the author, named as Nennius, disciple of St Elbotus (presumed to
be Elfod, bishop of Bangor), blamed the dull British for casting away
knowledge of their past. He tells us that he had heaped together all he
could find, ‘partly from traditions of our ancestors, partly from writings and
monuments of the ancient inhabitants of Britain, partly from the annals of
the Romans, and the chronicles of the sacred fathers’.43 The result was
mainly pseudo-history. One modern authority deems the prologue a
forgery and the authorship of the work therefore unknown. Yet the
prologue’s strictures on the failure to record the past find echoes
elsewhere.
Here is an exasperated Irish would-be historian, writing in the late 9th
or early 10th century:

The foolish Irish race, forgetful of its history, boasts of incredible

or completely fabulous deeds, since it has been careless about
committing to writing any of its achievements.44
 Geoffrey of Monmouth (died 1154/5) was similarly frustrated. He was
probably of Breton stock. William the Conqueror had granted Monmouth
Castle to a Breton and it descended in his family until after Geoffrey’s
time. So Geoffrey’s parents may have been part of the entourage of the
Breton lords of Monmouth.45 The Bretons and Welsh spoke similar Celtic
languages. No wonder Geoffrey of Monmouth wallowed in nostalgia for a
golden Celtic past. As a cleric in Oxford, he searched keenly for any clue
to that past. According to him, he could find little until his friend Walter,
archdeacon of Oxford, gave him a ‘very ancient book written in the British
language’. Geoffrey tells us that he set himself to translate it into Latin,46
claiming that the result was his History of the Kings of Britain, completed
about 1138.47 Geoffrey traced the royal line from one Brutus of Troy,
whom he imagined to be the founder of Britain, through a host of
supposed pre-Roman sovereigns to three genuine 7th-century kings of
Gwynedd starting with Cadfan, whose gravestone is in Llangadwaladr
church on Anglesey (see also p. 178). [6] Geoffrey devoted most space
to a detailed and loving treatment of Arthur, that shadowy symbol of
resistance by the Britons to the Anglo-Saxons, which has ensured the
eternal popularity of his book. The phenomenal success of the Arthur
legend is fascinating. It has been endlessly adapted over the centuries to
appeal to different audiences. Geoffrey converted a Dark Age battle-
leader into an imperial Arthur who could rival the creators of Continental
empires, thus giving the British as proud a history as that of the Greeks
and Romans. Some scholars of Geoffrey’s time entertained their readers
at his expense,48 but such critical voices were drowned in the wave of
delight.
Alas, the History of the Kings of Britain was one long exercise in
fantasy writing for the greater glory of the nation. Far from being the
translation of a single book, it draws on many sources, both Classical and
Welsh, changing and elaborating to suit Geoffrey’s purpose.49 His key
British source may have been a manuscript similar to one now in the
British Library. [7] This contains the Historia Brittonum, from which he
took the pseudo-history of Brutus and a list of Arthur’s battles, the
Annales Cambriae (Welsh Chronicles), which mention Arthur, and Welsh
royal genealogies.50
6 The gravestone of King Cadfan (d. c. 625) built into the north wall of the nave in Llangadwaladr
church, Anglesey. The inscription reads Catamanus rex sapientisimus opinatisimus omnium
regum (‘King Cadfan, most wise and renowned of all kings’).

7 The Historia Brittonum was the first attempt at a history of Britain, written in Latin. In this copy
dated 1105 in the British Library, the large capital ‘B’ halfway down the page begins a section on
the origins of the British and Irish. In translation, it begins: ‘The island of Britain is so called from
one Brutus, a Roman consul.’

Much of Geoffrey’s work was simple supposition from place-names.

For example, he imagined that London must have been founded by a
King Lud and Leicester by a King Leir. Both cities were actually of Roman
foundation.51 Geoffrey was following in a long tradition of creating origin
stories from names. Places were indeed sometimes named after the
person who first settled there. So it was tempting to imagine that from
every place-name, country name or tribal name an ancestor could be
conjured up. Then storytellers wove legends about him.
Isidore of Seville (d. 636), who wrote an ‘encyclopedia’ of information
gathered from Classical sources, wryly reported that ‘Some suspect that
the Britons were so named because they are brutes [brutus in Latin]’.52
The Historia Brittonum deftly turned insult into boast, asserting that
Britain derived its name from one Brutus, a Roman consul. Brutus is
represented as the grandson of the Trojan prince Aeneas. There could
be no clearer proof that this tale drew on Classical sources rather than
local folk memory. The Aeneid was the Roman answer to the Greek Iliad,
that stirring epic poem peopled with gods and heroes, and telling of the
10-year siege of Troy. Virgil created his own epic around Aeneas, a
character in the Iliad who survives the fall of the city and was therefore
available for further poetic adventures. Virgil has his hero buffeted around
the Mediterranean until he found shelter at last in Latium, home of the
Latins. There Aeneas became king by a combination of conquest and
marriage politics. Thus Virgil provided a glorious origin for the Roman
people. Tagging a Brutus on to the genealogy of Aeneas created an
equal status for the British with much less effort.53
Since Christianity and literacy were so closely linked, it was generally
the scholarly religious who began to shape origin stories for the Celts of
Britain and Ireland in the Middle Ages. It was natural for them to turn to
the Bible for ancient history. Genesis narrates the story of a deluge which
only one family survived – that of Noah, with his three sons Shem, Ham
and Japheth and their wives. Their descendants supposedly populated
the earth.54 The biblical offspring of Ham covered an impressive swathe
of lands from Mesopotamia through Palestine to northeast Africa. The
biblical Shem is presented as the forefather of the Assyrians, Elamites
and Hebrews. Japheth’s brood seem to be those Indo-Europeans who
were known to the Hebrews. They lived in a semicircle around the Fertile
Crescent: to the east were the Medes, to the north on the steppe ranged
the Scythians and Cimmerians, while the Hellenes lay to the northwest in
Greece, Cyprus and western Asia Minor.55
The Romano-Jewish historian Josephus (AD 37–c. 100) provided some
more geography:

Japhet [sic], the son of Noah, had seven sons: they inhabited
so, that, beginning at the mountains Taurus and Amanus, they
proceeded along Asia, as far as the River Tanais [Don], and
along Europe to Cadiz; and settling themselves on the lands
which they light upon, which none had inhabited before, they
called the nations by their own names.56

As newly Christian nations began to search for their origins,

increasingly complex genealogies from Noah were created.57 The
Historia Brittonum provides a descent of Brutus from Japheth via his son
Javan, which is the word used throughout the Old Testament for the
Ionian Greeks of western Asia Minor.58 This no doubt seemed logical to
those trying to forge a link to the people of Troy.
Another pseudo-history painted on a biblical backdrop has fascinated
generations. Lebor Gabála Érenn (The Book of the Taking of Ireland),
compiled in the late 11th century, tells a stirring story of invaders battling
for Ireland. It opens with a synopsis of Genesis, starting with the creation
of heaven and earth and proceeding inevitably to the offspring of
Japheth. This time two sons of Japheth are mentioned: Magog, whose
‘progeny are the peoples who came to Ireland before the Gaedil’, and
Gomer, ‘of him are the Gaedil and the people of Scythia’.59 Gomer is the
name used in the Bible for the Cimmerians, occupants of the steppe
north of the Black Sea, later known as Scythia.60 Isidore thought Gomer
was the ancestor of the Gauls, possibly because no other son had a
name beginning with G.61 Similar thinking may account for the choice of
Gomer as ancestor of the Goidel, a name that the Irish used for
themselves no earlier than the 7th century AD, which helps to date the
Lebor Gabála Érenn. The name was adapted from the Brittonic name for
the Irish.62
Magog is mentioned in Ezekiel (38–39) as the land of a powerful
potential enemy:

Gog of the land of Magog.… You will come from your place in
the far north, you and many nations with you, all of them riding
on horses, a great horde, a mighty army.

The description fits the Scythian horsemen who took over the
European steppe from the Cimmerians. Their incursions south of the
Caucasus were the bane of established states of the Near East.
Josephus made the logical identification: ‘Magog founded those … who
are by the Greeks called Scythians.’63 So Scythia was seen as the
starting point for both the Gaedil and their predecessors in Ireland.

Genetics: the first clues

Genetically the Irish do not cluster close to Iberians, despite the claims in some origin myths
(p. 28). Instead they overlap with their nearest neighbours, the British.64 How do we know?
The studies that came to this conclusion compared samples of DNA from living people. But
what exactly did they look at? In the nucleus of every human cell are the 23 pairs of
chromosomes that hold the code for the creation of a human being. One of each pair is
inherited from each parent; 22 pairs of your chromosomes are gender neutral. The other
pair dictates whether you are born male or female: two X chromosomes and you are a girl,
but an X and a Y for a boy. Together these 23 pairs of chromosomes are known as the
genome. Each chromosome is made up of two strands of DNA that coil around each other
in the famous spiral staircase or double helix.
The code itself is composed of just four nucleobases, written as A, T, C and G. Most of
our DNA is shared with all other human beings, but there are locations where the genetic
code varies between individuals, for example I might have a T where you have a C. Such a
location is known as a Single Nucleotide Polymorphism (SNP, pronounced ‘snip’). The
studies which found that the Irish cluster with the British compared many SNPs across the
genome. It may make matters clearer to focus on one particular chromosome.
Only males carry a Y-chromosome. So mutations on this chromosome enable us to track
descent from father to son. Picture an unbroken chain of life from your earliest male
ancestor through countless generations to your paternal grandfather and your father down
to you, if you happen to be male. Your Y-DNA should be exactly the same as your father’s.
But sometimes there are faults in replication. You could see it as a typing error in the chains
of letters along the DNA. Such errors, often called mutations, can tell us a lot. The pattern of
mutations in your Y-DNA places you in a haplogroup.
 Western Europe is saturated with a particular Y-DNA signature, which has been labelled
R1b1a2a1a2. [see 23] This name fits the haplogroup on to a Y-DNA ‘family tree’ or
phylogeny. [8] From R descends R1, from R1b descends R1b1 and so on. Since these
‘relative’ names change as new mutations are discovered and the tree changes, it is
common to identify a haplogroup also by the mutation which defines it, which is unchanging.
So you may see R1b1a2a1a2 written as R1b1a2a1a2 (P312) or R1b-P312 for short.
The subclade (subgroup) R1b1a2a1a2c (L21) is overwhelmingly common in Ireland and
north and west Britain. [9] It is found at its densest concentration in those parts of the British
Isles where the Celtic languages lasted longest. In France it is strongest in Brittany, named
after the Britons who settled there. Y-DNA from a burial at Hinxton, Cambridgeshire, dating
to the last decades before the Roman conquest belongs to R1b-L21.65 A link is clear
between R1b-L21 and the Insular Celts.

8 A section of the phylogeny of Y-DNA showing the haplogroups common in Europe, which
descend from the ancient CT, mainly via F. The root of the tree (not shown here) is even
more ancient, going back to ancestral haplogroup A in Africa. R1b, common in western
Europe, is a relatively young haplogroup.
 9 The distribution of Y-DNA R1b-L21 suggests that it travelled down the Rhine and into the
British Isles, where it is now densest in the regions least affected by later arrivals. The high
level in Brittany may reflect the Dark Age migration of Britons, after whom Brittany was
named.

Outside the British Isles, R1b-L21 of British and Irish types can be explained by migration
from the Isles, but this does not account for all the R1b-L21 in France and the Low
Countries. It is a working theory that R1b-L21 entered the British Isles from this direction.
Ancient DNA is needed to confirm it. The light dusting of L21 in Galicia may be a relict of
those Britons who settled there in the Post-Roman period, or those who may have moved
there in the late Bronze Age. Some subclades of R1b-L21 can be associated with specific
surnames or families (see Appendix).

As it chances, the European steppe is now regarded as the homeland

of the Indo-Europeans, as we shall see in Chapter 4. But in the Lebor
Gabála Érenn the idea perhaps built on the geography of Josephus for
the sons of Japheth moving westwards into Europe from the River Don to
Spain.
A related idea appears as early as AD 731, when Bede began his
Ecclesiastical History of the English People. He records a tale that the
Picts came from Scythia.66 The Historia Brittonum culled from Irish
scholars an Irish origin story that starts with a Scythian nobleman and his
kin exiled in Egypt at the time that the Israelites escaped Egyptian
captivity. After many years wandering in Africa these supposed exiles
landed in Spain. There they stayed and multiplied for a thousand years
before moving to Ireland. Thus a parallel was created with the wandering
Israelites seeking the Promised Land. Waves of arrivals from Spain
culminated in three sons of a Spanish soldier (miles hispaniae) reaching
Ireland with thirty ships.67
The Lebor Gabála Érenn elaborates on this story. The Scythian
nobleman becomes Fénius Farsaid, who brought the Irish language from
the Tower of Babel. His character is based on one Fenech, who appears
as the leader of the descendants of Japheth at Babel in an obscure
Hebrew text. Fenech was evidently seized upon because of the fortuitous
similarity between his name and the word Féni, an Old Irish term for the
Irish people. The Latin miles hispaniae (soldier of Spain) was converted
into a name, Míl Espáine, and a mass of fake genealogy was grafted on
to the scheme. His sons Éber and Erimón divide the kingship of Ireland
between them. Both these names are also artificial, being derived from
the name of Ireland in Latin (Hibernia) and Gaelic (Ériu). Éber, presented
as the founding father of the Eóganacht, takes the southern half, while
Erimón takes the north. This division supplants a more ancient concept of
Ireland being divided into five parts. So it was probably cooked up in the
8th century AD to give a respectably ancient ancestry to the newly
dominant dynasties of the Uí Néill and Eóganacht. Essentially it is a
learned fiction, but it incorporates some genuine Celtic names, such as
Fir Domnann.68 For them we have some independent evidence (see p.
156).
Like Geoffrey of Monmouth’s fanciful work, the Lebor Gabála Érenn
was accepted for centuries as an accurate history. Modern scholars are
less credulous. R. A. Stewart Macalister, who translated the Lebor
Gabála Érenn into English in the mid-20th century, declared that: ‘There
is not a single element of genuine historical detail, in the strict sense of
the word, anywhere in the whole compilation.’69
Linguistic and genetic evidence conclusively rule out a Scythian
ancestry for the Celts, as we shall see (p. 78). Though some probable
input from Iberia into the Celts of Ireland or Britain will emerge in Chapter
5, modern DNA suggests that this was overlain by a stronger migration
pulse down the Rhine (see Genetics box, pp. 26–27). So in the next
chapter we cross the choppy English Channel to what was once Gaul.
Overview
• Ireland and Wales preserve the earliest literature in Celtic
languages.

• Key features of the Insular Celts that emerge from this literature are:
• Heroic values and leadership by warrior chiefs.

• Love of nature, feasting, language, beauty and wisdom.

• High status accorded to druids as prophets, sages, teachers and

priests.

• The Táin Bó Cúailnge is fiction, but its core conflict between the
peoples of Ulster and their southern neighbours fits archaeological
evidence.

• The medieval Irish, Welsh and Picts had no true knowledge of their
origins, which were too far in the past for recollection. So origin
stories were developed by early Christian scholars, using Genesis
and Classical sources.

• The genetic evidence links Ireland to Britain more than Iberia, but
complications will emerge in Chapter 5. In Chapter 2 we follow the
more obvious trail to Gaul.
CHAPTER TWO

The Gauls and Celtic

The Gauls are very tall with white skin and blond hair, not only
blond by nature but more so by the artificial means they use to
lighten their hair. For they continually wash their hair in a lime
solution, combing it back from the forehead to the back of the
neck … this treatment makes the hair thick like a horse’s mane.
Some shave their beards, while others allow a short growth, but
nobles shave their cheeks and allow the moustache grow until it
covers the mouth.… In both journeys and battles the Gauls use
two-horse chariots which carry both the warrior and the
charioteer. When they encounter cavalry in battle, they first hurl
their spears then step down from the chariot to fight with
swords. Some of them think so little of death that they fight
wearing only a loincloth, without armour of any kind.… Their
trumpets … are of a peculiar and barbarian kind which produce
a harsh reverberating sound suitable to the confusion of battle.1

Thus wrote Diodorus Siculus between about 60 and 30 BC, but he was
relying on older sources. When Caesar conquered Gaul in 58–57 BC, he
did not encounter chariots in warfare. They had dropped out of use in
favour of cavalry. But when Gauls battled the Romans in northern Italy in
225 BC, they did indeed use chariots in battle and some fought naked.2
[10]
We tend to visualize Gaul as equivalent to modern-day France. We talk
of Gallic cuisine or expressive Gallic gestures, when we mean French.
The Gaul of Caesar’s day did include all of mainland France, but it also
stretched eastwards to the Rhine, encompassing what is now Belgium
and Luxembourg, the Netherlands south of the Rhine, and those parts of
Germany and Switzerland west of the Rhine. From the Roman
perspective, this region was Transalpine Gaul (Gaul on the other side of
the Alps). Since Gauls had spread over the Alps into northern Italy by
then, there was also a region known to the Romans as Cisalpine Gaul.
[see 1]

10 A denarius minted by the Roman moneyer Lucius Hostilius Saserna in 48 BC. The bust of a
Gallic warrior appears on the obverse, his hair thickened with lime and combed back. The reverse
depicts a naked Gallic warrior standing in his chariot, holding a spear and small shield, while a
crouched charioteer, perched on the pole, drives the horses with a whip.

Caesar famously declared that Gaul was divided into three parts,
inhabited by the Belgae, the Aquitani and ‘a people who call themselves
Celts, though we call them Gauls’. Each, he said, had a different
language.3 In fact the name Belgae has a Celtic etymology and there
were Celtic place-names in their territory in Roman times,4 so their
linguistic differences from the Gauls cannot have been dramatic. By
contrast the Aquitani of southwestern Gaul appear to have spoken a
language ancestral to modern-day Basque. Yet another language, not
noted by Caesar, was spoken by the Ligurians along the southern coast
of Gaul. Enough evidence survives of the language of the Gauls
themselves for linguists to recognize it as belonging in the same family as
Breton, Welsh and Gaelic. Centuries earlier, seagoing Greeks had
encountered Keltoi on the Mediterranean coast at Narbo (Narbonne) and
near Massalia (Marseilles). The earliest surviving record of these Keltoi
comes from Hecataeus of Miletus (c. 500 BC).5
Ancient Greek knowledge of Central Europe was much vaguer.
Herodotus (c. 484–425 BC) had heard of Celts beyond the Pillars of
Hercules (Strait of Gibraltar) and he knew, probably from Hecataeus, that
the River Ister (Danube) rose in the land of the Celts, but his geography
seems absurd to modern readers. He thought that the river rose at the
city of Pyrene, which he envisaged somewhere in Iberia.6 Aristotle (384–
322 BC) made it clear that Pyrene in this passage was intended to be the
Pyrenees.7 Before we mock such ignorance, we should recall that
ancient Greek geographers were pioneers, trying to make sense of
information coming from sailors and traders. There were huge gaps in
their knowledge. That could lead to a mental map of one region being
stitched directly to that of another, without realizing that there was a lot of
territory in between. The Danube actually rises from two sources in the
Black Forest mountains of southwestern Germany, and is swelled by
tributaries from the Alps. The Alps are not mentioned by Greek authors
until the 3rd century BC.8 Herodotus, however, does list the tributaries of
the Danube from east to west, the last one of which he names as the
Alpis, rising north of the land of the Ombricians (Umbrians of Italy) and
flowing north to join the Danube.9 Here it seems we have one of the
Alpine tributaries. So we can ignore the red herring of Pyrene and accept
that Celts were living north of the Alps around the head of the Danube.
11 The Iron Age phases of the Hallstatt culture (Hallstatt C and D) encompassed the upper
reaches of the rivers Danube and Rhine. Early La Tène culture centres sprang up on the northern
fringes of the Hallstatt elites.

In the 19th century scholars identified two successive cultures at the

right place and time to correspond to these historical Celts. Johann
Georg Ramsauer was in charge of the salt mines at Hallstatt in the
Austrian Alps in 1846 when he discovered a cemetery nearby. [see 46]
Instead of simply plundering it, as antiquity-hunters of his day were wont
to do, he carefully excavated and recorded his findings. The people
buried there had been salt miners too. Odd items that they left in the
mines – clothing and wooden tools – were wonderfully preserved by the
salt. So this early foray into archaeological methodology remains one of
the most remarkable sites ever linked to the Celts. We now know that it
belongs mainly to the 7th and 6th centuries BC.10 When similar artifacts
were found elsewhere, they were naturally identified as in the style of
Hallstatt, so the site gave its name to a culture eventually found to be
widespread over Central Europe. [11] It began in the Bronze Age,
centuries before the salt mine at Hallstatt, but its later phases (c. 750 to
c. 475/450 BC) fall into the Iron Age.11
The same process decreed that a site at La Tène, on Lake Neuchâtel
in Switzerland, should be forever associated with the next phase of the
Central European Iron Age (c. 460 BC to mid-1st century BC). In 1857
local collector Hansli Kopp noticed some timber piles driven into the mud
of the lakeside. Groping between them, he found about 40 weapons. It
was a sign of the riches to come. Over the following decades, this
extraordinary site yielded over 3,000 artifacts. The flowing forms of the
decorative metalwork attracted most attention, but the oxygen-free lake
mud also preserved wooden objects, including a complete chariot
wheel.12 [12, 13] The chariot was to prove a key distinction between the
earlier Hallstatt culture and La Tène.

12 Stylized animals enliven an iron scabbard (3rd century BC) from La Tène in Switzerland. The
style is confidently fluid and playful.
13 This spoked wheel with iron tyre was discovered at La Tène. It was preserved by the oxygen-
free mud of this lake site.

By the time of the 1871 International Congress of Prehistoric

Anthropology and Archaeology in Bologna, objects in the same style as
found at La Tène had been dug up in France and northern Italy. The
Italian finds were crucial. The Roman historian Livy described the Gauls
crossing the Alps in waves to settle in the Po Valley, driving out those
Etruscans who lived there.13 So the cemetery of an intrusive people in
the ruins of the Etruscan town of Marzabotto could be firmly identified as
Celtic.14 The La Tène culture could thus be seen as the material
manifestation of the Celts. It could be tied not only to historical events,
but also to descriptions of the Celts, such as that which opens this
chapter, mentioning the Celtic chariot.

Chariots
The essential feature of the chariot was its speed. The first wheeled
vehicles were heavy wagons with solid wheels. They were tough
structures, ideal for transporting farm produce or use as a mobile home.
The invention of the spoked wheel reduced the weight. Given a stripped
down superstructure without a driver’s seat, a single axle and the power
of fast horses, these wheels could flash along.
The chariot was the sports car of its era for its speed and
manoeuvrability. In the Bronze Age of the Near East, Egypt and Greece it
was the vehicle favoured by royalty and a warrior elite. Since these
civilizations left records behind, we know that maintaining a chariot force
was hugely expensive, requiring specialist chariot warriors, charioteers,
horse-trainers, grooms and chariot-makers. Chariots were used in war
and peace. A mobile archery platform could be used for hunting. More
ornate versions featured in ceremonies. It was a versatile vehicle.
Among the Celts the chariot appears in high-status burials, mainly of
warriors with their weapons.15 Since wood usually decays in the soil, the
remains of chariots are more often recognized by their metal fittings.
Celtic chariot burials were discovered in the Marne department of France
in the 1870s, such as that of La Gorge Meillet.16 [14] Marne has
subsequently proved to be rich in chariot burials of the early La Tène
period.17 [see 55]
The Continental Celts left us no descriptions of chariots. So we are
faced with a paradox. Some of the best Celtic-language sources for
chariots were written in Ireland, where no complete chariot has been
found, though wooden wheels have survived at the remarkable wetland
site at Edercloon, Co. Longford, starting with a Late Bronze Age solid
wheel and continuing to Iron Age and medieval spoke wheels.18
Furthermore the earliest surviving version of the Táin Bó Cúailnge may
have been first committed to writing in the 8th century AD.19 So there is a
gap of over a millennium between the earliest Celtic chariot burials and
this literary epic. That gap in time narrows if we consider the last
evidence for chariots, rather than the first. Though Caesar mentions no
chariots in Gaul, his forces came up against massed ranks of them in
Britain in 54 BC.20 Chariots were still in use by Britons resisting the
Romans in AD 60–61. Who can forget the image of an enraged Boudica
rousing up rebellion ‘in a chariot with her daughters in front of her’.21
Even in post-Roman Britain, one of the warriors celebrated in Y
Gododdin had a war-chariot.22 In Ireland a king escaped in a chariot from
a battle in AD 563.23
14 Chariot burial of La Gorge Meillet, in the French department of Marne, excavated in 1876. The
body of a warrior had been laid on the floor of his chariot, and slots cut in the ground for the
wheels to fit into. The wooden wheels had decayed, but the iron tyres and bronze axle bands and
hub caps remained. The rectangular cut of the burial is typical of those found in Marne. The grave
was richly equipped with iron sword, javelin and knife, and a bronze helmet.

Sadly, the texts of the Táin Bó Cúailnge that survive are not the original
manuscript, but versions written down in the 12th century, garnished in
glorious detail, wherein scholars detect the hand of medieval scholarship.
A wonderful description of the war-chariot of Cú Chulainn, laden with
precious metals, comes into the category of such later embroidery.24 By
this time Irish scholars were well versed in Latin literature, which included
versions of Greek epics. Cú Chulainn was dubbed ‘the Irish Achilles’ by
Celticist Alfred Nutt at the beginning of the 20th century, for his
correspondences to Homer’s hero.25 The chariot of Cú Chulainn was
drawn by two swift horses, one grey and one black, who are given a
supernatural origin.26 In the Iliad, Achilles has yoked to his chariot the
‘fleet horses, Xanthus and Balius, that flew swift as the winds, horses that
the Harpy Podarge conceived to the West Wind, as she grazed on the
meadow beside the stream of Oceanus’.27
Xanthus means ‘yellow’ and Balius ‘piebald’, but they can be depicted
as white and black, for example on a Greek hydra of the 6th century BC.28
Isidore of Seville refers to one black horse and one white being yoked
together in the two-horse chariot.29 So is the Irish chariot merely a literary
device?
Chariots are carved on six high crosses in Ireland. That might seem
evidence enough that they existed. However, these scenes seem to
depict a story from the Ulster Cycle, ‘The Phantom Chariot of Cú
Chulainn’. In this tale the hero reappears 450 years after his death to
convince Loegaire, king of Tara, to convert to Christianity.30 Proof that
the Irish chariot was no phantom comes from the Irish law codes. They
refer to a vehicle called a carpat. This is the same word translated as
‘chariot’ in early Irish literature. We can imagine a more workaday vehicle
than that driven by the hero of the Táin.31
Another lively debate has surrounded chariot burials found in Britain.
Doubts have been expressed about any connection with the Continental
Celts, yet a tribal link leaps to the eye. Since the Celts were largely
illiterate in pre-Roman days, it was ancient Greek and Roman authors
who first recorded tribal names. The Geography of Claudius Ptolemy,
written in Greek c. AD 150, provides the framework of our knowledge. It is
a shaky scaffolding by comparison with a modern atlas, but it was
revolutionary in its day. Crucially, Ptolemy supplied geographical co-
ordinates to locate towns. This was the start of scientific mapping.32
Ptolemy located the Parisi in two areas, both of which have chariot
burials. The Continental tribe, called the Parisii by Caesar, is easily
identified by the town that Ptolemy assigned to them, Lutetia Parisiorum
– present-day Paris. The British tribe had a settlement called Petuaria,
usually identified as Brough-on-Humber in Yorkshire.33
At the end of the last century, two Celtic chariot tombs dating to 300 BC
were uncovered during runway construction at Charles de Gaulle Airport
at Roissy, on the northern outskirts of Paris. In layout both were similar to
the chariot burials which begin earlier in the Marne region. [see 14] They
were rectangular in plan, with the deceased laid flat on the platform of
their chariots. Slots were cut in the ground to accommodate the wheels.
One burial was that of a warrior with weapons.
The other was exceptional. [15] Here had been laid to rest a man
without weapons. His chariot was sturdier than that of the warrior and
ornamented with decorative bronzes of rare quality. So it was built for
status rather than speed. A strange collection of objects lay scattered
between the wheels. They were probably amulets, which had been kept
in a pouch around the man’s neck. Could this be the tomb of a druid?34
The high status of druids in the Irish tradition (see p. 16) is confirmed by
Classical writers. Thus Diodorus Siculus:

15 The most remarkable object from an unusual chariot burial at Roissy, on the outskirts of Paris,
which may be the last resting place of a druid. This perforated bronze, a swirl suggesting fantastic
animals with outstretched wings, was created by the lost wax method. Surviving fibres show that it
formed the cover of a wooden vessel.

The Gauls have highly honoured philosophers and theologians

called druids.… They do not sacrifice or ask favours from the
gods without a druid present, as they believe sacrifice should be
made only by those supposedly skilled in divine communication.
… The Gauls obey with great care these druids.35

On the other side of the English Channel we find the Arras culture of
the Yorkshire Wolds, in the region where Ptolemy located the Parisi. [see
55] This culture combines La Tène material with chariot burials. Where
we find not only artifacts but also burial rites transplanted from one place
to another, it is reasonable to suspect migration. People tend to prefer
the burial practice familiar to them. Yet these burials are distinctly
different from those at Roissy. Typically, the Arras vehicles were
dismantled and the deceased buried in a crouched position. So it has
been argued that there is no connection with the Parisii of Gaul.36 Six of
these typical burials were certainly of people who lived locally all their
lives. The clues lie in the isotopes in their bones and teeth, which are
characteristic of those living on a chalk soil like that of the Wolds.37 So
they were not a group of migrants from Gaul. Who their ancestors were is
another matter.
The burials most typical of the Arras culture can be dated to around
200 BC.38 Earlier chariot burials have now been discovered in Britain,
causing considerable excitement. One dating to c. 355 BC was found at
Ferry Fryston in Yorkshire.39 This chariot was buried intact with slots dug
to accommodate the wheels, like those of the Marne region and Roissy.
Ferry Fryston seems connected to the Arras culture – perhaps the
earliest manifestation of it. The occupant of the chariot came from a
region of ancient granites.40 That might fit Scotland, in view of the
discovery of an even earlier chariot burial at Newbridge, near Edinburgh.
[see 55] This is dated to the 5th century BC – placing it within the earliest
La Tène period. Here, if anywhere, we might hope for a migrant from
Gaul. Unfortunately all traces of a human body had completely
disappeared. So there was no opportunity for either isotope or DNA
analysis.41

Trump of war
The Romans … were dismayed by the discipline of the Celtic
army, and intimidated by the blare of countless enemy horns and
trumpets. With the Celts all chanting battle-hymns at the same
time as well, there was so much noise that the sound seemed
not just to ring out from the trumpets and the men, but to echo
from the very hills around them.42

Thus wrote the Greek historian Polybius about the Gauls advancing
into northern Italy in 225 BC. The eerie sound of the war horn has now
been recreated, thanks to an astonishing discovery. It has long been
known that the Celts used a long, slender trumpet called a carnyx, with a
bell shaped like an animal head. They were played upright, as we see on
the Gundestrup Cauldron. [16] Carnyces were not exclusive to the Celts.
They are depicted with captured Dacians on Trajan’s Column in Rome.
The Dacians may have adopted them from Celts who had expanded into
their region in the 4th century BC (see Chapter 7).

16 An interior panel of the huge Gundestrup Cauldron shows a procession of warriors (below)
towards a god-like figure (on the left), with (on the right) trumpeters playing the carnyx. Though
found in Denmark, the cauldron looks like Thracian silverwork, but the carnyces and the helmets
topped by animals are typically Gaulish. Probably the cauldron was made where Celtic and
Thracian peoples lived close together at the time of its production between about 150 BC and 0
BC.

As we know from warfare in times closer to our own, the trumpet can
be used to relay the orders of commanding officers over the din of battle.
So it was with the carnyx. The great hero of the Gauls, Vercingetorix (see
pp. 151–52), who rallied Gallic tribes in one last attempt to repulse
Caesar in 52 BC, sounded the trumpet (presumably the carnyx) to lead
his men into battle.43
The problem for eager re-enactors was that only a few segments of
carnyces had been found prior to 2004. The finest fragment was a boar’s
head from Deskford, northeast Scotland, found in 1816. Since it was
made from recycled Roman metal, it can be dated to the Roman period in
Britain. In 1992 a replica of the complete instrument was created,
reconstructing the missing parts mainly from depictions. This was the
brainchild of musicologist Dr John Purser, keen to bring the music of the
past alive.44 But without a complete ancient instrument, could we be
certain of the replica’s accuracy?
Then suddenly a wealth of evidence was unearthed from just one
findspot. In September 2004 a mass of metalwork was discovered in a pit
at a Gallic shrine at Tintignac in the Limousin region of central France.
Objects had been broken to put them out of use by mere mortals. They
included swords and scabbards, iron spearheads, the metal parts of a
shield and armour, and bronze helmets, one in the form of a swan.
Finally, in the bottom of the pit, were fragments of no fewer than seven
carnyces, as tall as a man. Six have boars’ heads and one a snake’s
head.45
The Limousin region takes its name from the Limovici tribe, whom
Ptolemy identified, together with their town of Augustorium, now
Limoges.46 Southeast of Limoges lies the present-day town of Naves.
The remains of the sanctuary are nearby. The Romans were generally
tolerant of the religions they encountered within their expanding empire.
So their conquest of Gaul did not see the shrine at Tintignac deserted.
On the contrary, a monumental Romano-Gallic complex was erected,
which included a theatre.
The earlier Gallic shrine of the 1st century BC at Tintignac was simpler.
Like most Gallic structures, it was timber-built. It seems to have played
an important part in the life of the tribe. Traces were uncovered of
gatherings for animal sacrifices and banquets, including libations to the
gods – usually of imported Roman wine. A small circular building was
probably a temple. Beside it was the pit with its extraordinary contents,
which propelled the site on to the international stage.47 The contents of
the pit look like the spoils of war. That fits Caesar’s description of the
habits of the Gauls:
When they have decided to fight a battle they generally vow to Mars the
booty that they hope to take, and after a victory they sacrifice the
captured animals and collect the rest of the spoil in one spot. Among
many of the tribes, high piles of it can be seen on consecrated ground;
and it is almost an unknown thing for anyone to dare, in defiance of
religious law, to conceal his booty at home or to remove anything placed
on the piles. Such a crime is punishable by a terrible death under
torture.48
The Romans generally equated Celtic gods with whichever member of
their own pantheon seemed closest in character. A bewildering variety of
Celtic deities is known from inscriptions to them. Many seem local or
tribal. Several were conflated by the Romans with Mars. In the absence
of an inscription, we cannot be certain which war-like god was honoured
at Tintignac. We can only be thankful that religious feeling preserved so
many carnyces for modern scholars.
Though none of the Tintignac carnyces was buried complete, it proved
possible to piece together one boar’s head carnyx from the hoard. [17]
The result may seem surprising. The Gundestrup Cauldron creates the
impression that the carnyx-players were facing forward, which would
suggest that the instrument was curved at the base where it was blown.
The Deskford reconstruction and a number of others were therefore
given a curved mouthpiece, though depictions on coins show a straight
tube. [18] The Tintignac mouthpiece was not curved. Christophe
Maniquet, the archaeologist in charge of the excavation, was curious to
find out exactly what sound a carnyx produced. In 2011 a brass copy was
made of the restored Tintignac carnyx so that it could be tested. It proved
difficult to play, and not as powerful as a modern brass instrument. The
closest modern instrument is an alto saxhorn.49
17 The bell of a carnyx found at a Gallic shrine at Tintignac, France, is in the shape of a boar’s
head. Fragments of seven carnyces were found in a pit at this site.

18 A coin of Tasciovanus, king of the Catuvellauni c. 20 BC to c. AD 9. A riding warrior holds aloft a

carnyx.

Migration mystery
Just the few examples above show that in the last centuries BC there was
a similarity of culture between the Gauls and the island Celts. Indeed,
that similarity was remarked upon by Julius Caesar.50 There might seem
to be a simple explanation. It has been supposed by many authors that
the Celts spread out from Central Europe with the La Tène culture. In this
model the first Celtic-speakers to set foot in Britain would have arrived c.
450 BC and in Ireland not much before 300 BC.51 This theory had a strong
appeal. It was underwritten by Classical sources and La Tène is solidly
identified as Celtic (p. 34). As we shall see in Chapter 7, Greek and
Roman historians vouch for movements of the Gauls in various directions
during this period. The arrival of chariot burials in Britain we can certainly
date to the La Tène period. It seemed a neat fit. Yet La Tène had an
insignificant impact on southern Ireland and most of Iberia, where Celtic
languages were spoken from as far back as we can delve with historical
sources.
Here we turn to the ancient Greeks, literate and adventurous before
the Romans took to the seas. Herodotus tells us that the Phokaeans
were the earliest Greeks to make long voyages by sea. Phokaea (now
Foça, in modern Turkey) was one of a number of Greek colonies in Asia
Minor. The Phokaeans opened up Iberia to trade with the Greek world.
Perhaps they were enticed by a traveller’s tale. In around 630 BC Kolaios,
a merchant captain on the Samos to Egypt route, was blown so badly off
course that he ended up beyond the Strait of Gibraltar at Tartessos. This
wealthy city was virgin territory for Greek traders at the time, and Kolaios
returned home with a huge profit from his cargo.52 The fame of Tartessos
had already spread across the Mediterranean. This mineral-rich region in
what is now southwestern Spain had attracted the Phoenicians before
the Greeks. By the 8th century BC the Phoenicians had founded the city
now known as Cadiz in order to trade with Tartessos around the Lower
Guadalquivir. Phoenician traders were visiting the Huelva area in the
previous century.53 Phokaea created its own colony c. 600 BC on the
Mediterranean coast of what is now France at Massalia, modern
Marseilles, facilitating their trade outside the Mediterranean. In the 4th
century BC the Greek historian Ephorus describes ‘a very prosperous
market called Tartessos, a famous city, with much tin carried by river, as
well as gold and copper from Celtic lands’.54
The Phokaeans reported that the king of Tartessos was called
Argantonios.55 This name is Celtic, derived from the word arganto
(silver).56 Names from this root are attested in Roman times all over
Hispania and further afield, such as the Caledonian chieftain
Argentocoxos, meaning ‘silver-limbed’. On the other hand Argantodannos
is found on Gaulish coinage, referring to the official in charge of minting
coins.57 So Argantonios could also be a title. That might make sense of
the strange claim by Herodotus that he ruled for 80 years and lived for
120. New traders arriving might not realize that they were dealing with a
series of men with the same title. However, Herodotus also reports that
Argantonios was dead before the Phokaeans abandoned their home city
in Asia Minor, when it was under siege by the Persians. That was in
about 540 BC.58
This does not necessarily mean that the language of Tartessos was
Celtic. Celtic-speakers could simply have taken control of trade in those
precious metals that came from their own territory further inland. Celticist
John Koch has argued that inscriptions in the Paleohispanic
southwestern script, adapted from the Phoenician alphabet and dating
from between the 8th and 6th centuries BC, are in the Tartessian
language and that this is Celtic.61 Both suppositions are contested. Only
four of the around 90 inscriptions in this text come from the Tartessian
region. Most are from southern Portugal. Their content is so meagre as to
severely hamper identification of their language, and Koch’s solution is
controversial.62 The fact remains that we have a man with a clearly Celtic
name in Iberia long before the La Tène period.

Genetics: ancient DNA

Today scientists can extract genetic code from the remains of people who walked the earth
thousands of years ago. Early studies of ancient DNA focused on mitochondrial DNA
(mtDNA), since it was easiest to extract. This is found not in the cell nucleus, but in energy-
generating mitochondria throughout the rest of the cell. Since there are up to a thousand
mitochondria per cell, it is the most abundant human form of DNA in terms of copies. That
made it the natural first choice for pioneers trying to extract DNA from ancient bones and
teeth. MtDNA is passed down from mother to child. Men do have mitochondrial DNA – it
comes from their mothers. But they cannot pass it on.

Table 1 The first appearance of each Y-DNA haplogroup so far found in European ancient DNA up
to the Copper Age.
 The results overturned cherished beliefs. The standard view at the end of the 20th
century was that Europeans are mainly descended from European hunter-gatherers.
Ancient mtDNA revealed an influx of farmers from the Near East in the Neolithic, whose
haplogroups predominate in modern Europeans.59 A new generation of technology has
made it possible to extract ancient Y-DNA, which confirms the influx of farmers. Haplogroup
I, found in hunter-gathers, represents less than one-fifth of the present European
population, while C1a2, found in a Spanish hunter-gatherer, is vanishingly rare in modern
Europeans.60 The first farmers brought new Y-DNA haplogroups such as G2 from the Near
East, while the R1a and R1b that dominate Europe today are first found in Russian hunter-
gatherers and did not arrive in Central Europe until the Copper Age.

The earliest known names for Britain and Ireland are also Celtic. These
names were familiar to seafarers before La Tène material arrived in
either island. Or so it appears. As we head backwards from the La Tène
period, historical sources disintegrate into tantalizing snippets, acquired
secondhand. In the 4th century AD a Roman named Rufus Festus
Avienus laboured over a poem designed to display his erudition rather
than personal knowledge: Ora Maritima. It describes the coast from
Brittany to Marseilles, not in his own day, but drawing on authors from the
6th and 5th centuries BC. He credits Himilco with knowledge of what lay
north of Brittany. This Carthaginian explorer was sent to investigate the
remote northwestern shores of Europe around 500 BC.63
The Ora Maritima begins the poetic voyage by describing a high ridge
known in an earlier age as Oestrymnis, facing south and sloping down to
the sea. Beneath it lay the Oestrymnic Bay, in which islands called the
Oestrymnides were widely spread. A tribe in what is now western Brittany
was known as the Ostimioi to the Greek traveller Pytheas in the 4th
century BC. So a modern translator is no doubt correct to identify
Oestrymnis as the island of Ushant (Uxisame to Pytheas) off the extreme
western point of Brittany. To the east of it is the Bay of Douarnenez. A
chain of islands spreads from Ushant to the bay.64 The Ora Maritima
goes on to say:
From here it is a two-day voyage to the Sacred Isle, for by this name the
ancients called the island. It lies rich in turf among the waves, thickly
populated by the Hierni. Nearby lies the island of the Albiones. The
Tartessians were accustomed to trade even to the edge of the
Oestrymnides. The Carthaginian colonists and people around the Pillars
of Hercules frequented these waters. Four months scarcely is enough for
the voyage, as Himilco the Carthaginian proved by sailing there and back
himself.65
‘Sacred Isle’ is a misconstruction of the name of Ireland. In ancient
Greek hieros meant ‘sacred’. The Hierni are easily identified as the Irish
and the Albiones as the people of Britain. The Irish name for themselves
can be reconstructed as Iwerni, from the Irish name for Ireland, Iverio
(‘the fertile land’). That evolved into Old Irish Ériu and modern Irish
Éire.66 Albion corresponds to the Old Welsh elbid and Middle Welsh
elfydd, with meanings ‘world, earth, land, country, district’.67 Here are
Celtic names dating from two or three centuries before La Tène material
appeared in Ireland.
Awareness of these flaws in the concept of Celtic spreading only with
La Tène has generated new interest in an alternative model, proposed
intermittently since the 1930s, in which the Bell Beaker culture was the
vector for the earliest forms of Celtic.68 Support for this model is starting
to arrive from ancient DNA. Yet we shall see in the next chapter that a
straightforward equation of Bell Beaker with Celtic may over-simplify a
messier reality.

Overview
• Two successive Iron Age archaeological cultures, Late Hallstatt and
La Tène, are found in the right place and time to correspond to
Classical references to the Celts.

• The Gauls called themselves Celts. They spoke a language of the

same family as Breton, Welsh and Gaelic.

• The La Tène culture spread to Britain. There are striking similarities

in culture between Gaul and Britain, such as: • The use of chariots.
• The use of a type of trumpet called a carnyx.

• This similarity encouraged the idea that Celtic languages spread

from Central Europe with the La Tène culture. There are two
problems with this theory: • La Tène material is barely found in
Iberia or southern Ireland.
 • Celtic place-and personal names appear in the British Isles and
Iberia before the La Tène period.

• So was the Bell Beaker culture the vector for the Celtic languages?
Or is that an over-simplification? These questions are tackled in
Chapter 3.
CHAPTER THREE

Bell Beakers and Language

The change in the properties of copper by heat is really very

startling; it is distinctly more dramatic than the effect of baking
upon potter’s clay.… Even more startling and mysterious were
the transmutations involved in the extraction of the metal.… It
was a stupendous feat … to connect the green crystalline
stones with the tough red metal. The recognition of the
underlying continuity marked the beginning of chemistry.1

The seeming magic of metallurgy has seldom been more vividly conjured
up than by the passage above from Vere Gordon Childe, published in
1930. This influential archaeologist understood that the earliest smiths
were privy to powers that would have been awe-inspiring in their era.
Metal management lay at the centre of the Bell Beaker culture.
Bell Beaker (c. 2800–1700 BC) leapt across Europe, not settling
everywhere, but picking certain patches, often ones with ores to exploit.
[19] It is recognized by its characteristic pottery, shaped like an inverted
bell. Some later cultures of the Bronze and Iron Ages were quite
widespread within western Europe, but none extended over the whole
area anciently Celtic. This is why a series of archaeologists have looked
towards Bell Beaker for the origins of the Celts.2
Is it really that simple? Bell Beaker in Italy covers an area where Celts
did not appear until they drove out the Etruscans from the Po Valley in
historical times (see Chapter 7). It also appears as far south as Sicily and
Calabria. Could it be that in its earliest stages Bell Beaker was
associated with both Celtic and Italic languages, or with an ancestor to
both? If so which ancestor? Could it be a type of Italo-Celtic,3 or
something even earlier? We shall consider this question below, but first,
what exactly was Bell Beaker?
19 Bell Beaker pottery spread by sea and river routes. It was probably made by women. So its
dispersal may be partly linked to a search for marital partners among the scattered kin of a mobile
Copper Age people.

The bell-shaped pot

Bell Beaker pottery is distinctive. The inverted bell shape is decorated
with horizontal zones of patterns, which can be incised or impressed into
the wet clay. Where such pots are particularly well preserved, the white
paste that was rubbed into incisions to pick out the pattern may still be
visible. [20]
 The concave sides of the typical Bell Beaker would make it easy to
grasp, so archaeologists have usually seen it as a drinking vessel, hence
the name ‘beaker’. That image led to a vision of male drinking rituals
cementing the culture.4 Beer was thought to have arrived in Europe with
Copper Age cultures such as Bell Beaker. Indeed the powerful appeal of
alcohol was seen as one source of the influence of the Beaker folk. Vere
Gordon Childe thought the heady fumes of beer could have drawn the
earlier farmers of Europe into the Bell Beaker net.5
The popularity of alcohol can scarcely be denied. It features in
countless cultures. Therein lies one problem with seeing it as the Bell
Beaker liquid weapon. As scientific techniques in archaeology have
advanced, the evidence for fermenting and brewing has been pushed so
far into the past that beer is challenging bread as the presumed staple of
early farmers. Perhaps as early as 9000 BC people gathered to feast and
drink beer at the world’s first megalithic monument, Göbekli Tepe in
Anatolia.6 In that case brewing techniques could have travelled into
Europe with early farmers. Residues of alcoholic beverages have indeed
been found in Neolithic Scotland.7
So Beaker folk could not have introduced beer to western Europe.
They certainly drank it though. Chemical analysis of Bell Beaker vessels
from tombs in the Ambrona Valley in Spain found residues of a primitive
wheat beer.8 Beer and mead (made from fermented honey) have been
identified from other examples too, but not all Beakers were drinking
cups. Some were used to smelt copper ores, others contain food
residues, and yet others were funerary urns. Bell Beaker pots vary
considerably in size. The largest, with a capacity up to 20 litres (over 4
gallons), were scarcely ideal for drinking.9 So perhaps we should see the
waist of the bell shape as a more general lifting aid. There would be less
likelihood of such a pot slipping out of the hands of its bearer.
20 Replica of a Bell Beaker found at Ciempozuelos, Spain. The incised patterns are picked out in
white paste made of crushed bone.

Bell Beaker ware is found as far east as Poland,10 as far south as

northern Morocco and as far north as Scotland,11 northern Denmark12
and even the southern tip of Norway.13 Either its makers were mobile or
this was a popular trade item. These days scientists can probe a pot’s
origins. In 2012 an almost complete bell-shaped beaker was uncovered
at Cranbrook near Exeter in England. Under the microscope a geologist
could identify the minerals in the clay as so local that the pot was
probably made on the site where it was found.14 This is typical for Bell
Beaker in southwest England,15 and fits a wider pattern. Local origin was
the most common conclusion from testing a selection of Bell Beaker
pottery in France.16 This discovery triggered a systematic programme of
testing Bell Beaker pottery from over a hundred sites in France, Portugal,
Spain and Switzerland. It found that most pots were locally made, varying
from just over 50 per cent to 100 per cent.17 So while there was some
movement of actual pots, in the main what moved was knowledge of how
to make them.
This pottery was not wheel-thrown in a workshop. It was made in the
home. So it is most likely that it was made by women. In a society in
which women generally moved on marriage to the home of their spouse,
the knowledge of particular pottery techniques and styles would travel
too.18 The genesis of Bell Beaker pottery has been much debated. That
fraught question is discussed in Chapter 5, along with the origin of the
whole culture.
Way of life and death
Bell Beaker folk ushered the Bronze Age into western Europe. They were
the first metalworkers to enter the British Isles, homing in on the copper
belts of Ireland and Wales. Around 2400 BC they left their characteristic
beakers at a copper mine on Ross Island, in Lough Leane, Co. Kerry.
This is the earliest known copper mine in northwestern Europe. There
can be no doubt that it was created by incomers, for they brought with
them an already advanced knowledge of metallurgy. These experts were
probably looking especially for arsenic-rich copper ore, and they certainly
found it at Ross Island. An arsenic-copper alloy made a tougher metal
than pure copper. The prized ore was smelted on site into copper ingots,
which could be moved elsewhere to be cast into finished objects.19
Analysis of chemical composition shows that copper from Ireland was
traded into Britain.20
From around 2200 BC Bell Beaker interest in Britain intensified as
Cornwall was discovered to be a prime source for tin, the rare and
precious component of true bronze. It has recently been realized that
Ireland too had tin in the Mourne Mountains.21 These resources gave the
British Isles a head start in western Europe in making bronze.22
Tin was also available in western Iberia and the Erzgebirge Mountains
in Central Europe. The abundant tin and copper of Tartessos (p. 43)
might lead you to expect the earliest Iberian bronze in southwestern
Spain. Instead it appears in the northwest of the peninsula, which was
linked into a trade network that included the British Isles. At one site the
whole process of metalworking is laid bare. High in the mountains of the
District of Bragança in Portugal, the detritus of bronze-working has been
found on the hilltop of Fraga dos Corvos. Here a small rural community
tended their flocks, hunted deer, and made their own pottery and bronze
axes around 1750–1500 BC. The pottery includes Epi-Bell-Beaker, a final,
crude representative of the type. So in Iberia too we see Bell Beaker at
the beginning of bronze-making.23 This interesting site has inspired a
novel in Portuguese titled The First Alchemist.24
Bell Beaker burials vary by region and status. It has long been noticed
that in Iberia and Ireland, Bell Beaker people were willing to re-use
Neolithic megalithic tombs. In addition, wedge-shaped tombs were dotted
along the west of Ireland in the Bell Beaker period. These are undeniably
megalithic in construction, but of Bell Beaker date and peculiar to Ireland.
They usually contain more than one interment or cremation.25 It has been
supposed that this pattern of collective burials was exclusive to the
southern and Atlantic zones of the Bell Beaker culture, while individual
graves were the rule in the eastern Bell Beaker region. A recent review
overturns this idea. It seems that the re-use of earlier monumental tombs
was common throughout the Beaker territory. It is simply more visible in
the zone with most monumental graves. Likewise, individual graves are
also found throughout.26
Most people were buried quite simply, but burials of high-status
individuals were more elaborate. A burial chamber could be timber-lined.
Finely decorated Bell Beakers were placed in it by mourners. The
deceased was clothed, but since fabric rots away, only imperishable
objects such as the characteristic v-shaped buttons and archers’ wrist-
guards survive. Tanged arrowheads are usually found with wrist-guards.
Archery equipment is typical of male burials, but one woman buried at
Tišice in central Bohemia had arrows and a wrist-guard.27
A few burials are extraordinary. In the Bell Beaker necropolis at Hulín-
Pravčice, Czech Republic, a group of large graves have inner structures
and rich equipment. Besides the usual high-status set of objects, they all
have copper daggers. (Copper knives are found in less than 6 per cent of
Bell Beaker graves.) A profusion of jewelry made from gold, silver,
electrum, amber and copper included a pair of gold spirals, probably
used to bind the ends of hair plaits. Sets of stones among the grave
goods are interpreted as symbolic of tools used by smiths. Mastery of
metal could reap rewards.28
On the other side of Europe we find another honoured metalworker.
The Amesbury Archer was buried near Stonehenge around 2350 BC. [21]
His skeleton reveals that he was a man aged between 35 and 45, who
must have been in constant pain in the last few years of his life. An
abscess on his jaw had eaten into the bone. His left kneecap had been
ripped off, perhaps in a riding accident, resulting in a bone infection. Had
he come to Stonehenge in hope of a cure? Tim Darvill’s vision of
Stonehenge as a healing shrine has won little support among his fellow
British prehistorians, yet the idea of places with special powers runs deep
and it seems this massive monument drew people from far and wide.
Isotope tests on the Archer’s teeth and bones show that he spent his
younger years in the Alpine region. Altogether around 100 objects were
buried with the Archer, making his the richest Bell Beaker grave found in
Britain. His mourners placed beside him five Beaker pots and two wrist-
guards. Significantly, he was also buried with a cushion stone, thought to
be used by metalworkers. He had no fewer than three copper knives.
Two could be from northern Spain and the third from western France; it
seems the Archer had travelled widely. His gold hair binders are the
earliest gold artifacts found in Britain.29
A similar gold hair binder and cushion stone were found in a burial
mound at Kirkhaugh in Northumberland in 1935, together with a bell
beaker. The matching hair binder was found there in 2014. The grave lies
on the edge of an orefield. So the man buried at Kirkhaugh could have
been part of a team prospecting for copper in the north Pennines.30
Gold was prized by the Bell Beaker elite. It is too soft for practical uses,
but its eternal glitter is ideal for ornament. Southwestern Iberia abounded
in gold in the 3rd millennium BC. It was worked using the same
technology as for copper.31 Ireland had gold in the Mourne Mountains.32
Britain too had several sources of the precious metal. No wonder then
that we find a wealth of gold objects in the British Isles from between
2400 and 1400 BC. A type of golden collar or pectoral known from its
crescent shape as a lunula is not found in burials and so was probably
not a personal item, but something akin to a priestly pectoral which would
be passed on to a successor. The finest are incised with complex
geometric patterns related to Beaker pottery.33 Lunulae and golden discs
are found along the entire Atlantic region from Portugal to the British
Isles, though are particularly common in Ireland. We may see them as
symbols of the moon and sun. Indeed discs and diadems appear right
across Bronze Age Europe, suggesting a common perception of their
significance.34
21 The burial of the Amesbury Archer. This man was buried near Stonehenge around 2350 BC
with a rich array of symbols of his prestige. His grave goods included the arrowheads and wrist-
guards that led his discoverers to name him the Amesbury Archer.

Mobility
When the Bell Beaker culture was first recognized, it was taken for
granted that it arrived with immigrants. The anti-migrationism that arose
in British archaeology in the 1970s changed the perspective. For
decades afterwards Bell Beaker was viewed as a purely cultural
phenomenon. The present century has seen the pendulum of opinion
swing back, as evidence of mobility mounts.35 We have already noted
above (pp. 51, 53) that Bell Beaker arrivals at Ross Island must have
been incomers, and that the Amesbury Archer certainly was.
Unfortunately isotope testing can only detect first generation migrants.
A few metres away from the Amesbury Archer was the burial of a
younger man, who was born in Britain and died a generation or two after
him. The presence of a rare trait in the bones of their feet showed that
the two were related.36 Given that the Bell Beaker culture lasted many
centuries, we would expect to find that the overwhelming majority of Bell
Beaker burials were not of first-generation migrants. Indeed an isotope
study of 250 individuals from all over Britain has shown exactly that,
though local and regional mobility seems common.37 A similar pattern of
mobility appears in a study of Bell Beaker burials in Austria, Bavaria,
Czech Republic and Hungary. Out of 81 individuals, 51 had moved
during their lifetime.38
It has long been recognized that Bell Beaker skeletons in northern,
central and eastern Europe had a different skull shape from previous
people in these regions.39 Initially this was taken as straightforward
evidence of new arrivals. However, the position in which an infant is
placed to sleep can affect skull shape, so DNA is a better guide to the
degree of relationship between populations.

Celtic and Italic languages

The Celtic and Italic language families both belong within the huge family
that linguists label Indo-European (Chapter 4). So they have similarities
that go back to the common parent of all Indo-European languages. That
is without question. As we shall see, there are some similarities shared
just between Celtic and Italic which suggest a closer relationship.

Genetics: Y-DNA R1b-P312

R1b1a2 (M269) was found in human remains from a Bell Beaker site in Kromsdorf,
Germany.40 This Y-DNA haplogroup has many subclades, generated by an explosive
population growth. The growth spurt was assumed to date from the Neolithic,41 before
ancient DNA results made the Copper Age more likely. One subclade predominates in
western Europe today: R1b1a2a1a2 (P312) or R1b-P312 for short. [23]
Any mutation happens first in just one person. So all men today carrying R1b-P312 or
one of its descendants have a common ancestor. Where and when he lived we shall never
know exactly. However, R1b-P312 has been found in a man of the Bell Beaker culture at
Quedlinburg, Germany (see box pp. 44–45). Today the concentration of Y-DNA R1b-P312
falls most heavily where Celtic languages were once spoken and in Italy, the home of the
Italic language family. We often find a correlation between Y-DNA haplogroups, which are
passed down from father to son, and languages.

23 Y-DNA R1b1a2a1a2 (P312) is a large and widespread haplogroup, which predominates

in regions of Europe where Celtic and Italic languages were once spoken. Some of the
highest levels are in Wales and Ireland.
24 The languages of the Italian peninsula and neighbouring territory c. 550 BC. Italic
languages are picked out in colour.
 25 The density of Celtic place-names recorded in Classical sources. Such sources are more
plentiful within the former Roman empire.

Today Romance languages such as French, Italian, Spanish and

Portuguese are spoken by many millions of people. They all spring from
Latin, which was spread by the Roman empire. Before the rise of Rome,
Latin was just one of a group of Italic languages confined to central Italy,
sandwiched between Etruscans in the north and Greeks who had
colonized southern Italy. [24] That picture was to change dramatically as
the Romans gradually dominated the other Italic-speakers, then the
Etruscans, and finally burst out of Italy to create an empire that engulfed
much of Celtic-speaking Europe (Chapter 8).
We need mentally to turn back that Roman tide if we are to reconstruct
the linguistic pattern of western Europe as far back as we can. Roman
geographers recorded a wide scatter of Celtic place-names inside the
empire and even beyond it. [25] The wealth of Celtic place-names in Gaul
supports Caesar’s observation that most of Gaul was Celtic-speaking
before the Roman conquest. Iberia too was well supplied with Celtic
place-names in Roman times, though there are interesting gaps, which
will be discussed later. The Romans also recorded many place-names of
Celtic origin within their province of Britannia. The absence of Celtic
place-names in western Ireland and northern Scotland on the map should
not disturb us. It simply reflects the way that it was created. Patrick Sims-
Williams analysed, by one-degree squares, what percentage of place-
names noted in the Barrington Atlas of the Greek and Roman World
(2000) could be derived from Celtic.42 Those labouring on the Barrington
Atlas had to contend with the limited data-gathering by Romans outside
their own borders. Virtually all the information for Ireland comes from
Ptolemy. Few of the Irish place-names he recorded can be located today
with certainty, so the Barrington Atlas does not attempt to plot them all.43
The Scottish Highlands were thinly populated, with few settlements
known to Roman geographers, but Ptolemy tells us that ‘From the Bay of
Lemmanonia [Firth of Clyde] until the Varar estuary [Moray Firth] dwell
the Caledonians, above whom is the Caledonian forest’.44 The tribal
name Calidonii seems derived from Celtic calet, which has meanings
including ‘tough’ and ‘hardy’, entirely suitable for the highlanders.45 It is
preserved in Dunkeld (9th-century Dun Chaillden) and Schiechallon.46
Next we need to peel away the layers of known migration before the
Roman conquests. Celtic place-names in the Po Valley, the Balkans and
around the Black Sea can be attributed to movements of the Gauls south
and east into formerly non-Celtic territory after 500 BC (Chapter 7). By
contrast, the scattered Celtic place-names east of the Rhine tell us that a
Celtic language was in use there before the expansion of Germanic
speakers from 500 BC. The name of the River Rhine itself is Celtic in
origin.47 It lay well within the zone of influence of both the Hallstatt and La
Tène cultures. [see 11] According to an anonymous Greek poet of about
200 BC, the Rhine was a sacred river to the Celts:
The bold Celts test their children in the jealous Rhine And no man
regards himself as a true father Until he sees the child washed in the
holy river.48
Untangling the story for Britain and Iberia is more complex. Roman
writers tell us that Gauls moved into Iberia, and Belgae moved into Britain
before the Roman conquest of those regions. The very fact that these
migrations were known suggests that they were not in the far distant past
for Romans. That is supported by archaeological evidence of culture
changes in the early centuries BC (Chapter 7). Yet we saw in Chapter 2
that sea-going Greeks were recording Celtic names in Iberia and the
British Isles long before that. So the pattern seems to be one of waves of
Celtic migration, the earliest of which were before recorded history.

Italo-Celtic
The Celtic and Italic language families have similarities that suggest a
common ancestor more recent than the parent of all Indo-European
languages. They share the o-stem genitive singular ending in ‘i’, for
example the Latin viri (of a man), and Primitive Irish maq(q)i (of the son).
A joint innovation is the superlative suffix ismmo, in such formations as
Latin maximus (greatest), and the Gaulish place-name Ouxisame
(highest). There is also a subjunctive morpheme -ā-, as in Latin fer-ā-t
(he may carry) and Old Irish beraid (he may carry). So some linguists
argue for a common ancestor for the two families which they call Proto-
Italo-Celtic. Non-linguists have sometimes misunderstood this to mean a
type of Celtic. So it needs to be clearly stated that it is not. It is the
proposed ancestor of both language families. The alternative explanation
for their shared features is that Proto-Celtic and Proto-Italic developed in
such close proximity that they influenced each other.49
The case for a common ancestor was first made in 1861, was then
countered in 1929 and has oscillated in and out of favour since. Calvert
Watkins of Harvard University forcefully made the case against in 1966.
He pointed out the many dissimilarities between Proto-Celtic and Proto-
Italic.50 Indeed these suggest a comparatively short period of common
evolution, followed by a long period of separate development, as argued
by Dutch linguist Frederik Kortlandt.51 This does not dispose of the
common ancestor, whose existence has continued to garner support
among linguists.52 One good reason is that there are several lost
languages, such as Ligurian and Lusitanian, that do not fit into either the
Italic or Celtic branches, but are related. We could see them as
descending from Proto-Italo-Celtic. They were also spoken within the
regions colonized by Bell Beaker.
So could Proto-Italo-Celtic be the first language carried by the people
of the Bell Beaker culture? This was the solution I previously favoured,
since ancient Greek observers reported a coastal band of Ligurians all
the way from northwestern Italy to western Iberia, where the Romans
encountered Lusitanians.53 But recent work suggests that the Lusitanians
and their kin did not arrive in Iberia until the Late Bronze Age.54 So we
must look to a yet earlier language.

Old European IE (Alteuropäisch)

Hans Krahe noticed river-names across Europe which appear to be Indo-
European, but do not fit any known Indo-European language. He called
this the Alteuropäisch (Old European) hydronymy.55 Krahe saw such
names as evidence of a lost language ancestral to the western branches
of Indo-European. Critical dissection left this idea bleeding to death.
Specific archaic Indo-European river-names range from reflecting the
original Indo-European parent to dialects of it that had not quite become
fully fledged daughter languages.56 Nevertheless Krahe’s work is
important. It helps us to realize how complex the process of language
spread can be. Many parts of Europe seem to be like a linguistic layer
cake. One wave of Indo-European was succeeded by another.
Place-names that can be identified as Indo-European, but neither
Celtic nor Italic, are found in Iberia, particularly in the south. Most
obviously there are many place-names starting with the letter ‘p’, which
was lost in early Celtic. Modern place-name scholars generally class
these as Old European.57 This means that the first language of Bell
Beaker communities in the west may have been a form of Indo-European
too early even to have features specific to both Celtic and Italic.58 So we
need to press on further back, to the origins of the Indo-Europeans.

Overview
• The Bell Beaker archaeological culture (c. 2800–1700 BC) is
characterized by: • Pottery shaped like an inverted bell.
 • Metalworking in copper (later bronze) and gold.

• Mobility. It spread widely in Europe, and even into Morocco.

• A focus on archery.
• The Bell Beaker culture is found in all regions of Europe where
Celtic languages were subsequently spoken.

• Bell Beaker is also found where Italic languages were subsequently

spoken, as well as languages such as Ligurian, which appear to be
related to Italic and may derive from a common ancestor of Italic
and Celtic.
• Place-names over the Bell Beaker region include some which are
classified as Old European (Alteuropäisch), a dialect of Indo-
European too early to fall into any particular branch of the family.

• So the earliest Bell Beaker makers may have spoken Old European.
To better understand its origins, we turn in Chapter 4 to the Indo-
European language family.
CHAPTER FOUR

The Indo-European Family

The Sanskrit language, whatever be its antiquity, is of a

wonderful structure; more perfect than the Greek, more copious
than the Latin, and more exquisitely refined than either, yet
bearing to both of them a stronger affinity, both in the roots of
verbs and in the forms of grammar, than could possibly have
been produced by accident; so strong indeed that no philologer
could examine them all three, without believing them to have
sprung from some common source, which perhaps no longer
exists: there is a similar reason, though not quite so forcible, for
supposing that both the Gothic and the Celtic, though blended
with a very different idiom, had the same origin with the
Sanskrit; and the old Persian might be added to the same
family.1

Thus spoke Sir William Jones (1746–1794) as he celebrated the third

anniversary of the learned society that he had founded in Calcutta. The
date was 2 February 1786, only about six months after he had begun to
study Sanskrit, the ancient literary language of India. He had been a
linguistic prodigy even as a child. The son of a celebrated Welsh
mathematician nicknamed ‘Longitude Jones’ and a gifted English mother,
the young William Jones not only learnt the customary Latin and Greek at
Harrow School, but taught himself Hebrew and the Arabic script. At
Oxford University he improved his Arabic and became fascinated by
Persian. In his vacations he read European classics in Italian, Spanish
and Portuguese. In 1770 William Jones entered the Middle Temple,
London, to study law. As a barrister in Wales from 1775, he became a
keen Celticist. So when he arrived in India in 1783 to take up a position
on the Bengal supreme court, he was an outstanding linguist.2
 The correspondences between Sanskrit and the Classical languages of
Europe are striking. For example, the word for ‘father’ in Latin and Greek
is pater and in Sanskrit is pitar. Likewise the word for ‘mother’ in Latin is
mater, in ancient Greek is meter and in Sanskrit is matar. Such
resemblances had been privately noted by others. Indeed a proposal that
Latin, Greek and Sanskrit had a common origin had been sent to the
French Academy by Jesuit missionary Gaston Coeurdoux (1691–1777)
some years before the famous statement by William Jones, though it was
not published until 1808. Coeurdoux and Jones took the same approach,
looking for similarities not simply between words (for words can be
borrowed between languages), but also in grammar. A methodology was
born. Linguistics had entered the realm of science. By 1813 linguists had
formulated a model of a language family labelled Indo-European.3 It has
been intensively studied since. [26]
26 A tree of Indo-European languages, adapted from Nakhleh, Ringe and Warnow 2005. The
date scale indicates the estimated time that a group broke away from the Proto-Indo-European
parent, so that its speech developed independently and became a daughter language, and then
the estimated time of any splits in that daughter language. The first appearance of a language in
writing is indicated by the names in small capitals. The names in bold, such as ‘Balto-Slavic’ and
‘Proto-Indo-European’ are the creation of linguists; these languages were not recorded in writing.

So now we know that Celtic belongs to the Indo-European family of

languages. [27] How do such families arise? In the days before modern
communications, any language had to be spoken face to face, with the
people you met day to day. The language of a community would
gradually change. For example new words would be created, perhaps for
new inventions, and everyone in the community would soon know them.
If a group split from the community and moved so far away that it could
not talk constantly to the parent group, then it would no longer be part of
the parent language evolution. Instead it would develop its own dialect.
Eventually the parent and child communities would speak different,
though related, languages. As Sir William Jones surmised, the parent of
all the Indo-European languages was long gone by his day. Linguists
have painstakingly reconstructed it; it is known as Proto-Indo-European
(PIE).

27 Indo-European languages in AD 1500. From a Copper Age homeland on the European steppe,
Indo-European languages spread far and wide. It is now the dominant language family in Europe.
Names of non-Indo-European languages are in italics.

Indo-European homeland
Where was the birthplace of PIE? The lack of inscriptions or other ancient
writing in PIE set free the wildest imaginations. As J. P. Mallory put it:
This quest for the origins of the Indo-Europeans has all the fascination of
an electric light in the open air on a summer night: it tends to attract every
species of scholar or would-be savant who can take pen to hand.4
 He took on the unenviable task of sifting the evidence. A scholar of
both linguistics and archaeology, he gave equal weight to both. Other
theorists have tended to make a linguistic case for which there is little or
no archaeological support, or vice-versa.
Cambridge archaeologist Colin Renfrew argued that the Indo-
European languages were brought to Europe by farmers from Anatolia in
the Neolithic.5 This was a bold and attractive hypothesis. There is little
doubt that many of the language families spoken today spread with
agriculture.6 PIE certainly includes farming terms, but it reflects later
innovations as well. The first farmers used digging sticks rather than
ploughs. They had no wheels or wagons, no gold or silver. They kept
cattle for beef, not milk and cheese. They did not make wine. They did
not spin wool. Yet PIE had words for all these things.7 So dating the
spread of Indo-European as early as the first farmers did not appeal to
linguists.
Regardless, the Anatolian model continued to have its supporters.
Some biologists have claimed to prove it by applying a methodology from
epidemiology.8 Would that it were so simple. Any assumption that
humans will behave exactly like viruses is doomed to crash against
reality. Humans do not move randomly. They are capable of weighing up
geography, climate, hazards and convenient routes. Worse still, the
conclusion relied on computer models with a built-in bias. Advocates of
the Anatolian homeland had always placed great weight on the primacy
of the Anatolian branch of the Indo-European family in two respects: it
was the first to split away from the parent and the first to be committed to
writing. By devising a computer model that placed the Anatolian branch
at the root on these grounds, the devisers simply gave the Anatolian
homeland theory pseudo-scientific cladding.9 An independent analysis
failed to replicate their results.10
It is always exciting to find tangible evidence of a language – an
inscription, a papyrus, a clay tablet with signs that we can decipher. It is a
message from the past. It is easy to forget that for millennia our
ancestors spoke to each other without putting anything in writing. Literacy
goes hand-in-hand with civilization: a complex, centrally organized
society run by a bureaucracy. The vocabulary of PIE does not fit an
urban society with all the trappings of a legal system, officials and taxes.
So we should not be surprised that it was never written down. The
earliest languages attested in writing were those of the first civilizations.
These included languages in Anatolia, such as Hattic, that were
unrelated to Indo-European. As speakers of other languages came into
contact with civilization, they too could adopt literacy. This was the
pattern in Anatolia.
The Indo-European-speakers we call Hittites called their own language
Nešili, meaning ‘the language of Neša’. Kaneša (shortened to Neša) was
the Hittite name for a town in central Anatolia known as Kanesh to the
Assyrians. The site of Kanesh at Kültepe has been excavated, so we
know the Assyrians had a merchant colony there. Assyrian merchants
mention Hittite names in their texts, but the Hittites themselves left us no
word until their king Pithana of Kussara conquered Kanesh ‘in the night,
by force’, but without doing any damage. His son Anitta had an inscription
made to record that moonlit drama and his own conquest of the Land of
Hatti in central Anatolia.11 In short, the Hittites turned to writing as they
rose to power and acquired literate bureaucrats.
The crucial point linguistically is that Hittite took features from Hattic,
but PIE did not.12 There is no indication of PIE being in contact with any
of the languages of Anatolia. The PIE homeland lay elsewhere.
The chief attraction of the Anatolian theory was its simplicity. If we look
for a massive cultural upheaval across the whole of Europe, the coming
of farming leaps immediately to mind. The pattern emerging from ancient
DNA does indeed support mass migration in the Neolithic. Yet Copper to
Bronze Age movements also stirred the European gene pool with
startling results in Y-DNA (see box pp. 44–45), now supported by full
genome comparisons (see box p. 73).
J. P. Mallory, and more recently David Anthony, championed the
alternative hypothesis that PIE spread later, along with metallurgy, from
the Pontic-Caspian steppes.13 Put simply, PIE can be located in time by
its vocabulary and in place by its neighbours. PIE-speakers created their
own words for wagons and wheels, from Indo-European roots. These
words were retained in a number of daughter languages, with changes
characteristic of those languages, which is how linguists trace them to the
mother language. So PIE cannot date before the invention of the wheel
around 3500 BC.14
Crucially, PIE evolved in contact with Proto-Uralic, parent of the
language family which includes Finnish and Saami. Farming vocabulary
was absorbed from PIE and its offspring by Uralic. The Uralic family
takes its name from the consensus view that the parent language
developed near the Ural Mountains. So we can deduce that PIE was
spoken somewhat to the south of Proto-Uralic, closer to the sources of
farming.15 PIE in turn borrowed words from more southerly languages.
The words for ‘bull’ in PIE (*tawro-s) and Proto-Semitic (*tawr) are clearly
related. This formed the crux of linguistic arguments in favour of a
homeland for PIE in Neolithic Anatolia. Yet Proto-Semitic does not belong
in that niche. Its lexicon places it in the Copper Age Levant.16 So contact
between PIE and Proto-Semitic fits neatly into the Copper Age, but what
about the geography? There are linguistic clues that such words trickled
through the Caucasus.17 Culturally that is feasible. The Maikop culture of
the North Caucasus (see below) was in contact with both Mesopotamia
and the North Pontic steppe.18 The conclusions from linguistics can be
anchored in archaeology. The Yamnaya culture on the European steppe
fits the type of society PIE leads us to expect (see pp. 72, 74). Then we
see the influence of Yamnaya moving east and west to places where
Indo-European languages later emerge (see pp. 75–78).
For the moment, though, let us continue our detective journey back in
time. Indeed we take a massive leap to the age of mammoths. Today Y-
DNA R1 dominates Europe, but it first appeared on its eastern fringes. Its
ancestor R has been found in an ancient burial in Siberia. With this
discovery another piece of the puzzle may have fallen into place. The
relationship of Uralic to PIE runs deeper than the adoption of farming
terms. The two languages share such fundamental vocabulary as the
words for ‘water’ and ‘name’. The explanation could be contact at an
early stage, before Uralic and PIE were fully formed. A language
ancestral to Proto-Uralic was probably spoken somewhere in the Sayan
region of south-central Siberia.19

From Siberia to Europe

In the valley of the Angara River, which drains the massive Lake Baikal in
central Siberia, a four-year-old boy was buried 24,000 years ago. His
remains were discovered in the late 1920s near the village of Mal’ta and
taken to the Hermitage Museum in St Petersburg (then Leningrad),
Russia. It looks as though he was much loved. He was buried with a
beaded necklace, several pendants and an ivory diadem. His people
were reindeer and mammoth hunters. They carved Venus figurines from
mammoth tusk, some of which are marked to suggest fur clothing. They
lived at a time when the north was in the grip of vast glaciers and
Siberian mountains were ice-covered. Hardy hunters clustered in the
shelter of the pine-forested valleys of southern Siberia. At Mal’ta they
lived in round huts with a central fire that would have given both heat and
light. Perhaps the long winters encouraged art and craft, though no other
Siberian site of the period can compare to Mal’ta in richness and diversity
of art objects. Apart from the female figurines, there are bird carvings and
the engraving of a mammoth.20 [28] Mal’ta boy’s people were creative.
Mal’ta boy himself has given us something of great significance – his
DNA. This has been analysed, with some surprising results.
Some 4,000 years after Mal’ta boy a new stone tool-making technique
arose in his region, called pressure blade-making. It spread both east into
North America and west to the borders of Europe.21 [29] So could it have
been invented by Mal’ta boy’s relatives? The team who tested his
remains also managed to extract DNA from a man who lived about
17,000 years ago at Afontova Gora, a site with pressure blade-making.
Afontova Gora man was indeed similar genetically to Mal’ta boy (see box
p. 70).

28 Around 24,000 years ago a hunter scratched this image of a mammoth on a slab of mammoth
tusk. It was found at Mal’ta in Siberia, along with other evidence of the creativity of the small band
of people who sheltered there, such as female figurines and bird carvings.

Pressure blade-making was a remarkable innovation. For millennia

humankind created tools by directly striking a piece of flint. Breaking
stones without striking them is more complicated. Obsidian was the
preferred material. This volcanic glass fractures into pieces with curved
surfaces and very sharp edges. Freshly cut obsidian is sharper than a
surgical steel scalpel. By firmly clamping a piece of obsidian and then
using a tool to apply pressure to the right spot a regular blade with
parallel edges and constant thickness could be produced. The technique
was still in use by Aztecs in Mesoamerica in historic times, as described
by Spanish friars.22
The complexity of the technique makes it likely that it was passed on in
families. So we may suspect migration along the trail of technology. [29]
The pressure technique emerged around 20,000 BC in Mongolia, northern
China and the Lake Baikal area of Siberia.23 Moving westwards it arrived
in the upper Volga region around 9600 BC.24 Pressure blade-making
spread north of the Black Sea from about 9000 BC.25 The technique also
travelled south of the Caspian into the Near East, where it was embraced
by farmers. Long obsidian blades have been found at Çayönü Tepesi in
layers dating to around 7000 BC.26 Within 500 years the technology had
arrived with farmers on the coast of the Sea of Marmara in western
Anatolia.28 Dairy farming was developing in that region and crossed into
Europe shortly afterwards. Around 4800 BC the innovatory Cucuteni-
Tripolye culture formed north of the Black Sea, with pressure blades and
great herds of cattle.29

Genetics: Mal’ta boy

We now know that the ancestor of the Y-DNA R1 so common in modern European men
lived far from Europe. The Palaeolithic boy from Mal’ta in central Siberia was found to have
carried Y-DNA R*, the ancestor of the whole R lineage. That alone would have made
headline news. The multinational team who tackled Mal’ta boy’s DNA went further,
however, obtaining a full genome. The result was sensational. The boy was related not only
to modern Europeans but also to Native Americans. Perhaps we should not be too
surprised. The Y-DNA haplogroups R and Q are brothers, both descended from P. Some Q
was to enter America across the Beringian land bridge, but it seems that R1 went
westwards instead, to appear in western Russia from around 5600 BC (see box pp. 44–45).
Mal’ta boy was unrelated to modern East Asians, whereas modern Native Americans have
a clear relationship to the latter. So those relatives of Mal’ta boy who moved into Beringia
must have mixed with an East Asian group somewhere along the way. A second genome
was extracted from an adult male at Afontova Gora, a site to the west of Mal’ta and later in
date. It proved similar to that of Mal’ta boy.27 Together these two genomes were used to
represent ancestral north Eurasians (ANE) in an analysis of the source populations of
Europe (see box on p. 73).
29 Pressure blade-making spread east and west from an early centre in Mongolia, northern China
and the region around Lake Baikal in Siberia. Its westward movement took it into the region
around the Caspian and Black seas. Triangles mark findspots of pressure blades.

A second craft innovation also spread west from Lake Baikal to the
European steppe. The earliest pottery was made in the Far East,
thousands of years before farming. The idea was carried westwards
across Siberia by hunter-gatherers. This type of pottery reached the
Samara region in the middle Volga River valley by 7000 BC. It diffused
south and west into the valleys crossing the European steppe.30 Genetic
characteristics associated with Mal’ta boy and Afontova Gora man were
present at Samara along with pottery (see box on p. 73).
 The survival of male forager lineages as the Neolithic engulfed a
territory would depend on the willingness of foragers to adopt farming. So
let us home in on a visible transition from foraging to farming. In the
forest-steppe zone the pottery-making foragers of Dnieper-Donets I
transformed themselves into Dnieper-Donets II cattle farmers around
5000 BC.31 Their ancient mitochondrial DNA is revealing. They carried not
only a typical haplogroup for European foragers, but also some
haplogroups found in European farmers, and, intriguingly, others
normally found in Central Asia.32 A pottery-making wife might have been
sought after when the first pottery arrived in the region from Asia. A
farming wife could have been equally welcome as Dnieper-Donets
foragers encountered cattle-keeping neighbours. Indeed a wife could
have arrived with a cow as dowry. This already mixed culture seems to
be just one of the ingredients of the cultural bombe surprise that is
Yamnaya, which raced across the steppe absorbing previous cultures. It
eventually melded with the remnants of the Cucuteni-Tripolye culture.

The Indo-European lifestyle

There are about 1,500 reconstructed PIE roots and words. This must fall
far short of the full language. Yet the PIE lexicon reveals a great deal
about the lifestyle of its speakers. They were familiar with agriculture and
metallurgy. As we have seen, they coined words for wheels and wagons.
They talked of dairy products, sheep and wool. They had a concept of
social ranking, but few words for specific occupations, or other clues to
urban life. The lexicon reveals a Copper Age society, but not an
urbanized state.33 That is a match for the mobile, pastoralist Yamnaya
(Pit-Grave) culture of the European steppe.34
Yamnaya (3400–2800 BC) is a rich blend of influences. At the time
when Dnieper-Donets foragers were turning into farmers, the most
advanced cultures in Europe were in the Balkans. Farmers had
prospered on the rich, silt soils of the Lower Danube Basin. Hamlets in
what is now Bulgaria, Romania and Serbia grew into solidly built villages
of multi-roomed houses. This was the first region in Europe to take up
copper-working. It was also the first in the world to work gold.35 Before
these Balkan cultures could evolve into civilizations, the sun set on them.
A cold period afflicted Europe from 4200 to 3800 BC.36 Balkan
settlements were abandoned and Balkan metallurgy collapsed. Mines in
Bulgaria ceased production. Some Balkan copper-workers may have fled
westwards into Italy and Sardinia.37

Genetics: three main sources for Europeans

The last few years have seen a great leap forward in studies of ancient DNA. It is now
possible to obtain entire genomes from ancient individuals. From comparison of these to
modern Europeans, a pattern has emerged. The modern European gene pool was formed
from three main source populations: western hunter-gatherers (WHG); early European
farmers (EEF); and ancestral north Eurasians (ANE), who arrived in Europe later than the
farmers.38 Modern European populations are mixtures of the three components, in varying
amounts. [30]
ANE represents a Siberian hunter-gatherer lineage, as represented by Mal’ta boy and the
man from Afontova Gora. ANE admixture has been found in remains from the Samara
district of Russia, between the Volga and the Urals. The earliest examples, from about
9,000 to 6,000 years ago, belong primarily to mitochondrial DNA haplogroups typical of
European hunter-gatherers, but later ones of the Yamnaya culture include those associated
with farmers.39 Likewise, mitochondrial DNA from the Dnieper-Donets II culture and its
Yamnaya successor both show mixed forager and farmer origins.40 Yamnaya people did
not carry 100 per cent ANE, but had enough of it to enable geneticists to trace their
descendants in the Corded Ware and Bell Beaker cultures of the Copper Age.41
 30 Diagram to show the relationship of samples from various nations to the three main
sources of present-day European populations: western hunter-gatherers (WHG); early
European farmers (EEF); and ancestral north Eurasians (ANE).

The Cucuteni-Tripolye culture had emerged between the Carpathian

Mountains and the Middle Dniester around 5200–5000 BC and spread
gradually northeast until it bordered the Dnieper-Donets II culture.42 It
survived the climatic crisis by adapting its economy. The Cucuteni
farmers were already keeping cattle for milk. So they had taken one step
towards what archaeologist Andrew Sherratt called the ‘Secondary
Products Revolution’. Instead of just killing animals for meat, farmers
began to keep them for renewable secondary products, such as milk,
cheese and wool, and for transport and traction. Horses and donkeys
could be ridden or carry a pack; oxen could pull a plough or a wagon.
Thus more could be gained from stock and soil with no increase in
human effort. Any society adopting this new way of life had a marked
advantage in wealth and mobility.43
The earliest evidence of the wheel comes from the Late Cucuteni-
Tripolye culture in the form of wheeled toys. The Cucuteni-Tripolye
villagers by this time had begun to merge with the indigenous people to
their east. Around 3600 BC this culture produced models of sledges
harnessed with oxen.44 By the inventive stroke of adding wheels, it
seems that the sledge became the cart.45 Meanwhile a simple type of
plough called the ard, made of elk antlers, was devised by the Cucuteni-
Tripolye farmers. Traces of yoking and harnessing on steer bones
provide convincing evidence of animal traction.46 Just as oxen had pulled
sledges and the first ploughs, they were the early choice for wheeled
vehicles.47 Horses ran wild on the wide grasslands of the Eurasian
steppes. They were tamed at around the same time that the wagon was
invented, but not to run in harness. The first carts were heavy and slow-
moving. Horses were fleet of foot and ideal for riding. Riders could control
much larger herds of animals, and venture further with them.48
The Maikop culture (4000–3100 BC) of the North Caucasus introduced
to the steppe the tough arsenic-copper alloy that did duty as bronze
before the invention of true bronze (copper alloyed with tin). Maikop
marked the burials of their chiefs with a mound of earth.49 At the same
time, the habit appeared nearby on the steppe. The Repin culture
developed from around 4000 BC between the Volga and Ural rivers. Here
began the characteristic burial type of Yamnaya, with the body laid in a
crouched position under a round tumulus or barrow (kurgan in
Russian).50
Around 3400 BC Yamnaya archaeology begins to be seen. This was a
mobile, wagon-and tent-based herding economy adapted to the
grasslands. Evolving between the Volga and Don rivers, Yamnaya
spread rapidly over the whole European steppe.51 The Yamnaya cultural
package is distinctive. The most visible element today is the kurgan. It
placed a new emphasis on the individual by being a single grave, rather
than a collective grave often re-used. The difference is perhaps more one
of initial intent, since many house secondary burials. We can guess that
those few with rich grave goods, especially wheeled vehicles, were for
honoured leaders of the community. The Kargaly copper ore deposits in
the foothills of the southern Urals were exploited by Yamnaya people.
Burial with tool-kits marked the special status of metalworkers.52
New weapon designs included the tanged dagger and the shaft-hole
axe, which had been introduced by the Maikop metallurgists. The
Yamnaya people wore woven clothes, gold or silver spiral hair rings
(lockenringe), distinctive bone toggles and decorated bone discs.53 The
hair binders are found in pairs with both men and women, and would
have been worn on the end of braids to keep them from unwinding. In the
Iliad we find the Trojan hero Euphorbos with his tresses bound with gold
and silver.54 Cord decoration was common on pottery. The technical
innovations of horse-riding, wheeled transport and metalworking were
gradually adopted across Europe and Asia. Often they are accompanied
by other Yamnaya characteristics, which consolidate the link to the
cultural progenitor.55

Indo-European dispersal
Swings of climate played a large part in triggering movement from the
steppe. The collapse of Balkan farming in the cold period from 4200 BC to
3800 BC left land vacated in the Balkans. Some steppe herders pushed
into the marshes and plains around the mouth of the Danube for winter
fodder and cover. The Cernavoda culture (4200–3500 BC) of the Lower
Danube and eastern Bulgaria may represent the development of Proto-
Anatolian.56 This ancestor of ancient Hittite and Luwian had a PIE-
derived word for thill or harness-pole, but seems not to have had a PIE-
derived word for wagon or wheel, so it is logical to suppose that it left the
parent language community after animal traction was in use for drawing
sledges or ploughs, but before wheeled vehicles appeared. There are
strong similarities in pottery and metal finds between the later stages of
Cernavoda and some sites of northern Anatolia. Coastal seafaring
around the Black Sea would be the obvious explanation. The Balkan-
Anatolian network appears to be continuous and long-lasting.57 So we
may picture a gradual drift across into Anatolia.
A shift in the climate after 3200 BC may have encouraged another
exodus from the European steppe. Conditions became colder and drier.
In the forest-steppe belt the forest was reduced and the steppe
expanded. The region was at its most arid between 2700 and 2000 BC.58
The Yamnaya culture and its descendants spread far and wide along
multiple routes.
 The first Indo-European move east had all the boldness that would
come to characterize the steppe nomads. A group set out from the Volga-
Ural region to trek some 2,000 km (1,240 miles) to the high steppe of the
Altai Mountains c. 3300–3000 BC. There they created the first mobile
pastoralist culture east of the Ural Mountains.59 An Indo-European
language with archaic features crops up millennia later along the Silk
Road, named Tocharian after the people known to the Greeks as
Tokharoi.60
The next movement visible in the archaeology flowed to the western
end of the steppes, integrating Yamnaya herders and Late Cucuteni-
Tripolye farming communities into the Usatovo culture around the mouth
of the River Dniester. This culture could be the first step along the road
leading to the Pre-Germanic dialect splitting away. The next step was
migration up the Dniester, which blended with the descendants of Balkan
farmers to create the widespread Corded Ware or Single Grave culture
(2750–2400 BC).61 Ancient DNA samples suggest that people of the
Corded Ware culture had on average three grandparents descended
from Yamnaya ancestors.62
Proto-Germanic (the immediate ancestor to the Germanic language
family) did not develop until about 500 BC. It has an interesting feature.
Up to one-third of its lexicon is non-Indo-European. As Sir William Jones
said in the quotation that heads this chapter, Germanic seems ‘blended
with a very different idiom’. Many of these non-Indo-European words are
agricultural, and so must have been borrowed from a farming culture.
Furthermore, traces of this same language can be found in the Greek,
Latin and Celtic languages.63 The common feature seems to be contact
with Balkan farmers or their descendants.
We can certainly imagine some such contact within the massive
migration up the Danube between about 3100 and 2800 BC. Yamnaya
herders passed through the Usatovo culture into the Danube Valley,
ending up in what is now eastern Hungary. The evidence lies in their
kurgans. This was a true folk movement that left ten thousand burials.64
In fact it is best considered as an extension of Yamnaya. In that
Danubian arm of the expanding culture we can visualize the gradual
development from Old European (Alteuropäisch) to Proto-Italo-Celtic.
31 Yamnaya movement up the Danube into the Carpathian Basin is generally taken to represent
the spread of Proto-Italo-Celtic. Yamnaya movement towards Thrace could represent the
ancestor of the ‘Balkan group’ of languages: Thracian, Greek, Armenian and Phrygian.

Yamnaya settlements west of the Black Sea [31] suggest movement

towards Thrace, where the ancestor of the Balkan group of languages
(including Greek and Armenian) could have developed. Long before
appearing in the south Caucasus around 600 BC, Armenian seems to
have had a developmental period close to the ancestor of Greek.65 At
Plachidol in Bulgaria a particularly interesting Yamnaya-type cemetery of
six tumuli was excavated in 1979. Two anthropomorphic stelae were
found there. One female grave contained two solid wooden wheels.66
From Plachidol and Ezerovo, near Varna, there is a trail of
anthropomorphic stelae to Greek Macedonia and on to the Aegean island
of Thassos and Soufli Magoula in Thessaly.67 [see 33]
Steppe groups penetrated Late Cucuteni-Tripolye towns on the Middle
Dnieper, together with elements of Corded Ware, creating a hybrid that
gradually became its own distinct culture. This seems to represent the
dialect which became Proto-Balto-Slavic.68
Towards the end of the most arid period on the steppe, a final
expansion east of the Ural mountains apparently set the Indo-Iranian
languages on their way. Here the first fortified settlements appeared on
the Asian steppe, such as Sintashta and Arkaim (2100–1800 BC). The
earliest evidence of chariots has been unearthed at Sintashta. These
light vehicles with spoked wheels were in demand by the princes of the
Near East by 2000–1900 BC.69 The culture of Sintashta merged into the
widespread Andronovo culture.
Andronovo gave birth to languages spoken today in India and Iran, but
let us focus on the fate of those who remained on the steppe to emerge
in history as Scythians, since several medieval pseudo-histories that we
encountered in Chapter 1 claim them as the ancestors of the Insular
Celts. By the 8th century BC the whole Eurasian steppe was inhabited by
horse-riding nomads known to the Greeks as Scythioi. They were not a
nation, but rather numerous tribes, all speaking dialects of an East
Iranian language.70 So the Scythians were not the ancestors of the Celts.
They spoke languages from an entirely different branch of the Indo-
European family. Furthermore, the predominant Y-DNA signature found
in ancient Scythians was R1a,71 while that of Celtic-speakers is R1b (see
box pp. 26–27). In the next chapter we follow a trail of clues from the
steppe to the homelands of the Celts.

Overview
• Proto-Indo-European (PIE) was the parent language of a large
family of languages spoken today in both Europe and parts of Asia,
including the Celtic branch.
• In prehistory any language had to be spoken by a group of people
face to face. As groups of PIE-speakers moved away from the
parent group, they developed daughter languages.
• A theoretical PIE homeland in Anatolia has been preferred by those
who see the Neolithic as the most likely time for a new language
family to spread. Yet PIE appears to be a later, Copper Age
 language, which did not develop in contact with any language of
Anatolia.
• The theory of a PIE homeland in the Pontic-Caspian steppe is
supported by archaeological evidence of Copper Age movements
from this region to areas where speakers of Indo-European
languages emerge into history.

• Genetic evidence supports the Copper Age steppe homeland of PIE:

• Y-DNA haplogroup R1 predominates in Europe today. It arrived
on the eastern fringes of Europe with hunter-gatherers, but spread
more widely in the Copper Age.

• Present-day Europeans are mixtures from three source

populations: ancient European hunters; farmers from the Near
East; and ancient Siberian hunters arriving later in Europe,
associated with Y-DNA haplogroup R.
CHAPTER FIVE

Stelae to Bell Beaker

The Nasamones consult private oracles by frequent and

lengthened visits to the sepulchres of their relatives … and the
Celts, for the same purpose, stay away all night at the tombs of
their brave chieftains.1

This description of the Celts communing with their ancestral chiefs comes
from the 2nd century BC. Naturally, the Greeks had a word for it. In the
Greek world a herõon was a shrine dedicated to the cult of a hero. Celtic
sanctuaries that seem designed for just such a purpose are known from
the Iron Age (see pp. 110–11). In this chapter we delve to the roots of the
tradition.
Keen-eyed readers may have picked up a small clue to the Yamnaya
ancestry of Bell Beaker. Hair binders in precious metals are found on the
steppe (p. 75) and in Bell Beaker graves (p. 53). The characteristic
Yamnaya type are spirals, and similar ones are found in the Bell Beaker
of Portugal.2 How did Yamnaya elements travel right across Europe?
Anthropomorphic stelae provide the missing link. Stylized human images
can be found all the way from the European steppe to the Atlantic coast
of Iberia.3 These stelae can be linked to confident and experienced
copper-working of the Yamnaya type. A new way of life was transported
to the far west of Europe, which developed into the first stage of the Bell
Beaker culture.4

People of stone
Some Yamnaya burials were individualized by a stone stela. The
beginning of the tradition can dimly be seen at the western end of the
European steppe. A culture named after the site of Mikhailovka on the
Lower Dnieper seems to be a cross between farming and steppe
influences in its first phase (3700–3400 BC). Its people adopted the
steppe habit of burial mounds. The graves under the kurgans were often
stone-lined cists. In a few, the covering slab was roughly shaped at the
top to indicate a head and shoulders. This seems to be the tentative start
of what became a long steppe tradition of anthropomorphic stelae.
Mikhailovka I kurgans with stelae were scattered as far west as the
Danube delta and as far south as Crimea. From there the fashion spread
to the central Caucasus. The custom of creating stelae flowered in the
western steppe, where successor cultures to Mikhailovka I gradually
came under the influence of Yamnaya.5
Such stelae could be carved on both back and front, and in
exceptionally massive cases, on all four sides [32], and were unworked at
the base, so it is clear that they were intended to stand upright. Beneath
one kurgan of the Mikhailovka group the fractured remains of an
anthropomorphic stone stela were found together with traces of ochre,
potsherds and animal bones. It would seem that sacrifices were made at
the stela.6
32 A stela 1.2 m (almost 4 ft) tall found at Kernosovka, Ukraine, is carved on all four sides. On the
front is a face with a drooping moustache. The arms are indicated together with an array of
weapons. A belt is marked with incised lines around all four sides. On the rear (not shown) a pair
of footprints cross the belt.

This does not necessarily mean that the ancestors were treated as
gods. A monument from a literate era provides the clue to purpose. An
8th-century BC stela from Zincirli in southeastern Turkey depicts a royal
official named Kuttamuwa at his funerary banquet. Kuttamuwa himself
tells us so. The text on the stela explains that Kuttamuwa commissioned
it during his lifetime, and that at its inauguration in the mortuary chapel
offerings were made to various gods. The most enlightening line explains
that one of the offerings was ‘a ram for my soul that will be in this stela’.
Ancient people could visualize their soul being transferred to a memorial
stone after death.7 Human flesh is mortal, but stone may stand eternal as
a symbol of the departed.
 Hundreds of Yamnaya stelae have been recovered. No two are exactly
alike. The overwhelming majority are simple slabs, shaped slightly to hint
at a head and rounded shoulders, with little or nothing in the way of
ornament, though many have a belt incised or painted on. Another
feature that can appear is a pair of footprints. Both belt and footprints are
common on the more interesting rarities, sometimes called statue-
menhirs. These not only indicate arms and facial features, but also
objects, which have been eagerly examined for clues to the status of the
figure, such as weapons or a shepherd’s crook.8 That footprints are so
prevalent suggests they have some importance. It has been argued that
they are one of the symbols of a divine guide of souls, a psychopomp,
who escorts the newly deceased to the afterlife.9

The stelae trail

Funeral stelae radiate out from the European steppe in several
directions. We have already noted the trail through Bulgaria to Greece
(pp. 77–78). Here we focus on the route that links Yamnaya to Bell
Beaker. [33] In Romania anthropomorphic stelae have been found at
Hamangia-Baia [34] and Ceamurlia de Jos, both sites on the coastal
plain leading to the Danube delta.10 Migration up the Danube brought the
concept of stelae into the Carpathian Basin. Within Romania a site at
Baia de Cris had two – a broken one with decorated girdle and necklace
was discovered there in 1881, and a similar one in 2000.11
A connection with metals can be seen in the stelae trail westwards
along the Mediterranean. The first attraction was the Alps. The rich
copper deposits of the Alps had been discovered by metalworkers c.
4500–4000 BC. Experiments were made in smelting the local ore at
Brixlegg above the Middle Inn Valley in the Austrian Tyrol.12 There are
similar dates for a fragmentary crucible and copper slag at Botteghino
(Parma), northern Italy, on a site otherwise comparable to Neolithic sites
in France.13 One can imagine a visiting metalworker or two at that stage.
At first surface ores were collected. Mining began once such easily
available sources ran out. The earliest known copper mines in western
Europe (c. 3500 BC) are at Monte Loreto (Castiglione Chiavarese,
Liguria) in northern Italy.14 Tuscany too has deposits of copper ore.
33 From a starting point north of the Black Sea, Yamnaya anthropomorphic stelae radiated in
several directions. One route led to Iberia and on to what is now Brittany.
34 This Copper Age anthropomorphic stela from Hamangia-Baia, Romania, has a belt with a
sporran-like purse hanging from it in the centre and long pockets at each side. On the rear are
carved five axes and two footprints.

Copper Age cultures sprang up in Italy: Remedello in the north,

Rinaldone in the west and Gaudo in the south.15 The dating of these
cultures has been an academic battleground. That is unfortunate, since
the chronology is crucial to our understanding of how metallurgy spread,
and who spread it. We can attribute the discovery of the Brixlegg ores to
Balkan copper-workers turning their sights westwards. It dates to before
the collapse of the rich Balkan towns of the 5th millennium BC. In Sardinia
the arrival of metallurgy around 4000 BC suggests a flight direct from that
collapse.16 The most recent dates for Rinaldone burials centre around
3500 BC, too late for the Balkan exodus, yet too early for Yamnaya
wanderings.17 Knowledge of metallurgy could have spread down from the
Alps into central Italy. So it had been suggested that Remedello was
probably at least as old as Rinaldone.18 Radiocarbon dating proves
otherwise. The dates for Remedello graves centre around 2900 BC. They
have been divided into phases: Remedello I (3300–2900 BC) and
Remedello II (2900/2800–2400 BC).19 That is compatible with arrivals
from the Yamnaya stream, which would explain the Yamnaya elements in
the Remedello culture, such as single graves and copper-arsenic alloys.
The distinctive Remedello II daggers are depicted on anthropomorphic
stelae of northern Italy and the western Alps.20 [35]
So copper sources in the eastern Alps were already known in the days
when the Balkans were the centre of European metallurgy, and continued
to attract attention long afterwards. A group of copper prospectors seems
to have left the Carpathian Basin to wend its way into northern Italy. The
easiest route would be to follow the River Sava, a tributary of the
Danube, to the area of present-day Ljubljana, capital of Slovenia, and
then wind around the southeastern Alps to the Adriatic coast. That would
mean passing through the Ljubljana Marshes. In the 4th millennium BC
copper-workers settled in pile dwellings in these marshes. Their copper
came from the Alps. A fragment of ray bone found amid their detritus
shows contacts also with the Adriatic coast 100 km (62 miles) away. The
oxygen-free bog around the pile dwellings preserved an ancient solid
wooden wheel in amazingly good condition, and nearby a wooden axle,
radiocarbon-dated to between 3160 and 3100 BC. The square-cut axle
would have rotated with the wheel. Early wheels of the same revolving-
axle design have been found only in Switzerland and southwest
Germany. It was presumably intended for a two-wheeled cart, suitable for
the rolling foothills of the Alps.21 So we have a trail of technology that
leads to the Alps, where a profusion of stelae is found. Copper deposits
in Sardinia were discovered much earlier, but also attracted stelae
makers.
35 Stela no. 2 from Petit-Chasseur, Sion, Switzerland, has a double spiral pendant and a dagger
of the Remedello II type, with a triangular blade and half-moon pommel.

The search for copper had spread westwards along the Mediterranean
by the end of the 4th millennium BC, taking stelae with it. The earliest
copper mine and metallurgical complex in France is in the mountains of
Languedoc at Cabrières and Péret, dating from the late 3rd millennium
BC.22 Remedello daggers are depicted in rock art of both northern Italy
and southern France. The daggers appear in engravings in the French
Alpine foothills at Chastel-Arnaud, Drôme, amid the interconnecting
valleys between the south Alps and the Rhône Valley, a clue to the route
westwards of copper metallurgy.23 There is a concentration of Copper
Age anthropomorphic stelae in the south of France. [36]

36 ‘La Dame de Saint Sernin’ is one of a number of stelae from the south of France. The lines on
the face of this female image are thought to represent scarifications. Legs are shown here and on
several other stelae from this region.
37 Key sites in the early Copper Age of Iberia. Zambujal and Los Millares had the sea and
sources of gold close at hand. Copper was mined inland at Mocissos and Cabezo Juré.

Another rich Copper Age culture appeared in Iberia. [37] The earliest
dates of copper-working there (c. 3100 BC) are for mining-metallurgical
complexes in southwestern Iberia, such as Cabezo Juré. It is revealing
that this site was colonized by a community already specialized in copper
production. These incomers lived within a fortified centre, dining well and
importing luxuries, while in a village outside lived the lower-status
workers. The well-protected elite controlled access to horses, used
probably in the transport of copper ore.24 At this time Iberia had wild
horses. Some of their DNA made its way into modern Iberian breeds.25
Horse bones are found together with Bell Beaker pottery throughout its
range, so the idea that domesticated horses spread from Iberia with Bell
Beaker has enjoyed a certain popularity,26 but sites such as Cabezo
Juré, which precede Bell Beaker, suggest that the knowledge of horse-
taming and copper-working arrived in Iberia together from the European
steppe.
The two foci for Copper Age Iberia became the lofty, fortified
settlements of Zambujal (Torres Vedras, Portugal) and Los Millares
(Almería, Spain). Both were set on promontories commanding
approaches by river or sea. Both began as small strongholds and
expanded with the creation of new walls enclosing greater areas.27
Zambujal had the most easily defended position. It was set on a
peninsula carved out by the great River Tagus where it meets the sea.
On the same peninsula other fortified sites include Vila Nova de São
Pedro (Azambuja) and Leceia (Oeiras). The fort at Leceia was built
around 2900–2800 BC. Like Zambujal, pottery of different styles was in
use there at different periods, ending with Bell Beaker.28 At these sites
ease of defence was combined with ease of access by sea. Today
Zambujal lies 14 km (almost 9 miles) from the Atlantic coast, but in the
Copper Age the nearby River Sizandro was a tidal estuary. Zambujal
imported ivory from afar. Alluvial gold from the Tagus could have been
traded in exchange, along with copper ware. Copper was imported from
the mining-metallurgical complex at Mocissos for the workshops of
Zambujal. As in the Remedello culture of northern Italy, copper-arsenic
alloys were deliberately used for daggers, saws and other artifacts that
required a harder metal than pure copper.29
38 This stela from Tapada, Guarda, Portugal, can be compared with ‘La Dame de Saint Sernin’
(p. 86) in the position of the arms and the depiction of a belt and complex necklace. It is one of a
group in Iberia with a headdress.

Iberia has a rich variety of decorated stelae, ranging in date from the
Neolithic to the Iron Age. Here we look only at the anthropomorphic type.
[38] These are generally found in a prominent position within megalithic
burial mounds. In Iberia some Neolithic burial chambers were re-used in
the Bell Beaker period. Dating is easier where a communal burial
chamber was both built and closed in the 3rd millennium. Sometimes the
stelae themselves provide a clue to chronology. Obviously those
depicting metal objects must be later than the Neolithic. For example,
one from Longroiva in northern Portugal carries an image of a halberd of
Copper Age type, together with a bow and dagger. Most are much less
informative. As with Yamnaya stelae, many barely hint at a human figure,
with either no objects depicted or only a shepherd’s crook.30
Anthropomorphic stelae continue up the Atlantic coast from Iberia to
Brittany, where there is a notable cluster, and even as far as the Channel
Isles. [see 33] Crucially, Copper Age anthropomorphic stelae along the
Mediterranean and Atlantic route herald the earliest Bell Beaker pottery,
that which falls into the southern sphere. [see 19]

Linking stelae to Bell Beaker

The fascinating complex at Petit-Chasseur, Sion (Valais), in western
Switzerland, has been much studied. A necropolis there continued in use
from the Late Neolithic through the Bell Beaker period to the Early
Bronze Age. Swiss archaeologist Alain Gallay spent years on its
excavation and identified 28 decorated stelae, which he sorted into two
types: early and Bell Beaker period.31 So a link was made here between
stelae and Bell Beaker. The wider meaning was explored when Richard
Harrison and Volker Heyd reassessed this important site, together with a
remarkably similar one at Aosta in the Alps to the south of Sion.32
39 Anthropomorphic stela no. 25 from the necropolis at Petit-Chasseur, Sion, Switzerland,
belongs to the Bell Beaker period and is decorated with patterns reminiscent of Bell Beaker
pottery.

Aosta was much better preserved, with most stelae found still in situ,
so comparison between the two sites made it easier to reconstruct the
original appearance of Sion. Around 2700 BC a necropolis was laid out on
a solar orientation. Two communal burial chambers were built on this
axis. The first anthropomorphic stelae stood before them in a line facing
southeast, creating a parade of sun-blessed ancestors. At both Sion and
Aosta, stelae with Bell Beaker motifs were added to the parade,
indicating continuity in these communities from the first arrivals through to
Bell Beaker. Then came a dramatic change, still within the Bell Beaker
period. All the stelae at Sion were broken and then re-used, when
needed, to create individual cist burials.33 The political implications of this
upheaval will be considered later (p. 97).
 The early stelae at Sion are therefore much damaged, but enough
remains to pick out features and reconstruct the whole. No. 2 has the belt
and stick-like arms we have seen on Yamnaya stelae, together with a
double spiral pendant and a dagger of the Remedello II type. [see 35]
Three other early stelae depict Remedello II daggers.34 So the first
creators of the necropolis were connected to the Remedello phase of
copper-working (see p. 84).
Then the new style of stelae decoration, linked to the Bell Beaker
culture, appeared. As we saw earlier, the classic weapon of Yamnaya
stelae was the axe. It is not unknown for the bow and arrow to be
depicted, but almost invariably together with an axe.35 With the Bell
Beaker culture came a new emphasis on the bow in grave goods, which
is reflected in the stelae of this period. A symbolic bow and arrow
appears on four stelae.36
The detail of the clothing shown is extraordinary. [39] These stylized
bodies wear highly patterned garments and decorated belts. The
geometric patterns are reminiscent of those on Beaker pots. A few stelae
have small sporran-like purses hanging from the centre of their belts,
together with longer pockets at either side.37 The greater detail of these
depictions enables us to make sense of the belt features of the Yamnaya
statue from Hamangia-Baia. [see 34]

The origins of Bell Beaker pottery

The Bell Beaker culture was identified over a century ago. There has
been lively debate ever since over its origins, with homelands suggested
in Iberia, southeastern France, Sicily, Central Europe and the Low
Countries. Since the key object that appears over the whole Bell Beaker
range is the distinctive inverted-bell pot, there was a strong focus on
ceramics in these searches for an origin point. Several different styles of
Bell Beaker have been defined by the type of ornamentation. Most are
local or regional in distribution. The two international types have attracted
most attention. One widespread type, known as All Over Corded (AOC),
is entirely decorated with impressions made with cord. The Corded Ware
culture in central and eastern Europe (see p. 76) did not have a
monopoly on the simple and effective technique of winding a twisted cord
around a wet clay pot. The other type is labelled Maritime, since it is
found predominantly along the Atlantic seaboard from Portugal to Brittany
via Galicia. In the 1970s Dutch archaeologists presented a case in favour
of the Netherlands as the home of Bell Beaker. The ‘Dutch Model’ was
the most widely accepted for the rest of the century. In the lowlands
around the mouth of the Rhine both AOC and Maritime Bell Beakers
have been unearthed. This region is on the western edge of the
distribution of the Corded Ware culture and a type of Corded Ware
pottery is found there, with similarities to AOC, called the Protruding Foot
Beaker (PFB). A sequence was proposed from PFB through AOC to
Maritime.38
Radiocarbon dates put paid to this idea. Carbon-14 has dated the
earliest Bell Beaker pottery to the 2700s BC and the earliest Corded Ware
to c. 2750 BC. These two widespread cultures were contemporary.
Moreover the earliest dates for Bell Beaker come from southern
Europe.39 Bell Beaker did indeed follow Corded Ware in central and
eastern Europe, but these were different communities.40 Their similarities
reflect a shared cultural parent in Yamnaya: cord-impressed pottery was
part of the Yamnaya assemblage.
So it seems that copper-working descendants of Yamnaya developed
bell-shaped pottery in southern Europe. Though Bell Beaker pottery was
generally made in the home (p. 51), there were some specialist potters.
Expertise can be detected by regularity of form and decoration. French
archaeologist Laure Salanova perceives the Maritime Bell Beaker as the
classic form, made to a strict standard, identical wherever it is found. [40]
The vessel has a uniform S-shaped profile, and is tall with a flat base. It
is decorated with neat horizontal bands from top to bottom, edged by
incised lines or cord impressions, and filled with simple patterns,
commonly of hatched lines, incised using a shell or comb. Maritime was
the model that inspired a variety of styles that spread across Europe. For
example the banded decoration of the Ciempozuelos style of Spain [see
20] is more irregular than the classic Maritime style, but is clearly heir to
it. So Salanova argues that we should look west for the source of Bell
Beaker pottery, especially to the Tagus estuary in Portugal, which has
the highest density of these classical vases.41 At one time it was thought
that Iberia had no AOC pottery, which seemed a point against it as the
Bell Beaker homeland. Now corded Bell Beaker is known from four
Copper Age fortified sites in Portugal and a scattering of sites in Spain,
though it remains rare in Iberia in comparison with Maritime.42

40 A Maritime type of Bell Beaker from Zambujal, Portugal.

The people who made the earliest Bell Beakers were copper-workers
and could only be a few generations away from immigrants. At Zambujal
there is a clear continuity from the earliest copper-workers to the
beginnings of Bell Beaker.43 A problem lies in a lack of obvious local
antecedents. Several authors have argued that local Copper Age copos
(cups) with a gently waisted outline and horizontal grooves as decoration
developed into Bell Beaker.44 This fails to convince. It is more likely that
the Bell Beaker style was a synthesis of influences from more distant
connections, as Jan Turek has proposed. He puts forward certain
curvaceous pots with impressed decoration from Late Neolithic Morocco
as a possible inspiration.45 Yet key ingredients of the Bell Beaker design
have precise predecessors on the stelae route from Ukraine to the
Carpathian Basin. Inverted bell-shaped pots were made before 4000 BC
north of the Black Sea.46 [41] Cord impressions have a long history in the
same region. Bell Beaker ware commonly had its decoration picked out
with white paste made of crushed bone.47 [see 20] This technique was
used earlier in the Carpathian Basin.48
41 A curvaceous pot of the Cucuteni culture, from Draguseni, Romania. Its decoration is very
different from that of Bell Beaker, but was the shape an inspiration for the later pottery?

So the influences that culminated in this pottery could have travelled

over time along the same route that brought copper-working to Iberia. A
common pattern of migration is repeated movements along the same
route. We can imagine pioneers scouting out metal sources and then
returning home to trade or collect family members. If so they would
spread the knowledge of opportunities, which could attract more
migrants. At Leceia in Portugal we see new arrivals expanding the
settlement. Nestling just outside the fortifications were two huts with
radiocarbon dates centring in the 2700s BC in which the pottery was
exclusively Bell Beaker, while within the walls earlier local pottery
gradually mixed with Maritime Bell Beaker material, after the new styles
were introduced.49
Though huge efforts have gone into tracing the origin of Bell Beaker
pottery, there is much more to the culture. The German prehistorian
Edward Sangmeister deserves credit for his insight into the Bell Beaker
phenomenon. He worked at the German Archaeological Institute in
Madrid in the 1950s and excavated Zambujal from 1964 to 1973.
Recognizing the differences between the early Bell Beaker of Zambujal
and the later Bell Beaker of more easterly Iberia, he argued that the
pottery spread from Iberia into Central Europe and then returned to Iberia
together with new cultural influences. His ‘reflux’ model is followed below
in spirit, if not in all its details.
Bell Beaker routes and the development of Celtic
It has long been supposed that Proto-Celtic developed around the heads
of the Danube and Rhine. That fitted the identification of the Hallstatt and
La Tène cultures as Celtic. Now that attention is turning to the Bell
Beaker culture as the vector for Celtic, Celtic specialists John Koch and
Barry Cunliffe have teamed up to propose that Celtic spread from
Iberia.50 The evidence in Iberia of both Celtic and related Indo-European
languages makes this a tempting thesis, but there are insuperable
difficulties. Proto-Celtic developed in contact with an early precursor to
Germanic.51 That presents no problems if we site Proto-Celtic north of
the Alps, but does not fit an Iberian homeland. Later contact between
Proto-Celtic and Iranian, discussed in the next chapter (p. 104), can also
be explained by the traditional homeland, but not an Iberian alternative.
[42]
Koch has put forward two linguistic contact arguments as a counter-
weight. He suggests that the key change which defines early Celtic, the
weakening of the Indo-European ‘p’ sound (see p. 62), could have
occurred in contact with p-less Iberian. It is a common assumption that
Iberian was the pre-Indo-European language of Iberia, but it appears
actually to be a relatively late arrival, intruding into the band of Ligurian
along the Mediterranean coast.52 This seems too late and limited a
contact to account for Celtic.
42 Routes proposed in the present work for Old European (Alteuropäisch) and its descendants in
the Copper and Bronze Ages. Old European, the oldest linguistic layer of Indo-European origin,
was overlaid by various languages developing from Italo-Celtic.

Koch’s other suggestion suffers similarly from a chronological problem,

but also a geographical one. Some linguists argue for an Afro-Asiatic
influence, specifically Berber, Coptic or Semitic, on the Celtic languages
of the British Isles. Its most obvious expression is seen as a change in
the standard word order from the original Indo-European subject-object-
verb (SOV) and the mixture of SVO and SOV which has been
reconstructed for continental Celtic, to the rare VSO order.53 The
Phoenicians spoke an Afro-Asiatic language of the Semitic branch and
were present in Iberia by the 8th century BC, so for Koch they could
represent the supposed link, but if this affected the development of Proto-
Celtic, VSO should appear in all the Celtic languages, not just those in
the British Isles. It is logical to suppose that VSO in the British Isles arose
from contact with a language present in the islands when Celtic speakers
arrived. This need not be any member of the Afro-Asiatic family, but
simply a now lost language of Neolithic farmers among the many in
Europe that we can guess disappeared under the wave of Indo-
European.54 The farming languages Berber and Coptic survived because
Indo-European languages did not significantly penetrate North Africa until
colonial times.
Nevertheless, Koch and Cunliffe have usefully drawn attention to
Iberia, a region often neglected in Celtic studies. Any model of the spread
of Celtic needs to explain the Iberian linguistic mosaic. The proposal here
is that the language which spread from Italy to Iberia with the Stelae
People was an Indo-European dialect too early to be Celtic, and that
Proto-Celtic developed where traditionally supposed, around the heads of
the Danube and the Rhine, beginning during the Bell Beaker period. Thus
the Late Bell Beaker movements described below could have carried
early forms of Celtic into northeastern Iberia, the British Isles and the
southern Alps.
The Bell Beaker communities of Iberia were in crisis from c. 2500/2400
BC. Violence and disorder may have encouraged migration. Some large
Copper Age settlements such as Zambujal were abandoned. Maritime
Bell Beaker travelled eastwards to southern France c. 2500 BC and also
appears on the southern coast of Brittany.55 The distinctive copper
spearheads named Palmela points which seem to originate with Bell
Beaker in Portugal are also found in Brittany and southern France.56 [43]
Bell Beaker also arrived at Csepel Island in the Danube around 2500
BC. Among the finds here was a gold disc embellished with concentric
circles, similar to the gold discs found on western Bell Beaker sites and
interpreted as solar symbols.57 Csepel Island has given its name to sites
of the Bell Beaker Csepel group, clustered around Budapest. One
attraction of the Danube Basin was access to steppe horses. Indeed the
dominance of horse bones on the sites of the Csepel group suggests that
they were horse-breeders.58 Hungary has no other Bell Beaker.
Anthropologically and culturally the isolated Csepel group appears an
intrusion. Isotope tests showed that Bell Beaker people on Csepel Island
were migrants.59 A study of inherited tooth characteristics links the Bell
Beaker folk of Csepel to those of western Switzerland, while the latter in
their turn cluster with the southern Bell Beaker group in Iberia and
southern France.60 So it seems that some descendants of the Stelae
People returned to their ancestral home in the Carpathian Basin.

43 Palmela points were so named because they were first found at Palmela in Portugal. These
are from a Bell Beaker burial at San Román de Hornija (Valladolid, Spain) dated to 1800 BC.

The Bell Beaker culture took on a different configuration in the eastern

Beaker sphere. Boar’s tusks were worn as pendants or garment
fasteners, a fashion which can be traced back to the steppe.61 This
suggests that the incomers mixed with distant relatives who had
remained in the Carpathian Basin. Begleitkeramik (accompanying
pottery) is typical of the eastern Bell Beaker group; it reflects local pottery
styles. The handled pitcher and pedestalled and polypod (multi-footed)
cups or bowls appear in pre-Beaker groups in Hungary and Slovakia,
were absorbed into eastern Bell Beaker and then spread out of the
Carpathian Basin with it. This helps us to track the influence of the
eastern Beaker tradition west into northern Italy and southern France,
south to Sardinia and north down the Rhine.62 Around the mouth of the
Rhine, the Rhenish style of Bell Beaker decoration developed, which
spread into Britain.63 Lastly, islands of Bell Beaker appear across a
northern sphere including Jutland, northern Germany and Poland.64 [see
19]
Some of the eastern group entered territory that had previously
belonged to the southern Bell Beaker sphere. There was an abrupt
change at the Alpine sites of Sion and Aosta around 2425 BC. The former
Bell Beaker stelae were smashed. Objects distinctive of the eastern Bell
Beaker group appear, such as boar’s tusks.65 Isotopes reveal a distant
origin for one man, who had the type of cranium typical of the eastern
Bell Beaker group.66 The evidence adds up to a power shift in the middle
of the Bell Beaker period from the mouth of the Tagus to the head of the
Rhine. The Bell Beaker communities of the Rhône corridor who had
previously looked to Iberia shifted their gaze eastwards and northwards.
Bow-shaped pendants appear, along with pottery designs from the
eastern Bell Beaker group.67
An echo of eastern Bell Beaker even reached Iberia. Around 2200 BC a
new phase of Bell Beaker began in central Spain, the Ciempozuelos
horizon. The predominant pottery is influenced by Maritime Bell Beaker,
yet from one site came a type of pedestalled cup or bowl derived from
those in the Carpathian Basin.68 An ancient language of the Iberian
meseta may reflect an influx at this time of speakers of the developing
Proto-Celtic. Celtiberian is the most archaic form of Celtic.69 If indeed it
arrived c. 2200 BC, then it becomes plausible that Proto-Celtic developed
during the Bell Beaker period. If the very earliest Bell Beaker arrivals in
the British Isles spoke a pre-Celtic Indo-European language, later ones
arriving down the Rhine would bring the earliest Celtic.
It is intriguing that around 2400 BC anthropomorphic stelae were
sometimes used to close up megalithic monuments in Iberia and Brittany.
This is just the period when Bell Beaker makers first entered Ireland.70
Did migrating families say farewell to their ancestors in a ceremony of
closure? There are clues that the first Bell Beaker makers in Ireland
arrived from Brittany or Portugal. An earring or pendant found at Benraw
may be an early import, for it is not made of Irish gold, and it is very
similar to a pair of earrings found at Estremoz in Portugal.71 Wrist-guards
or bracers for archers are part of the Bell Beaker assemblage. The only
type found among the southern Bell Beakers is narrow with two holes,
one at each end. Broader, four-holed types predominate in Central
Europe. Ireland has almost exclusively two-holed types.72
According to the model proposed here, that would mean that the first
Bell Beaker arrivals in the British Isles spoke some form of Old European
or Alteuropäisch. Certain river-names have been taken to support a very
old stage of Western Indo-European in Ireland and Britain (see p. 61).73
Then from about 2200 BC the earliest Celtic would start to appear with
new arrivals drawn to Cornish tin and perhaps Irish gold.
It has been argued that the Celtic languages of the British Isles and
Gaul are too similar to have diverged this early.74 This case depends on
them actually diverging at this point, never to meet again. The
archaeological record suggests continuing contact, with recurrent bursts
of migration right up to the Roman period, giving ample opportunity for
new linguistic developments to pass from Gaul into the Isles. For
example iron-working technology arriving in the British Isles from its
Celtic Continental centres was no doubt accompanied by the appropriate
vocabulary. That would help to explain the complexity of the relationships
between the recorded Celtic languages. [see 54]

Genetics: R1b flows into the British Isles

If the pattern of Y-DNA haplogroups seen in Continental Europe (box pp. 44–45) is matched
in the British Isles, the haplogroup G2a probably predominated in the first farmers, while the
R1b that dominates Europe today arrived in the Copper Age. G2a is rare today, but there is
no need to picture genocide. The first farmers in the British Isles thrived initially, but then
encountered problems. Populations seem to have fallen before they were boosted by Bell
Beaker arrivals.75 The latter brought new technologies, giving them an economic
advantage, which could ensure the better survival of offspring.
Today two subclades of R1b1a2a1a2 (P312) seem to echo the two Bell Beaker routes
into the Isles that we see in the archaeology, though this can only be speculation in the
absence of early Bell Beaker DNA from the British Isles. The predominant one is
R1b1a2a1a2c (L21) [see 9], which probably moved up the Rhine, across the Channel and
from Britain to Ireland.
R1b1a2a1a2a (DF27) is common in Iberia.76 So the rare cases of R1b-DF27* (the basal
form of DF27) in Ireland today may be a remnant of Bell Beaker movements up the Atlantic.
If R1b-DF27 also travelled with Bell Beaker from Iberia into the Carpathian Basin, that
would explain its present wide range. It has some subclades whose bearers cluster in
southwestern Europe, but others which are almost exclusive to northern Europeans and
their descendants. Scandinavian branches of R1b-DF27 could have arrived in the Isles as
late as Viking or Norman times.

Bronze Age mobility

Language contacts between wide-spaced Celtic-speakers would have
continued throughout the Bronze Age, an era remarkable for its long-
distance trade networks across Europe. However, the traffic along the
Atlantic littoral petered out, becoming patchy and light c. 2100 to 1300 BC.
In this period Iberia developed largely in isolation from lands to the north.
Britain and Ireland had more contact with northwestern France and the
Low Countries.77
Then came the Atlantic Bronze Age (c. 1300–700 BC), which saw
prestigious items exchanged via the Atlantic seaways. The major centres
were southern England and Ireland, northwestern France and
northwestern Iberia.78 This was precisely the period in which
northwestern Europe suffered an increasingly wet and cold climate.79
Relocation to the sunnier south would offer attractions. This might help to
account for the Celtification of northwest Iberia.
The Castro (castle) culture of this region has its origins in the Late
Bronze Age and continued into the Iron Age. Rather than starting in the
far north of the peninsula, as we might expect if it was the product of
Atlantic arrivals, it seems to begin in northwest Portugal. Yet it absorbed
influences from both the Atlantic and the Mediterranean.80 In the Late
Bronze Age, cremation became the standard treatment for the dead over
a large stretch of Central Europe. The ashes were buried in urns, hence
the name Urnfield culture (c. 1300–750 BC). The cremation habit edged
into Iberia from what is now southern France. It has often been assumed
that Urnfield was solidly Celtic-speaking. Yet early Greek travellers found
Ligurians along the coast of both France and Iberia.81 [see 42] So did
seafaring colonists spread an Italo-Celtic tongue? It seems that both the
Lusitanians south of the Douro and their northern neighbours the Callaeci
or Gallaeci arrived in the Late Bronze Age. There is no sign of a local
origin for them. There are differences between the two peoples, yet they
shared three deities, suggesting some joint ancestry.82
This does not rule out an Atlantic component in the Gallaecian mix. By
the time that records appear of place-and personal names, Gallaecia was
divided into two regions: in the north the Lucensis, with a centre at what
is now Lugo, Spain, and in the south the Bracarensis, with a centre at
what is now Braga, Portugal. [see 64] They were divided by the river
Minho/Miño, which forms the northern border of Portugal. Gallaecia
Lucensis appears to have been mainly Celtic-speaking, in contrast to the
more Lusitanian affinities of Gallaecia Bracarensis.83 So trade contacts
during the Atlantic Bronze Age may have spread Celtic to northwestern
Iberia.

Overview
• Standing stones – stelae – with hints of the humanoid link Yamnaya
to Bell Beaker, accounting for the spread of Indo-European to the far
west of Europe.

• Stelae of this series started at the western end of the European

steppe. The tradition was taken up by the Yamnaya horizon from
3300 BC.

• There is a Yamnaya stelae trail up the Danube to the Carpathian

Basin. Similar stelae appear in northern Italy, southern France and
western Iberia.

• A connection with Bell Beaker is made at Sion, Switzerland, where

earlier stelae continue to be supplemented in the early Bell Beaker
period.
• Bell Beaker pottery first appeared in Portugal in the 2700s BC, but its
influences seem to be from the European steppe and Carpathian
Basin.

• The first language to travel with Bell Beaker was probably Old
European IE (Alteuropäisch). Proto-Celtic probably developed
around the Upper Danube in the Late Bell Beaker period, spreading
c. 2200 BC to Iberia and the British Isles.

• Movement during the Atlantic Bronze Age (c. 1300–700 BC) may
have spread Celtic to northwestern Iberia.
CHAPTER SIX

The Iron Sword

We should call a man mad, or else insensitive to pain, if he

feared nothing, neither earthquake nor billows, as they say of
the Kelts.1

Thus Aristotle cast a disapproving eye upon the fearless Celts in the 4th
century BC. Courage for its own sake he deemed irrational. Yet the
warrior ethos was far from unknown to the Greeks. The Iliad and
Odyssey are songs in praise of heroes and battle glory. Swords, shields
and armour not only feature in these tales, but also turn up in Greek
tombs from many centuries earlier.2
Swords were first made in the Bronze Age. Unlike bows and axes,
initially devised for hunting and chopping, swords had no other function
but to fight other human beings. Likewise, shields and armour would only
be made by communities who foresaw combat. Though helmets and
body-armour could be of leather and shields of wood, metal gave better
protection. For these bronze was first poured into a flat cast to create a
metal sheet, and the sheet was then hammered into the required shape,
using annealing (re-heating) to keep it malleable. In Central Europe the
first armour appears c. 1300 BC at the beginning of the Urnfield culture.
Approximately 120 helmets, 95 shields, 55 greaves and 30 cuirasses are
known from the European Bronze Age. The distribution area of each type
of armour is different; only in the Carpathian Basin and a little further to
the north do we find all of them.3
44 This magnificent Hallstatt C iron sword comes from a cremation grave in Gomadingen,
Germany. It is of the Mindelheim type, with a long, leaf-shaped blade and hat-shaped pommel.
The hilt is decorated with sheet gold.

With the appearance of iron, the paraphernalia of war began to take on

the shape familiar right up to the Middle Ages: the mounted fighter with
an iron sword. [44] The transformation of warfare began with nomads of
the steppe.

Steppe nomads
About the 11th century BC a long arid period began on the steppes of
Europe and Kazakhstan. The Black and Caspian seas shrank. Zones of
vegetation shifted. The population of the steppe faced ecological crisis.
Economies reliant on a mixture of agriculture and cattle-breeding
collapsed and a fully nomadic lifestyle took their place. Mobile herders
could avoid over-grazing any area, moving on periodically to greener
pastures. A culture sprang up on the European steppe which lasted long
enough to be known to ancient writers as Cimmerian. [45] Its origins were
mixed. Influences from the Asian steppe blended with those from North
Caucasus and the remnants of the dying cultures of the western steppe.4

45 The Cimmerians fled west and south under the pressure of Scythian advances. Both peoples
were Indo-European steppe nomads.

For these nomads horse riding was crucial. They developed a new
bridle. The Cimmerian bits had two movable parts, meant for riding,
unlike the rigid bits of the Urnfield culture, which were more suitable for
traction. From 900 BC weaponry and horse harness that we can trace to
the Cimmerians appear in the Late Urnfield and related cultures of the
Upper and Middle Danube region. The characteristic weapons were
swords and daggers with iron blades and bronze hilts. Iron-working had
begun early on the Pontic steppe. From about 800 to 700 BC ceramics
and ornaments from the same source are found in the graves and hoards
of the Early Hallstatt in Central Europe. Cimmerians had settled in the
Carpathian Basin with their steppe-bred horses and the habit of wagon
burial. Their influence on the developing Hallstatt culture was profound.5
Meanwhile metal-smiths in the Late Urnfield tradition of the Carpathian
Basin seem to have taken their skills to Etruria, creating new centres of
metal production for local elites and foreign exchange.6 This placed the
Hallstatt culture on a trade crossroads.
Could the Cimmerians understand the Celts? The effort to
communicate may have had a linguistic result. Celtic shares one
linguistic feature with Iranian (the syncretism of plain-voiced and ‘voiced
aspirated’ stops) that is not shared with Proto-Indo-Iranian or Proto-Italo-
Celtic. This points to a meeting between Celtic-and Iranian-speakers
sometime after 2000 BC. The feature is also shared by Baltic, Slavic and
Albanian.7 Cimmerian contacts on the steppe and up the Danube might
explain this, though there is so little evidence of the language spoken by
the Cimmerians that we can only guess that it was a member of the
Iranian family, if we recall the contribution to the Cimmerian culture from
the Asian steppe.

Hallstatt aristocrats
Richly furnished wagon burials of the Hallstatt culture are found in
southern Germany, eastern France, Switzerland, Bohemia and Upper
Austria. They conjure up a society dominated by an elite. Wealth and
power were concentrated in the hands of warriors entrenched on fortified
hilltops.
Their wealth came from control of trade routes. Amber was prized far
and wide. Its look of translucent honey trapped in time has long
fascinated people. Its use for jewelry dates back deep into the past.
Although it occurs in various places in Europe, the Baltic is by far the
leading source of amber. Chemical testing can identify Baltic amber, so
we can follow its progress from source to destination. In the early stages
of the Hallstatt culture, it crossed the Hallstatt zone from the Danish
island of Funen (Fyn) to Etruria. Sometime in the 8th century a more
easterly route was established from the amber-rich Pomeranian coast to
Italy and the Balkans.8 [46] In return the Hallstatt chieftains craved the
best that Mediterranean civilizations could supply. To judge by imported
wares designed for drinking or transporting wine, the Hallstatt elite fully
appreciated this particular grape derivative.9

46 The key to understanding the rise of Celtic chiefdoms is control over trading routes. Here we
see the Hallstatt culture and its trade connections.
 One item seldom traded from the early Hallstatt culture was the iron
sword. This new weapon was taken up with enthusiasm by Hallstatt
chiefs and is the standard type in their wagon burials, but it did not travel
far from the Hallstatt zone initially. The Hallstatt people had adopted iron-
working in the 8th century BC as an addition to bronze-making, rather
than an immediate replacement. Some other parts of Europe were even
slower to switch to the new metal. So instead of a sharp transition
between the Age of Bronze and that of Iron in Europe, there was a
gradual shift over centuries. Iron had the advantage of being readily
available, but the technology for working it was complex and labour-
intensive. Rather than being cast, iron was forged and hammered on an
anvil at red-heat. This process was not totally unfamiliar to bronze-smiths
accustomed to hammering objects from sheet metal. So now to armour
could be added the iron sword.10
The best-studied Hallstatt hillfort is the Heuneburg in southern
Germany. [47] It is superbly sited. [see 46] The hill overlooks the Upper
Danube, thus controlling east–west trade from the Danube to the Rhône
Valley, and also a route south from Pomerania to Italy. Extensive
excavations have revealed a major political and commercial centre.
Extraordinary power required extraordinary protection. When it was first
laid out the Heuneburg had houses and a defensive wall of traditional
type for the region, but around 600 BC this first wall was replaced by a
Mediterranean-style defence with bastions made of sun-dried bricks,
unique in Central Europe. Was a southern architect enticed north by a
prudent prince?11
Recent work has uncovered more of a fortified lower town, and in the
process overturned the prevailing view of it. Scholars had long thought
that these secondary fortifications were medieval. The research project
‘Early Celtic Princely Centres’ (2004–2010) made the startling discovery
that they were in fact constructed in the Final Hallstatt period. Remains of
a timber bridge crossing the outer ditch could be dendro-dated to 590 BC.
Also uncovered was a massive gatehouse of around the same date
guarding the entrance to the lower town. Outside the lower town, closely
spaced farm-steads were protected by palisades. They presumably
supplied food to the denser population within the walls. This
concentration of people gives the Heuneburg an urban flavour. A town
has its own economy. It may provide a market for agricultural produce,
but it is more than a farm or estate centre. Goods are made and traded
there. Local authority may be centred there. Pottery, jewelry and textiles
were certainly made at the Heuneburg, but territorial and trade power
was probably more significant than local craft. Heuneburg’s impressive
defences and the sumptuous burials around the settlement proclaim its
status as a political capital.12
Not all the rich burials nearby are of warrior chiefs. Mound 4 of the
Bettelbühl necropolis, 2.5 km (roughly 1½ miles) southeast of the
Heuneburg, has been yielding wondrously rich grave goods. The main
burial was that of a wealthy woman, dubbed a Celtic princess by her
discoverers. She lay in a chamber constructed of timber from a tree cut in
the year 583 BC; it proved to be a treasure-house of high-quality craftwork
in gold, amber and bronze. A secondary burial of a two-to four-year-old
girl was also furnished with golden jewelry. So young a child could not
have acquired status by her own actions. Here, then, is a hint that the
Celts were developing the idea of inherited status, the foundation of
aristocracy.13
47 The flat top of the Heuneburg was ideal for a hillfort. Below it lay a fortified lower town, with
farms outside it.

Of the 100 or so Hallstatt wagon graves, the one at Hochdorf stands

out for its conspicuous consumption. [49] The contents of the burial
chamber were exceptionally well preserved, so we know that a tall man
of about 40 had been buried in colourful textiles and a conical hat made
of birch bark. A similar hat (and little else) is worn by the Warrior of
Hirschlanden, the earliest life-size and lifelike statue known from a Celtic
sculptor. [48] It is what the Hochdorf chief wore on his feet, though, that
has attracted most attention. His pointed, slightly upturned leather shoes
had lavish gold leaf decoration. A BBC television programme in 1987
christened him ‘The Man with the Golden Shoes’. Gold was not spared
elsewhere about his person either. He wore a gold collar and bracelet.
His belt was decorated in gold, and the bronze dagger beside it was
completely covered in gold leaf. If this gold-laden costume strikes you as
impractical day-to-day wear, you would be right. All the gold leaf was
added specifically for the burial.
This glittering figure had been tenderly laid on blankets spread on a
bronze couch, unique for its place and time. It is decorated with warrior
scenes and supported on wheeled female figures. At the foot of his couch
was a massive bronze cauldron which had held about 450 litres (100
gallons) of mead. It is decorated around the brim with three lions, beasts
unknown to the Celts of this era. That alone should tell us that this
magnificent object was imported. Indeed it is Greek in style and its
copper has a high bismuth level, which suggests a Greek ore. It was
probably made by Greeks in southern Italy. On the other side of the
chamber was an impressive iron-bound four-wheeled wagon.14 The iron
tyres had been nailed in place, which must have made for a bumpy ride.
(From around the 2nd century BC such iron tyres were heated and then
allowed to cool on the wooden wheel, shrinking them into place.15)
The gold collar worn by the Hochdorf chieftain is sometimes described
as a torc, but that is misleading. It was made of sheet gold which had not
been rolled into the circular cross-section of a torc. It had no opening. It
was large enough to slip over the head of its wearer and sit on the
shoulders. The classic torc is a ring close around the throat, which opens
at the front. The Warrior of Hirschlanden wears one. So did a man buried
in some state at Hochmichele, 3.5 km (2 miles) west of the Heuneburg, in
the Final Hallstatt period.16 Particularly interesting is the burial of a torc-
wearing warrior at Saint-Romain-de-Jalionas (France) in the 8th century
BC,17 since that takes us back to the start of the Hallstatt culture. With
these Hallstatt-era burials, we have solid evidence that torcs were not
simply an item of jewelry for women, but were also worn by high-status
men.
48 The Warrior of Hirschlanden is the earliest life-size and reasonably lifelike stone statue known
from a Celtic sculptor. He wears a torc around his neck and a belt with a typical late Hallstatt
dagger. His pointed hat was perhaps made of birch bark, like that found in the princely grave of
Hochdorf, approximately 5 km (3 miles) from Hirschlanden. The sculptor gave him legs thick
enough to provide solid support and avoided the arms breaking off by portraying them clasped to
the body.

Rivalling the Hochdorf chief in magnificence is the Lady of Vix. She is

named for the French village where she was found. The Vix area is
dominated by Mont Lassois, a steep, flat-topped hill overlooking the
upper reaches of the Seine. Today the church of St Marcel of Vix stands
upon it, while the village itself lies in the valley by the Seine. Southeast of
the village, within a curve of the river, is a necropolis first used in the late
Bronze Age. Here was found the wagon tomb of the Lady of Vix, the
contents of which remain unparalleled. Her wealth can be explained by
someone’s shrewd grasp of geography. The top of Mont Lassois was
ideal for fortification and it stood at a route axis. [see 46] Diodorus
Siculus describes tin being transported from Cornwall to Gaul and then
south to the mouth of the Rhône.18 Tin travelling up the Seine could have
been disembarked near Mont Lassois, where the Seine ceased to be
navigable, and transported by land to the Saône Valley, leading to the
river Rhône.

49 Reconstruction in the Keltenmuseum Hochdorf/Enz of the chieftain’s wagon burial discovered

at Hochdorf in Baden-Württemberg, Germany. The chieftain was laid on a couch made of sheet
bronze, with a huge bronze cauldron at its foot which had been filled with mead. The wagon in the
foreground is heaped with bronze dishes and horse harness.

Luxury goods from the Mediterranean are lavishly displayed in the

tomb of the Lady of Vix. So rich a female burial has fascinated
generations of scholars. How could she fit into a society of warrior
princes? Some even wondered if the body could be that of a male
transvestite, until its gender was finally determined by DNA. This was
indeed a woman who died at the age of about 30 and was buried
between 500 and 450 BC.19 She lay on the body of a wagon, the four
wheels of which were placed against the walls of the chamber. She wore
a hollow gold torc unlike any other. Ball terminals are found on many later
examples, but these are unusual, being at right-angles to the ring. The
junction between ring and terminal is decorated with a lion paw on the
inside and a winged horse on the outside. [50] Pegasus leaps to mind
when we think of winged horses. On the Vix torc he stands on an intricate
filigree pedestal. Could that depict the Hippocrene (‘horse-fountain’)
spring on Mount Helicon, which Greek myth-makers declared had been
created by Pegasus stamping his hoof?20 The idea that those who drank
of the Hippocrene were granted poetical inspiration is not recorded until
long after the burial of the Lady of Vix. Since eloquence was so highly
valued by the Celts, it is tempting to think that we see here a symbol of
the flight of imagination, but perhaps that would be a flight of imagination
itself.
The Lady of Vix was amply supplied with other jewelry too, in gold,
lignite, coral, amber and bronze. The most spectacular feature of her
tomb, though, was a gigantic bronze krater, dwarfing even the Hochdorf
cauldron. The Greek krater was a vessel wherein water and wine were
blended. This particular krater is 1.64 m (5 ft 5 in) high and has a
capacity of 1,100 litres (roughly 240 gallons). Around the neck is a
parade of Greek warriors, some on foot, others driving four-horse
chariots. Most authorities agree that it came from the same Greek
workshop as the cauldron at Hochdorf.21
The Lady of Vix was evidently an important person, but what exactly
was her role? Was she a priestess or a political leader? Even within
patriarchal societies, women may sometimes inherit a powerful role from
father or husband, in default of a male heir. Boudica is a famous example
among the British Celts. She led her people after her husband died
leaving only daughters.22 Clues emerged nearly 40 years after the
excavation of the tomb of the Lady of Vix. In 1991 aerial photographs
revealed a square feature nearby. Excavation brought to light two
headless, seated statues, one depicting a warrior and the other a woman
wearing the distinctive torc of the Lady of Vix. This was a significant
discovery. These figures were placed within a ditched enclosure at the
heart of the necropolis, interpreted as a sanctuary.23 The pairing of the
Lady of Vix with a warrior suggests that they were man and wife, or
closely related. They could be the founders of a dynastic lineage. Some
political upheaval around 450 BC led to the beheading of these powerful
images at Vix. At the same time, Mont Lassois lost its dominant
position.24
50 The unique gold torc worn by the Lady of Vix, with a detail showing one of the winged horses
decorating its ball terminals.

Glauberg (Hesse, Germany) is a fine example of a late Hallstatt/early

La Tène (6th–4th century BC) princely seat. Like the Heuneburg, it is a
fortified hilltop with farms and burials on the slopes below it. In July 1996
archaeologists were excited to discover beside one of the burial mounds
the almost complete statue of a man soon dubbed the ‘Glauberg
prince’.25 [51] Intriguingly, the statue is crowned with something akin to
the laurel wreaths bestowed upon Roman emperors. A similar headdress
appears on several roughly contemporary statues and other depictions
from the Hallstatt zone.26
The mound beside this striking statue had already been excavated. It
contained the princely inhumation of a man in his 20s, which had
escaped the attention of tomb robbers. It was so well preserved that it
has yielded a mass of information. There can be no doubt that the statue
portrays this young lord, for the burial contained a magnificent gold torc
which matched that on the statue. [52] Other jewelry included arm-rings
and a finger-ring, also as seen on the statue. It was even possible to
detect the remains of his leaf crown.27
The buried prince was equipped as a warrior with an iron sword,
spearheads, arrowheads in a quiver and a shield. Bioarchaeological tests
revealed that he had enjoyed a high-quality diet with plenty of animal
protein. By contrast, individuals buried in pits nearby had numerous joint
lesions, indicating a strenuous lifestyle. Their diet was much poorer than
that of the prince. They ate a lot of millet and less animal protein. Their
burials were so informal that it even crossed the minds of their
excavators that they might be slaves. We cannot be sure. What stands
out is the contrast between prince and pauper.28
Like the Warrior of Hirschlanden, the Glauberg prince is not as
confidently representational as Greek sculpture of the same period. While
the legs are lifelike, the sculptor avoided piercing the upper section.
Leaving gaps between body and arms would risk broken arms. That
solidity gives the upper body a similarity to the anthropomorphic stelae of
the Copper Age, but the considerable gap in time makes it impossible to
claim a continuous tradition from the earlier stelae to these Iron Age
statues.
51 This sandstone statue from Glauberg in Hesse, Germany, is a rare depiction by Celts of a
Celtic warrior. He wears a three-pronged torc like one from the adjacent burial mound (below). On
his head is a curious crown of leaves, remains of which were found in the same grave.
52 This magnificent and unique gold torc comes from the princely burial of a young man at
Glauberg. Most torcs were simply rings around the neck, plain or twisted. The front half of this ring
is decorated with ten human face masks, and from it protrude decorative finials flanked by two
grotesque figures with enlarged heads.

Celtic caught in transition

The Golasecca culture of the north Italian lake region acted as a trade
gateway between Hallstatt and the Etruscans. Here Celtic-speakers were
in contact with literacy. They adapted an Etruscan script to write in their
own language, and so left us the first inscriptions that are generally
agreed to be Celtic. There are about 140 such inscriptions, beginning c.
600 BC and ending around 1 BC.29
Their language was labelled Lepontic by modern scholars, knowing
that a people called Lepontii lived within the area in Roman times. In 55
BC Julius Caesar noted that the Rhine rose in the country of the Lepontii
in the Alps.30 In the 1st century AD Pliny the Elder wrote that one group of
the Lepontii lived around the source of the Rhône.31 A century later
Ptolemy mentions the Lepontii living in the Alps and their town of
Oscela.32 This town (now Domodossola) has given its name to the valley
(Val d’Ossola) in which the Toce River runs. The upper reaches of the
River Ticino run through the Valle Leventina, which preserves the name
of the Lepontii. What all these locations have in common is rivers with
their source in or close to the Gotthard massif. [53]

53 The region in which Lepontic inscriptions are found is circled. Golasecca culture finds are
scattered over the same area and as far south as the River Po. Important Alpine passes are
marked by red triangles.

Today traffic hums daily through the high Gotthard Pass. It crosses the
saddle between the sources of two rivers: the turbulent Reuss, running
north into Lake Lucerne (Switzerland), and the Ticino flowing south
through Lake Maggiore (Italy) to join the Po. To the west the Simplon
railway tunnel connects Brig in Switzerland to Domodossola in Italy.
Before its construction the connection was via the Simplon Pass. Both
these passes were far more dangerous before modern roads and
bridges. Yet the Golasecca culture clearly thrived on trade through these
and other Alpine passes. Objects of Golasecca manufacture are found
north of the Alps, along with Etruscan luxury goods, while Hallstatt
influences trickled south.33
Naturally, scholars have wondered how Celts came to be in northern
Italy before the well-known expansions of the Gauls in the 4th century BC
(discussed in Chapter 7). The Golasecca culture began in the 8th century
BC, developing from the local Bronze Age culture. This can be traced
back to a variety of Urnfield that arrived in the 13th century BC. Given its
wide cultural connections north of the Alps, and its seemingly abrupt
arrival, this brand of Urnfield has been generally favoured by scholars as
the vector of the first Celtic language south of the Alps.34
Yet if we go further back in time we reach the Bell Beaker sites of Sion
on the upper Rhône and Aosta in the southern Alps. [53] It will be
remembered that they received a sudden influx of new, eastern Bell
Beaker material c. 2425 BC (see p. 97). So here we have an earlier wave
that could have brought Celtic to the southern Alps. If so, the continued
use of trans-Alpine trade routes would keep those Celts of the southern
Alps in touch linguistically with the developing core of Proto-Celtic north
of the Alps as it gradually turned into Gaulish. Yet Lepontic seems best
classified as a separate language from Gaulish.35
The relationship between the various Celtic languages is complex. [54]
One sound change is well known: the shift from kw to p in some Celtic
languages. The Proto-Indo-European and Proto-Celtic kw sound can be
seen in the PIE word for horse,*ekwos, which became equus in Latin and
equos in early Celtic, and passed down into Old Irish as ech. Equos
appears as a month on the lunar calendar inscribed on bronze which was
discovered near Coligny in France. The Coligny calendar uses the Latin
alphabet, which dates it to the Gallo-Roman period.36 By that time the qu
sound had been generally replaced in Gaulish by p, as we see in the
many dedications to the horse goddess Epona.37 Indeed this sound
change had taken place centuries earlier, for it appears in the earliest
name known for Britain (see pp. 169–70). Ptolemy’s Geography records
‘p’ forms of tribes in Britain and Gaul: the Epidii (people of the horse) in
Kintyre, and the Menapi of Belgic Gaul. Yet the archaic form Sequana
was retained for the Seine.38 The occasional archaism helps to uncover
the process of change.
54 This conjectural tree of the Celtic languages attempts to take into account the complexities
caused by wave after wave of Celtic dialects entering the British Isles and interactions between
speakers of the different languages.

So it has been customary to label the Brittonic group of Celtic

languages together with Gaulish, as P-Celtic [see 54], while the Gaelic
group is labelled Q-Celtic. Language trees have often simply split Celtic
into those two familiar categories. Linguist Kim McCone protested against
this, arguing that the important division is between Insular and
Continental Celtic. Insular in this context includes Breton (as a
descendant of Brittonic). There is certainly a major difference between
Insular and Continental Celtic. The latter was dead by the end of the
Roman period. Welsh, Breton and Gaelic have gone on developing since.
This means that they have had centuries to develop differences from the
Continental group. So comparing ancient with modern could be
misleading.39
The earliest form of Celtic to enter the British Isles was probably Proto-
Celtic, which would gradually diverge from its parent. La Tène
movements from Gaul to Britain would explain the similarities between
Gaulish and Brittonic. Continuing interaction between Celtic-speakers in
Britain and Ireland would blur the distinctions between languages. So we
can understand the similarities between Brittonic and Irish, without
ascribing them all to a very early date.40

La Tène warriors
There is no abrupt break between the Hallstatt and La Tène cultures. The
one flows into the other. We have already seen that the important site at
Glauberg flourished from Late Hallstatt to Early La Tène (see p. 111). At
Hallstatt itself there is a notable tomb of the early La Tène era. Continuity
seems quite common.41 Yet signs of power struggles around 450 BC crop
up here and there. The abandonment of Mont Lassois has already been
mentioned (p. 111). The Heuneburg was abandoned around 450 BC, with
no obvious explanation.42

55 Map of early La Tène chariot burials, 5th–early 4th century BC.

 Around the same period new centres of power appear along the
northern fringe of the Hallstatt zone [see 11], concentrated particularly
around the rivers Moselle and Marne and in Bohemia.43 [55] All these
areas not only had navigable rivers, but also access to iron ore, and
forest to provide the charcoal for iron-working. The Moselle was a
tributary of the Rhine already well known to the Celts. There are Hallstatt
wagon burials at sites along it from Trier to Koblenz (Germany). In the La
Tène period the wagon gave way to the chariot. Chariot tombs cluster
along the Moselle and the adjoining middle Rhine, and even more
densely between the rivers Marne and Oise, around present-day Reims
(France). In Bohemia they are mainly found along the River Vltava and its
tributaries. The Vltava is the longest river in present-day Czech Republic
and was a trade-route for the Celts.44
The chariot had become one of the key symbols of the Celtic warrior
aristocracy, along with the gold torc, wine and weaponry (see also pp.
34–38). Instead of burial under an impressive mound, we find flat graves
of this period in which males were buried with weaponry and females with
jewelry. The chariot itself became a fearsome thing in battle as the Celts
learned how to deploy it to great effect against their foes.
The Celts came late to chariot-building. Greek chariots are pictured on
the Vix krater, with their typical four-spoke wheels, while the Celts still
favoured wagons. Celtic chariot burials start in the 5th century BC.
Generally the wooden parts of Celtic chariots have decayed and all that
is left is metal, but the lake mud of the La Tène site preserved yokes and
a wheel [see 13], as well as metal parts of a chariot including cotters. A
cotter is a bolt or pin used to secure two other parts, a clue that the
chariot had some form of suspension. So Andres Furger-Gunti of the
Swiss National Museum in Zurich engaged craftsmen to reconstruct the
La Tène chariot as an exercise in experimental archaeology. He was
particularly interested in solving the problem of the suspension. He opted
for a longitudinal type, with two sidearms curved up and back from the
pole, to provide rests for the double cotters supplying the suspension for
the platform. The end result was fully functional. Only then was he told
that Irish literature supported him.45 A more recent analysis of early Irish
chariot terminology drew the same conclusion.46

Shape-shifting art
Shape-shifting art
There is more to the Celts than warfare and chariots. This is not a book
about Celtic art, yet we cannot pass over the artistic flowering in the La
Tène period. Contact with Mediterranean civilizations brought Classical
motifs into the Celtic repertoire. Even the earliest Celtic coins copied
Greek coins (see p. 129). The fantastic beasts of Scythian goldwork may
have been another influence on the Celtic aesthetic, though the parallels
are not close.47
What lifted Celtic art above imitation was the La Tène confidence to
innovate. A Celtic preference for curvilinear rather than angular design
can be seen in the Hallstatt period, but these patterns were regular and
could be drawn with compasses. Such work continues into La Tène, but
we also see a new freedom. Gradually the full glory of the Celtic visual
imagination emerged. Curving plant tendrils could sprout a sleek animal
head or melt into abstract twists and turns. A horse could be evoked in
fluid curves that make it look half seahorse. [56] Celtic art had parted
company from the drive towards ever greater naturalism within the
Classical world. The famous head from Bohemia [57] makes a perfect
contrast with the equally famous sculpture of the Dying Gaul.48 [see 63]
Both have the Celtic insignia of torcs and moustaches, but how different
they look. With the Dying Gaul a gifted Greek sculptor imagines reality, or
an idealized version of it. We feel that this man could have breathed and
hurt and died. The head from Bohemia is more of a mask. It conjures up
the haughty pride of an individual, yet sets that person apart from the
human realm, with his bulging almond eyes, twirling eyebrows, and ears
in the shape of a lotus flower.

56 Gold stater of the Parisii in Gaul, c. 70–60 BC.

57 Was this a leader of the Boii or a sacred figure? This stone head from the shrine at Mšecké
Žehrovice, Czech Republic, has all the hallmarks of the La Tène style, with curvilinear shapes
favoured over naturalism.

Oppida
By the 2nd century BC Celts were developing their own towns. Caesar
used the word oppidum (plural oppida) to describe the major Celtic
settlements that he encountered on his campaign of conquest. This Latin
word was used in Italy for the civic centre of a territory and its people. No
doubt Caesar perceived a similar status for settlements functioning as
tribal centres in Gaul. Much ink has been spilled debating whether oppida
were truly towns. Part of the confusion arises from modern
archaeological usage of the term. It has sometimes been so loose as to
include all fortified Celtic sites of the late La Tène period. A fort is not
always a town and vice versa. While most of the oppida Caesar knew
were indeed walled towns, a town of this date was not invariably walled.
Nor were early towns necessarily grid-planned, though they could be.
The key factors that point to urban life are trade, manufacture and
administration. Clues in the archaeology could be public buildings or coin
minting.
Some of these Celtic towns were the predecessors of Roman and
modern towns or cities. Others faded away before the Roman conquest.
Some were developments of earlier hillforts. Others were laid out in
lowlands. Generalization therefore would be folly. One common factor
though is a location on a trade route, usually riverine or coastal.
Zavist, now on the southern fringe of Prague (Czech Republic) was an
excellent choice of site: a large and easily defended hilltop, overlooking
the Vltava near its confluence with the Berounka. A town with massive
ramparts was built there in the 2nd century BC. The Celtic name of Zavist
is unknown, but it seems to have been the capital of the Boii tribe, who
gave their name to Bohemia. Zavist flourished long enough to see
several phases of rebuilding, growing to around 118 ha (290 acres) in
size, but was abandoned in the late 1st century BC, as the Boii came
under pressure from expanding Germani. The Boii left a trail westwards.
An inscription mentioning a member of the tribe was found in Manching in
Bavaria (Germany). The Boii settled a short distance to the east at
Boioduron (fortified settlement of the Boii), modern Passau.49
By comparison with Zavist, the oppidum of Manching was a giant, at a
final 380 ha (938 acres). It sprawled on a flood-free low terrace beside
the River Paar, which at that time flowed into an arm of the Danube
nearby. So it was well placed for trade. Iron-smelting sites dotted its
hinterland, and metal waste within the town attests to a plentiful
production of iron objects, but Manching did not depend on a single
industry. Tools for leather-and textile-working have been found within the
oppidum. The town was laid out with streets, squares, cult structures,
commercial districts and workshops. Defence was evidently not the first
consideration. When the settlement began around 300 BC it had no
surrounding walls. It acquired ramparts c. 120 BC. Despite all the effort
that must have gone into the creation of a 7-km (over 4-mile) long
fortification, Manching was abandoned c. 50 BC, before the Roman
conquest of the region.50
Was it the success of the La Tène culture that generated the will to
expand into new territories? Or was Late La Tène a culture in decline and
under threat, jettisoning streams of migrants to find a life elsewhere?
These are questions tackled in Chapter 7, as we focus on the wanderlust
of the Gauls.

Overview
• In the Iron Age, the paraphernalia of war began to take on the shape
familiar right up to the Middle Ages: the mounted fighter with an iron
sword.

• The transformation of warfare began with nomads of the steppe. By

moving up the Danube into the Carpathian Basin, they introduced
iron-working, swords, horse gear and wagon burial to the developing
Hallstatt culture.

• The Hallstatt culture was dominated by a warrior elite, entrenched

on fortified hilltops and buried often with wagons. Such burials are
found in southern Germany, eastern France, Switzerland, Bohemia
and Upper Austria.

• The Golasecca culture of the north Italian lake region acted as a

trade gateway between Hallstatt and the Etruscans. Celtic-speakers
adapted an Etruscan script to write in their own language, called
Lepontic by modern scholars. Inscriptions in Lepontic start from c.
600 BC.

• Some Hallstatt sites were abandoned around 450 BC. Around the
same period new centres of power appear around the rivers Moselle
and Marne and in Bohemia, which represent the start of the La Tène
culture.

• Key features that distinguish La Tène from Hallstatt are chariots and
greater artistic freedom and confidence.
• By the 2nd century BC Celts were developing their own towns, which
the Romans called oppida. The word has remained in the
archaeological vocabulary.
CHAPTER SEVEN

On the Move

The Gauls, when the land that had produced them was unable,
from their excessive increase of population, to contain them,
sent out three hundred thousand men, as a sacred spring, to
seek new settlements. Of these adventurers part settled in Italy,
and took and burnt the city of Rome; and part penetrated into
the remotest parts of Illyricum under the direction of a flight of
birds (for the Gauls are skilled in augury beyond other nations)
making their way amidst great slaughter of the barbarous tribes,
and fixed their abode in Pannonia. They were a savage, bold
and warlike nation, and were the first after Hercules … to pass
the unconquered heights of the Alps.1

So wrote Pompeius Trogus, himself a Romanized Gaul, in the 1st century

BC. These events were centuries earlier, but still fresh in the minds of
both Romans and Greeks. Where the Celts thrust into the civilizations of
the Mediterranean, their onslaughts were recorded for posterity. We hear
of their exploits from their enemies. The attack on Rome in 390 BC burnt
its brutal message into the collective consciousness of an emerging
power, and spurred the Romans to ever more conquests to bolster their
security. The Romans neither forgot nor forgave and were to turn the
tables centuries later.

58 Birds were a popular emblem in Celtic art. Here they decorate a bronze flesh-fork, used in
feasting, from Dunaverney, Co. Antrim (950 BC–750 BC). Two swans and three cygnets face two
ravens. Both swans and ravens appear in Irish myth.

Other Gauls marched southeast in the 3rd century BC even as far as

Greece and Anatolia. Their attack on the sacred site at Delphi deeply
shocked the proud Greeks and was likewise never forgotten. By contrast
Celtic movements into prehistoric Britain and Iberia have to be traced
mainly by the objects and place-names they left behind.

The forces driving expansion

What were the reasons for this burst of mobility? As mentioned in the last
chapter (p. 120), the advance of the Germani into what had been Celtic
territory put such pressure on the Boii that Zavist was abandoned in the
late 1st century BC. Germanic expansion had started centuries earlier,
bearing hard on Celtic-speakers east of the Rhine. Gaulish druids knew
that their land was partly inhabited by the descendants of refugees.
Fortunately, Roman historian Ammianus Marcellinus preserved their
account. People had ‘poured in from the islands on the coast, and from
the districts across the Rhine, having been driven from their former
abodes by frequent wars, and sometimes by inroads of the tempestuous
sea’.2 Caesar puts a name to these incomers. They were Belgae. He
says that their ancestors had long ago come across the Rhine from
Germany, expelling the former inhabitants from northeast Gaul.3 Their
recorded tribal, personal and place-names are Celtic (with very few
exceptions). They had a late La Tène culture. Thus their ancestry was
from what the Romans called Germania, but they were Celts.4
The reference by Gaulish druids to islands and the tempestuous sea is
significant. The North Sea coast of what is now the Netherlands and
Germany has been gradually subsiding for millennia, creating the
Wadden Sea, with its chain of islands separating it from the North Sea.
Periodic floods would have been catastrophic for Germani spreading out
from a homeland in north Germany and Poland. So in addition to floods
directly affecting Celts east of the Rhine, there would also have been
pressure from Germani trying to escape these disasters.
As the Belgae were displaced into Gaul, they in turn displaced Gauls.
The Parisii are a case in point. The Marne territory was thick with early
chariot burials. Yet in the later La Tène period chariot burials of the same
type cluster around Paris, territory of the Gaulish Parisii.5 We have
already seen the remarkable example at Roissy c. 300 BC (p. 37). It
seems that the Parisii were pushed westwards by the Remi, a Belgic tribe
who in Caesar’s day had an oppidum at Durocortorum,6 the site of
modern-day Reims. Such a crisis might explain why Ptolemy recorded
Parisi in Britain. The chariot burial at Ferry Fryston which seemed so
similar to Continental types was dated c. 355 BC (see p. 38). Why choose
East Yorkshire? There was easy access to iron ore and woodland for
charcoal. A concentration of iron-working in the area is contemporaneous
with the chariot burials and close to them.7
The chariot at Newbridge in Scotland (5th century BC) seems too early
to be part of this domino effect. Indeed, La Tène influences generally
started seeping across the Channel around 450 BC. Even before that,
Hallstatt material is found in Lowland Britain; it reached as far north as
the Forth–Clyde line. Then there is the surge southwards across the Alps
around 400 BC. Pompeius Trogus, as we saw in the quotation at the head
of this chapter, blamed overpopulation. He also mentions ‘civil discords
and perpetual contentions at home’.8 His contemporary Livy knew of two
entirely different colourful traditions about the coming of the Gauls to
Italy. He separated the two in time. One, which he set around 600 BC,
tells of an otherwise unknown Gallic king, Ambigatus, under whose sway
‘the harvests were so abundant and the population increased so rapidly
in Gaul’ that he was ‘anxious to relieve his realm from the burden of
overpopulation’. So he sent his nephews to lead the adventurous to
distant lands, including Italy.9 Overpopulation from agricultural surplus is
entirely credible for the period 650–450 BC, when the climate was warm
and dry. However, the major onrush of Gauls to Italy came around 400
BC, when there was a climate crisis and land marginal for cereal
production was abandoned.10
The other tradition Livy knew related to this later period. It claimed that
the Gauls were lured into Italy by its delicious fruits and wine, introduced
to them by a trader from Clusium (Chiusi) who wanted revenge on the
city. This story was told to explain the siege of Clusium in 391 BC.11 Livy
seems to have taken it from the works of Dionysius of Halicarnassus,
who goes into more detail, some of it seriously uncomplimentary to Gallic
cuisine:
The Gauls at that time had no knowledge either of wine made from
grapes or of oils such as is produced by our olive trees, but used for wine
a foul-smelling liquor made from barley rotted in water, and for oil, stale
lard, disgusting both in smell and taste.12
Pliny the Elder in the 1st century AD gives another tale of gastronomic
temptation:
It is related that the Gauls, separated from us as they were by the Alps,
which then formed an almost insurmountable bulwark, had, as their chief
motive for invading Italy, its dried figs, its grapes, its oil, and its wine,
samples of which had been brought back to them by Helico, a citizen of
the Helvetii, who had been staying at Rome, to practise there as an
artisan.13
Here Pliny captures the mechanics of migration. People often move for
economic advantage, but they need to know of good opportunities. Trade
is one route to hear of them. It can encourage a few migrant workers,
who then communicate with the people back home, encouraging more to
make the journey. So a trickle of migration can turn into a flood. Pliny
condenses the process into a simple story of one man. In fact these trade
goods were spreading much more widely than to just one tribe. Diodorus
Siculus tells us that the Gauls were so addicted to wine that they would
hand over a slave in exchange for just one jar of the intoxicating drink.14
Wine amphoras appear in Gaul and southern Britain with increasing
frequency in the generations prior to the Roman conquest. [59] (Not all
the Belgae succumbed. The Nervii refused entry to traders for fear that
wine and other luxuries would weaken their warriors.15)
59 A cremation burial of the Late Iron Age at Welwyn, southern England, was rich in feasting
equipment, including five amphorae. This extravagant type of burial is uncommon in Britain and
mainly restricted to the region controlled by the Catuvellauni, a Belgic tribe.

Wine was not the only lure of the south though. Wealthy warlords
attract warriors. The capacity of fighting Celts to inspire terror and panic
made them sought after as mercenaries. Celts were employed by the
tyrants of Syracuse, Sicily, as early as the 4th century BC.16

The surge southward

Through the channel of the Golasecca culture (pp. 113–14) the Gauls
had plenty of time to assess the power politics of northern Italy before
they ejected the Etruscans from the Po Valley around 400 BC. [60] Indeed
that is precisely the picture painted by Polybius, a Greek historian closer
in time to these events than Livy and his contemporaries. Polybius was
born around 200 BC. His great virtue as an historian was his preference
for first-hand evidence. He began his history of the rise of Rome in the
year 220/219 BC, so that he was able to interview people who lived
through events before he was born, and then to record an era of which
he had personal knowledge, ending in the year 146 BC. Conceding that
his Greek readership might not be acquainted with what modern authors
call the back-story, he used his first two volumes to cover crucial earlier
events. Polybius was also among the first to advocate the steely eye of
the seeker after truth in historical narrative. This dispassionate voice is
well worth heeding:

60 The movements of Gauls noted in Classical sources. Celtic-place-names mark the regions that
Celts actually settled, rather than raided.(Compare with [25].) Dates are BC.

Long ago … the Po plain was home to the Etruscans. The Celts
had plenty of dealings with the Etruscans, since they were near
neighbours, and they cast covetous eyes on the beauty of the
land. On some feeble pretext, they suddenly invaded with a
huge army, drove the Etruscans out of the Po plain, and took
the land for themselves. The first part of the territory, near the
source of the Po, became home to the Laevi and Libicii; after
them came the Insubres, the largest of the Celtic tribes; and
 next to the Insubres along the river were the Cenomani.… The
first stretch of land on the other side of the Po, the Apennine
side, became the home of the Anares, the next of the Boii; the
region next closest to the Adriatic was occupied by the
Lingones, and the coastal area by the Senones.17

Rome was peripheral to these events, rather than the prime target.
According to later sources such as Livy, it was the siege of Clusium well
to the south of the Po Valley and dangerously close to Rome, that drew
the Romans into a battle that proved so disastrous for them. [61] The
Gauls were able to plunder Rome itself in 390 BC before they were
themselves defeated.18
The Etruscans had an urban civilization, but what the Gauls wanted
was their fertile land on the plain of the Po. The towns were either left to
crumble or occupied on a reduced scale.19 Polybius tells us that the
villages of these Gauls were unwalled and lacked any civilized
amenities.20 So no sophisticated city should be imagined when we read
that the Insubres founded Mediolanum (Milan) and the Cenomani
founded Brixia (Brescia) and Verona, all Celtic place-names.21
Nevertheless, Italian archaeologists were quietly pleased to find that
Milan, noted in modern times as an industrial centre, can boast a
concentration of Iron Age metalworking sites. Some of these do indeed
pre-date the Roman period and can be linked to the La Tène culture.22
Enough Celtic material has been found over the whole region to support
the historical accounts.23 Furthermore, the incoming Gauls adopted the
alphabet used earlier for Lepontic to leave inscriptions in their own
language, known as Cisalpine Gaulish.24
At around the same time that we find Boii among the arrivals in Italy,
others of this numerous tribe were spreading eastwards into Moravia,
southwestern Slovakia and northeastern Austria. They could then drift
down the Danube to the Carpathian Basin.25 Here they could find good
grazing. From the southern Carpathian Basin the River Sava leads to the
southeastern Alps and ways down to the Adriatic. It seems that some
Celts used the Sava to join up with Celtic-speakers of the Alps, creating a
trade route from the Alps and the Adriatic to the Danube. La Tène
material in the northern Balkans, together with items from the lower
Danube region, tell the tale. [62] Celts who appear in history as the
Scordisci settled around the confluence of the Sava and the Danube.
Their fortress of Singidunum (modern Belgrade) was strategically placed
on a hilltop overlooking this meeting place of riverine routes.26
All this activity meant pushing in among Illyrians. One devious tactic
was recorded by a contemporary Greek historian. The story goes that the
Celts poisoned their own food and wine with noxious herbs, then left their
camp by night in pretended confusion. The Illyrians took possession of
their camp and gorged happily on the provisions found there. Soon they
succumbed to violent diarrhoea. While they were helplessly ill, the Celts
returned and slew them easily.27

61 Bronze figure of a Celtic spearman from the Rome area, 3rd century BC. His torc makes him
instantly identifiable as a Celt. Otherwise he wears nothing but a belt and helmet, in keeping with
Roman descriptions of the Celts fighting naked.
62 Detail of an iron scabbard from grave 6 at Dobova, Slovenia, dating from around 200 BC, with
fine decoration in the La Tène style. Dobova is within the area of the Taurisci.

The young Philip II of Macedon had driven Illyrians out of northwest

Macedonia in 358 BC. On the principle that my enemy’s enemy is my
friend, the incoming Celts may have been well disposed towards Philip.
There is no surviving record that Philip employed Celts as mercenaries,
yet his coinage became familiar to many of them. His silver tetradrachms
were copied by the Boii and Scordisci in minting the first Celtic coins.
Certainly, after Philip was assassinated in 336 BC, Celts from near the
Adriatic sent ambassadors to his son Alexander the Great, as he secured
his northern border the following year. The story is told that Alexander
welcomed them warmly, and while they were sharing a drink asked what
they feared the most, expecting them to confess to an overwhelming fear
of him. Instead they replied that they feared nothing except that the sky
might fall on them.28 One suspects a problem in translation. The Celts
saw the divine in nature. Evidence of a sky-god is found throughout the
Celtic-speaking world. So the response of the Celts to Alexander could
well have been the pagan equivalent of ‘We fear nothing but the Lord’.
That may have been a disappointment to a young man driven by a sense
of destiny, but these Celts had no reason to fear him. His eyes were on
the east. Which particular Celtic tribe had sent these sky-fearing
ambassadors to Alexander is not recorded, but when the northern
Balkans emerged more clearly into history, the Taurisci were living in
what is now central and eastern Slovenia and northwest Croatia. So they
are the most likely candidates.29
The Celtic friendship with Macedonia lasted until the death of
Alexander in 323 BC. A delegation of Gauls met him on his march to
Babylon in the year of his death.30 This brilliant commander had welded
together a huge empire, but without his charismatic presence it fell apart.
In its fragmented state it became vulnerable to Celtic expansions. Even
the throne of Macedonia could not be secured for Alexander the Great’s
infant son. In 317 BC Cassander, a distant relative, declared himself
regent. He shored up his own claim to the throne by marriage to
Alexander’s half-sister, and within a few years had executed Alexander’s
son. Meanwhile the Celts were making further inroads into Illyria, pushing
a major Illyrian tribe into the arms of Cassander in 310 BC.31
Macedonia was further weakened on the death of Cassander in 297,
when his sons squabbled over their inheritance. It was seized from them
first by a league of Greek city-states and then by Thrace. Into this chaos
of political rivalry the Celts struck in force. An army led by Bolgios in 281
BC attacked Macedonia and in 279 cut off the head of the foolish king
who had not taken their threat seriously. Macedonia was only saved by a
man named Sosthenes, of humble birth but with the mind of a general,
who raised an army to defend the country.32
Furious that Bolgios had been so easily thwarted by this hero of the
hour, another Celtic commander, Brennos of the Prausi or Tolistobogii,
ravaged Macedonia with a huge force, and led them south into Greece,
intent on plundering the oracle at Delphi. Brennos was defeated at Delphi
not by armies but by natural forces. An earthquake and thunderstorm
were followed by rock falls. The Gauls even fought each other in the
ensuing panic. Brennos advised his men to retreat, killing their wounded
to avoid delay. He led by example. Badly wounded, he drank deep of
unwatered wine and then stabbed himself to death. Greeks harried the
retreat all the way, gradually destroying the desecrators of Delphi until
not one was left alive, or so the Greek authors closest to the period
say.33
 Centuries later the Greek historian Athenaeus of Naucratis thought that
the Scordisci were a remnant of the force that attacked Delphi. Writing
around AD 190, Athenaeus tells us that the leader who brought them to
safety was one Bathanattus. ‘From him also the road by which they
retreated is called Bathanattia, and they call his descendants Bathanatti
to this very day.’34 Though this certainly sounds like reliable local
knowledge of a founding hero, we may doubt that Bathanattus ever saw
Delphi. The drama of Delphi so dominated Greek accounts of the Celts
that it would be easy to assume that he must fit into this well-known story.
The truth was probably more complex. Celts had been moving along the
Danube for some time before the burst of activity around 280 BC. Their
numbers may well have been swollen by an influx of newcomers in that
fateful period, but the hordes of arrivals were not a single fighting force.
Bands from different tribes might unite behind one leader for as long as it
suited them, but split apart just as easily.
The massive army under Brennos divided before it had even reached
Macedonia. Livy tells us that a quarrel arose and as many as 20,000
Celts parted company with Brennos between Thrace and Illyria. They and
their descendants remained a force in southeastern Europe for over half
a century and in Asia Minor until the Roman period. Moving into Thrace,
they carved out a kingdom for themselves in what is now Bulgaria, from
which they threatened Byzantium. This city founded by Greek colonists
on the European coast of the Bosporus had grown so rich on trade that it
could afford to pay massive tribute to keep the Gauls at bay. These
Gauls were massacred by Thracians in 212 BC, but not before they had
sent an offshoot into Asia Minor.35
Across the Bosporus from Byzantium was the kingdom of Bithynia,
which offered an opportunity for mercenaries just as the eager horde of
fighting Celts had arrived on the opposite shore. In about 280 BC
Nicomedes succeeded his father Zipoetes on the throne of Bithynia. He
immediately had two of his brothers put to death, but the third, Zipoetes
II, escaped and challenged his rule. Nicomedes engaged a large force of
these conveniently placed Celts to defeat his brother, in which they
succeeded in 276 BC.36
In Asia Minor these Celts gained the name Galatoi (the Greek version
of ‘Gauls’), which the Romans rendered as Galatae.37 On the loose after
the end of their contract with Nicomedes, they terrorized the whole of
Anatolia west of the Taurus Mountains into granting them tribute. None
dared refuse them until Attalus I of Pergamon (241–197 BC). Attalus saw
himself as the champion of the Greek cities of Asia Minor against these
barbarians, and was victorious in battle against the Galatians.38 [63] By
232 BC the Galatians were contained within an area of Anatolia roughly
centred on the hillfort of the Tectosagi at what is now Ankara.39 Strabo
knew of a tribe of Tectosagi in southwest Gaul near the Pyrenees and
assumed simply from the matching name that here was the origin of the
Galatian Tectosagi.40 This seems unlikely geographically. Celts moving
down the Danube were probably from east of the Rhine. Since Tectosagi
means ‘journey-seekers’, it would fit a band of migrants.

63 Detail of the sculpture of the Dying Gaul showing the face, hairstyle and torc. It is thought to be
a Roman replica in marble of a lost Hellenistic original dedicated to Athena at Pergamon by
Attalus I to commemorate his victories over the Galatians.
 Strabo tells us that the meeting place for the annual assemblies of the
Galatians was Drunemeton (sacred place of oaks).41 There could be no
clearer proof that they were Celtic-speakers. The element -dru (oak) is
the basis for the word druid. Nemeton (sacred place) appears in a
Gaulish inscription. Wherever Celts lived, -nemet-provided a component
of tribal names, gods’ names and place-names.42
Galatia remained an independent state until swept up into the Roman
empire. As the Celtic-speakers closest to Jerusalem at the time of Christ,
the Galatians lay within the scope of the earliest Christian drive to
convert. St Paul visited Galatia during his second missionary journey,43
and some years later (c. AD 51) wrote an impassioned letter to the
churches of the Galatians, preserved in the New Testament.
St Jerome remarked that the Galatians spoke a language almost
identical (apart from a Greek influence) to that of the Treveri,44 among
whom he had studied at their capital Trier (in present-day Germany). He
encountered the Galatians in AD 372/3 on his travels in the Near East.45
The Treveri were Belgae. This is interesting confirmation that the Belgae
spoke Gaulish, for what remnants we have of the Galatian language
appear Gaulish with a Greek overlay.46

Iberia: the tangled skein

The puzzle of how Celtic-speakers arrived in Iberia has occupied a series
of scholars. Influences from Central Europe are limited.47 It has been
suggested above (pp. 98–100) that the oldest layer of Celtic to enter
Iberia came with a late Bell Beaker influx across the Pyrenees into the
eastern meseta c. 2200 BC, which was to emerge in written form as
Celtiberian, while in what is now Galicia the initial contact with Celtic-
speakers may have come through interaction with northwest Europe
between 1300 and 700 BC.
By the Roman period Celtic place-names were scattered far more
widely across Iberia. [see 25] Those ending in -briga (‘hillfort’, ‘high
place’) have attracted particular attention. [64] Many fall within the region
where inscriptions have been found in the Lusitanian language. That
does not mean that Lusitanian must therefore be Celtic. Lusitanian and
Celtic have distinctly different ways of handling what linguists call the
vocalic r. Celtic is unique in converting it to ‘ri’, as in -briga. So it seems
that Celtic had crept into what had been Lusitanian territory. Both there
and in Galicia, place-names ending in the Celtic -briga or -bris can have a
non-Celtic first element, no doubt derived from a pre-existing place-
name. These are no less Celtic than cases where the first element is also
Celtic, such as Nemetobriga (Trives Viejo in modern-day Galicia). The
inclusion of -briga tells us that Celts created the name that was recorded
by the Romans.48

64 Place-names in Iberia ending in -briga, the Celtic word for ‘hillfort’ or ‘high place’. This
distribution shows that Celtic speakers had spread into areas that were previously Lusitanian-
speaking.

So who were the newcomers? The scatter of Celtiberian objects in

western and southwest Iberia suggest that the Celtiberi had been
expanding westwards for several centuries before the Romans arrived.49
In Roman times there was a mint producing typically Celtiberian coins in
Villas Viejas del Tamuja within ancient Lusitania.50 Pliny deduced from
their religious rites, their language and the names of their towns that the
Celtici of the Roman province of Baetica in what is now southern Spain
were descended from the Celtiberians and had come from Lusitania.51
This slightly confusing passage seems to mean that the Celtiberians in
his view had expanded first into Lusitania and then from there to Baetica.
Strabo claims that the Celts living around the River Guadiana, within the
Roman province of Baetica, were kinsmen to those of Gallaecia
(Galicia).52 Certainly there is at least one inscription in Galicia to Reue
Ana Baraego, the god of the Guadiana and Albárregas rivers. There are
also Latin inscriptions to individuals from Celtiberia, evidently arrivals in
Roman times.53
The last wave of Celtic-speakers to arrive were Gauls, though the
impact they had is hard to judge. A group of place-names in Iberia end in
-dunum (meaning ‘enclosure’ or ‘fort’), which is commonly found in Gaul.
However, the underlying word -duno is Celtic rather than specifically
Gaulish, so we cannot be certain that all of these mark the trail of a
Gaulish influx. On the other hand some place-names ancient and modern
leave no doubt of the existence of Gauls in Iberia, such as Forum
Gallorum, located probably close to modern-day Gurrea de Gállego in
Aragon. Likewise ancient Gallicum was probably on the Roman road
near present-day San Mateo de Gállego in Aragon. Gallica Flavia might
be modern Fraga in Aragon.54 This cluster of names in Aragon suggests
that some Gauls at least had crossed the Pyrenees.

La Tène in the British Isles

The La Tène culture spread deep into Britain soon after it arose on the
Continent. [65] Many archaeologists have preferred to explain this by
trade contact rather than migration, but that scarcely explains the
linguistic similarity of Gaulish and Brittonic. Movement there clearly was.
Climate change seems likely to have played a part in this. As already
mentioned above (p. 124), 650–450 BC was a warm, dry period in
Continental Europe. High crop yields could have fed a burgeoning
population. Britain might have seemed inviting around 400 BC, as
conditions improved there.55
 Ireland had become a backwater. To judge from the amount of
archaeological material, productivity hit a peak in 1000–900 BC and then
began to fall. During the early Iron Age (c. 800/750–400 BC) evidence of
human occupation is thin in the Irish landscape. This does not mean that
the population completely disappeared. Modern archaeological methods
reveal more subtle signs of activity, even if magnificent metalwork is no
longer to be seen.59 Yet the conclusion is inescapable that Ireland had
sunk into a poverty that would not support large numbers. Decline began
before the end of the Bronze Age, so we cannot simply blame a fall in
international demand for bronze or its raw materials. Naturally, Ireland
would have lost its pioneer advantage as bronze-making spread across
western Europe. The same is true of Britain. It is even possible that Irish
tin and gold were panned out, for the gold in objects from Iron Age
Ireland does not seem to come from the Mourne Mountains.60 Looming
over the period though are the grey skies of climate shift. The change to
a cooler and wetter climate around 800 BC would have had an impact on
farming.

65 The pommel of a sword hilt in the form of a male head, found near Cirencester, western
England, in 2006. It is in the Late La Tène style, but unique in its depiction of an elaborate
hairstyle of braids forming a loop. The face has lentoid eyes and a wide moustache above a slit
mouth.

Genetics: traces of the spread of the Gauls?

The distribution of Y-DNA haplogroup R1b-U152 is similar to that of the Urnfield culture, but
also has features suggestive of the Iron Age movements of the Gauls. [see 60] The density
of R1b-U152 is greatest in northern Italy and Corsica and radiates out from there.56 [66]
This pattern is interesting. A radiation in all directions from a high density centre is what we
would expect if a mutation occurs within a comparatively static population. With no mass
movement in one particular direction, the mutation will percolate gradually outwards from its
 origin point.57 R1b-U152 may have fanned out initially within Urnfield. This would not
necessarily be exclusively with Celtic-speakers. A link between Ligurians and U152 has
been suggested in Italy,58 and might explain some U152 in Iberia.
If we picture a gradual seepage of this haplogroup across Gaul in the Bronze Age, then it
would naturally spread with subsequent movements of Gauls. The presence of R1b-U152 in
central Anatolia is interesting, since part of that region was colonized by Gauls.
Given its concentration in Italy, the same haplogroup probably spread within the Roman
empire at a later date. The discovery of subclades of U152 offers promise that one day we
might be able to distinguish between one hypothetical movement and another, by testing
ancient DNA.

66 Y-DNA R1b-U152 radiates from a high density core in northern Italy and Corsica.

There is a sharp increase in activity in Ireland at around 400 BC,

followed by stability until the end of the millennium.61 Population growth
may have begun as local farmers enjoyed a warmer, drier climate. It
seems no coincidence though that the first La Tène objects appear in
Ireland by around 300 BC. Migrants would find a warmer Ireland more
appealing. The early metalwork imports have parallels on the
Continent,62 though this does not mean that they arrived direct from
there. People entering Britain from Gaul would no doubt bring treasured
objects with them, which might still be in the possession of those of their
descendants who chose to move to Ireland. The fact that La Tène is
largely confined to the northern two-thirds of Ireland generated the notion
that it arrived from northern Britain. A seminal study of beehive quern
stones in Ireland confirmed it. These querns are linked to the La Tène
culture by the decoration on some of them, and match the type of
beehive quern found in the region of Britain ranging from Yorkshire to
southern Scotland.63
If an influx from Gaul had already caused the famous sound-shift to P-
Celtic in Britain (see pp. 115–16), then newcomers from Britain would
have taken it to Ireland. It seems that they did. By the time Ptolemy wrote
his Geography in the 2nd century AD there were both Gaelic and Brittonic
place and tribal names in Ireland. [67] Not all the Brittonic names need
date as far back as 300 BC though. Ptolemy was writing after the Roman
conquest of Britain, which seems to have driven some unhappy Britons
across the Irish Sea (see pp. 155–56).64
On the face of it, Ptolemy’s Ireland bears little resemblance to the
prehistoric division of Ireland into five provinces (the Pentarchy) which
can be deduced from native sources.65 We can detect in Ptolemy the
familiar name of only one of these. The Ouolountioi (in Ptolemy’s original
Greek) lived in northeast Ireland. They appear later as the Ulaid, who
gave their name to the province of Ulster. We have already seen the
archaeological evidence for this province (see p. 18). The Iwerni (Iouerni
in Ptolemy’s Greek) tribe is placed in southern Ireland, and appears in
later sources as the Érainn, a group of dynasties who ruled in Munster
prior to the 7th century AD. The town named after them by Ptolemy could
be Teamhair ‘Erann, which appears in the Ulster Cycle as the muster-
place for the Érainn.66 The Iwerni were once the people of all Ireland, as
recorded by ancient Greek travellers centuries earlier (see p. 46). So it
seems that they had been squeezed into the south by incomers from the
La Tène period onwards.
67 The Celtic tribes of Ireland as noted by the geographer Ptolemy in the 2nd century AD. The
Brigantes of southeast Ireland were probably an offshoot of those in Britain.

Genetics: traces of the spread of La Tène to Ireland?

A likely genetic signature of La Tène in Ireland is Y-DNA R1b-M222 [see 102] carried by up
to 44 per cent of men in parts of Northern Ireland today.67 Given the drastic drop in
population in Iron Age Ireland, it would not have taken many incomers to have made such a
strong genetic impact. A study restricted to the counties forming the Republic of Ireland
found 20 per cent of men in Donegal were R1b-M222, much higher than in other parts of
the Republic. Northwestern Ireland was supposed to be the territory of the Northern Uí Néill,
descendants of the fabled 5th-century warlord, Niall of the Nine Hostages. It also appears
among the Connachta, supposed descendants of the brothers of Niall. So R1b-M222 was
initially labelled as the lineage of Niall.68
Alas this attractive idea rested on genealogies that were tampered with around AD 730 to
make the famous Niall the ancestor of unrelated kings based in Donegal, who then claimed
to be the Northern Uí Néill.69 Wider sampling subsequently showed the highest
concentrations of R1b-M222 in northeastern Ireland (Belfast 44 per cent) and western
Ireland (Mayo 43 per cent). Outside Ireland roughly 10 per cent of men carry M222 in
northern England (Yorkshire), western Scotland (Skye) and northeastern Scotland
(Moray).70 This is not the pattern that we would expect from Irish migrants into Britain.
 Recently a parent to R1b-M222 has been slotted into the ‘family tree’, defined by the
marker Z2961. It is too new a discovery for its distribution to be known, but one carrier has
the Welsh surname Powell.71 A working theory would be that M222 arose among the Celts
in Britain. Ancient DNA may one day settle the matter.

Belgae in Britain
In the two centuries before the Claudian invasion of Britain, the south of
the country was subjected first to raiding and then to settlement by an
earlier wave of invaders, the Belgae. Caesar learnt in 54 BC that Belgae
from Gaul had settled along the coast of Britain, many retaining the same
tribal names as their brethren across the Channel.72 This is compatible
with the archaeological evidence, if we are generous in our interpretation
of ‘the coast’. From 125 BC Gallo-Belgic coins appear over the whole of
southeastern Britain. [68] New tribal centres appeared, similar to those in
Gaul. These were large, fortified, lowland settlements. Among their
inhabitants were craftsmen making the first British wheel-thrown pottery
and minting the first British coins.

68 The earliest Belgic coins appear on both sides of the Channel. The first and most spectacular
type is shown here: dating from 125–100 BC this gold stater is identified with the Ambiani, since it
is found in their Somme Valley territory.
 Tribal coin issues and their distribution add to our knowledge of the
tribes of Britain.73 [see 74] The Atrebates had a centre at Calleva
(Silchester). They share a name with a tribe of the Gaulish Belgae living
in the neighbourhood of Arras, northern France. The Catuvellauni lived to
the north of them, with a town at Verolamion (St Albans). A minor tribe of
the same name lived in Gaul in the valley of the River Marne; they gave
their name to the town of Catalaunum (Châlons-sur-Marne).
Archaeologically, the Aylesford-Swarling culture is common to the British
Catuvellauni, their eastern neighbours the Trinovantes, with their tribal
centre at Camulodunon (Colchester), and the Cantiaci of what is now
Kent. It has parallels with that of the Continental Catuvellauni.74 So we
can confidently include the Trinovantes and Cantiaci among the Belgae
in Britain. It was the Belgic tribes who took the first shock of Roman
invasion.

Overview
• From c. 400 to c. 100 BC, the Gauls/Galatians expanded in many
directions. Their reasons for this may include pressure from the
Germani, overpopulation and desire for a better standard of living.

• East of the Rhine, Celts were under pressure from the expanding
Germani. Belgae moved into northeastern Gaul and southern
Britain. The Boii of Bohemia moved east, west and south.

• Pressure from incoming Belgae meant that some Gaulish tribes

were displaced, for example the Parisii, who seem to have moved
west to the region around present-day Paris and also to Britain.

• Brennos ravaged Macedonia with a huge force of Gauls and led

them south into Greece, intent on plundering the oracle at Delphi.

• A group split away from Brennos before he even reached

Macedonia. They and their descendants remained a force in
southeastern Europe for over half a century and in Asia Minor until
the Roman period.
• In Iberia the Celtiberians expanded west across much of the
peninsula.

• The La Tène culture spread into Britain and from there to the
northern two-thirds of Ireland.

• The expansion of the Celtic-speaking world had reached its greatest

extent in around 100 BC. It was soon to be challenged by the rising
power of Rome.
CHAPTER EIGHT

Celts vs Romans

Is there anyone on earth who is so narrow-minded or

uninquisitive that he could fail to want to know how and thanks
to what kind of political system almost the entire known world
was conquered and brought under a single empire, the empire
of the Romans, in less than fifty-three years?1

Polybius wrote this before the Roman conquest of Gaul and Britain. For
him the known world surrounded the Mediterranean. At its greatest extent
in the 2nd century AD, the Roman empire swallowed up all the Celtic-
speaking peoples in Europe and Asia except those of Ireland and
northern Britain. [69] It was indeed a wonder, even more so because it
survived more or less intact for centuries.
69 The Roman empire at its height encased the Mediterranean, making the Romans masters of
most of the world known to previous Mediterranean mariners, Greek and Phoenician.

Cisalpine Gauls
The rise of the Romans first put the Cisalpine Gauls in peril. Polybius was
well informed on the progress of the Roman push northwards, which was
within living memory in his day. The Senones were ejected in 295 BC and
lost their land to Roman colonists. The Boii, fearing a like fate, implored
the aid of Etruscans to attack the Romans, but were defeated and had to
come to terms. There followed 45 years of peaceful co-existence, until a
new generation had arisen that had no personal memory of defeat. The
Boii and the Insubres called for support from fellow-Gauls across the Alps
and marched on Rome. They were subjugated in 224–226 BC. Seeing
which way the wind was blowing, the Cenomani threw in their lot with the
Romans.2
 The wind seemed to veer in the opposite direction when the mighty
Carthaginians challenged Rome in the Second Punic War (218–201 BC).
At the time it must have seemed to the Cisalpine Gauls an excellent
opportunity to rid themselves of the upstart Latins. Rome might be flexing
its muscles, but Carthage was a long-established power in the
Mediterranean. Founded by Phoenician traders from Tyre, Carthage had
planted colonies of its own along the coast of North Africa by 550 BC,3
and at its height headed a Punic empire encompassing North Africa and
parts of Iberia, Sicily and Sardinia. Great was the fury of the Carthaginian
general Hamilcar Barca when he was forced to yield Sicily to the Romans
in 241 BC. Rome’s seizure of Sardinia in 237 BC was a second blow.
Hamilcar resolved upon retaliation and had his nine-year-old son
Hannibal solemnly swear lifelong enmity with Rome. Though Hamilcar
died before he could accomplish his aims, he ensured an heir just as
committed to them.4
Hannibal cherished high hopes of the Cisalpine Gauls as allies, given
their bravery in war and hatred of Rome. He secured their support,
including that of the Cenomani, before setting out on his extraordinary
journey with an army complete with elephants, starting from Cartagena in
Iberia. From the River Rhône onwards, he had Cisalpine Gauls as
guides. Surviving the crossing of the Alps, Hannibal in the spring of 217
BC crushed the Roman army in the battle of Lake Trasimene. In the
months that followed he teetered on the edge of total victory.5 The
Romans meanwhile chewed at the root of Carthaginian power in Iberia
and North Africa. It was the success of the Roman general Scipio
Africanus in Africa that would bring Carthage to peace talks and the recall
of Hannibal from Italy.
The implications for the Cisalpine Gauls were profound. It was only a
matter of time before they were absorbed into Rome’s nascent empire.
Yet their fate was less lamentable than they might have expected. The
Cenomani were defeated in 197 BC and the Insubres and Boii in 194 BC.
The Boii were scattered. The Cenomani were disarmed in 187 BC. The
Insubres retained their arms, but were stripped of the small empire they
had acquired by their conquest of other local tribes. So the Cisalpine
Gauls were humiliated, but they were not completely ejected from Italy. It
was more profitable to the Romans to leave pacified tribes in charge of
their own land, but in league with Rome. The surviving Cisalpine tribes
enjoyed a degree of autonomy until the region became a Roman
province in 81 BC. Perhaps the aggression that the Insubres once
expended in battle was channelled into commerce. They developed a
bustling commercial hub at Milan.6

The struggle for Iberia

During the Second Punic War, the Romans undermined Hannibal by
attacking the base of his power in the Iberian peninsula. This starved
Carthage of a source of wealth and warriors. The Roman conquest of
what they called Hispania was a slow process. It began in urgent self-
defence in 218 BC and ended in 19 BC as part of a policy of imperial
expansion. [70]
The immediate rationale for the Romans to throw their weight behind
Iberian resistance to Carthage was their friendly relationship with the
walled town known to the Romans as Saguntum (now Sagunto),
approximately 30 km (20 miles) north of Valencia. This town overlooked a
fertile plain and was conveniently close to the coast. It had flourished on
trade. Its citizens were Iberes, who spoke a non-Indo-European
language. The Mediterranean coast of Iberia which had once been home
to Ligurians had gradually been taken over by the literate and urban
Iberes.7 As Hannibal was advancing towards them, the citizens of
Saguntum sent repeated pleas to Rome for aid. In return the Romans
merely sent legates, who protested against any attack on Saguntum.
Undeterred Hannibal took the town just before he left for Italy. The
Romans had their casus belli.8
Roman troops arrived in Spain under the command of the brothers
Gnaeus and Publius Cornelius Scipio. Hannibal had taken the sons of
many leading Iberes as an assurance of their support, housing these
hostages within the walls of Saguntum. A wealthy and prominent Iberian
named Abilyx managed to rescue the children by deftly playing politics.
As the Scipios approached, Abilyx persuaded the local Carthaginian
commander that it was no longer possible for the Carthaginians to control
the Iberians by fear, as the Romans might soon have their hands on the
hostages. What was needed was a gesture of goodwill. Since Abilyx had
previously been seen as loyal, he was allowed to whisk away the
hostages under cover of darkness. He took them straight to the Roman
camp. There he persuaded the Scipios that returning the children to their
parents with Roman blessing would gain the friendship of the Iberes.
Indeed it did make a good impression.9

70 The gradual Roman conquest of the Iberian peninsula, also known as Hispania. The Roman
provinces of Hispania changed their borders as more territory was assimilated.

After some successes against the Carthaginians, both Gnaeus and

Publius Scipio were killed in 211 BC. The Romans laid the blame on the
Celtiberian mercenaries the Scipios had recruited, who were lured away
by the Carthaginians. The Roman Senate appointed another Scipio to
take command in Spain, the son and namesake of Publius Cornelius
Scipio, eager to avenge the deaths of his father and uncle.10 His daring
exploits in Spain and later against Carthage itself were to make him
famous as Scipio Africanus or Scipio the Great.
Planning in secret, the young Scipio prepared for a surprise attack on
the heart of Carthaginian power in Spain – the city of Cartagena. He took
it at speed in 209 BC. In Cartagena too the Carthaginians had been
holding Iberian hostages. Learning from previous experience, Scipio
restored them to their homes, gaining yet more support from the Iberes.11
By comparison with some later Roman campaigns, it was a charm
offensive.
A Carthaginian general named Hanno then arrived in Spain with a
fresh army and marched into Celtiberia to raise troops there. Scipio sent
a commander against him who first attacked the unprotected Celtiberian
camp, where the newly gathered levies were not expecting trouble so
soon. The result was a resounding victory for the Romans. One such
success followed another until the Carthaginians were driven from Spain
and Scipio returned to Rome in triumph.12
This could scarcely be the end of the story. Scipio is alleged to have
told his men ‘Our purpose and endeavour is not that we may remain in
Spain ourselves, but that the Carthaginians may not.’13 Yet a Roman
army had to be left in Spain to prevent the return of the Carthaginians.
The Romans were acquiring an empire, and at the expense of the Celtic
peoples.
Rome gradually extended its Hispanic territory in the 2nd century BC,
and shaped a new social order. The Romans had already begun to make
slaves of selected prisoners during the early campaign among the Iberes.
When Scipio took Cartagena, he made a class distinction between its
citizens and its working men. The former were granted their freedom. The
working men were told that for the time being they were public slaves of
Rome, but if they showed goodwill and industry in their crafts, they would
have their freedom if the war against Carthage ended in Roman victory.
Thus he ensured that even his captives would be zealous in the Roman
cause.14
As time went on, the Romans developed a policy of taking as slaves
those who resisted the Roman expansion. No doubt this was intended to
encourage peaceful co-operation in Rome’s imperial designs, but it also
produced a supply of slave labour that became crucial to the imperial
economy. Thousands of male and female slaves were shipped back to
Rome for sale. Celts might find themselves tending Roman fields or
dressing the hair of Roman ladies. After generations in Roman servitude
their very origins might be forgotten. They were Romanized just as
effectively as those chiefs coaxed to become clients of Rome by gifts and
favours.
 One figure stands out as a local hero of the resistance. Viriatus (or
Viriato) was a Lusitanian guerrilla leader whose memory remains green
in Portugal and Spain.15 Some Roman officials were becoming rapacious
in the arrogance of power. Those appointed to short-term posts had little
incentive to learn about the locals or build up a good relationship with
them. There was a grumbling resentment in the Roman province of
Hispania, which found an outlet in 171 BC in a deputation to the Roman
Senate, begging that the allies of Rome should not be treated more
shamefully than their enemies.16 The popularity of Viriatus among later
generations has led to an accumulation of legend around him. What we
know scarcely needs embellishment. Viriatus was remarkably strong,
healthy and agile, having been a shepherd on the mountains from his
childhood. For most of his life he lived under the open sky and was
satisfied with nature’s bedding. He was thus inured to spartan living. He
always carried weapons and was famed for his deeds of arms, but it was
his intelligence and tactical ability that lifted him to leadership. At first he
led a group of bandits. Progressing from raider chief to warlord in the
140s BC, he was successful in battle against the Romans. In the end the
Romans resorted to arranging the assassination of this charismatic figure
by some of his own kinsmen.17
Julius Caesar served as governor of Hispania from 62 to 60 BC. He
took Rome’s writ into the far northwest of the peninsula by conquering
the Gallaeci and Lusitani. It was here that he was first acclaimed
imperator by his troops,18 an honorific title for military commanders at the
time, but also perhaps a poignant foreshadowing of the imperial title his
heirs were to seize.

Caesar’s conquest of Gaul

Both Romans and Gauls were politically in transition. Most of the Gaulish
tribes were still led by kings, but some had abandoned hereditary
monarchy in favour of government closer to the Roman model.19 The
Romans had replaced their monarchy by elected officials long before. It
was still possible in emergencies for a dictator to be appointed for a
limited period, who had absolute control, civil and military. Julius Caesar
would be granted almost regal powers a few years after his Gallic
campaign.20
Before that he was elected as one of Rome’s two consuls for 59 BC.
The post of consul was for one year. By this time it was usual for the
consul to be rewarded at the end of his term by a lucrative post as a
provincial governor. Caesar gained Roman Gaul, together with
Illyricum.21 Both Cisalpine Gaul and Gallia Narbonensis were already
Roman provinces. The latter had been acquired in part to provide a land
corridor for troops from Italy to Hispania.
Caesar had no brief and no immediate plan to conquer the rest of
Gaul. Yet he would have been well aware of the enormous prestige and
popularity that would accrue to a commander who did so. Gaul was a
prize worth fighting for, with its swathes of good agricultural land blessed
with sunshine to ripen corn and grapes, yet well-watered by clouds from
the Atlantic and the Mediterranean. Given a suitable pretext first by the
Helvetii and then by the Aedui, Caesar seized the prize. [71]
The Helvetii were a warlike Gallic tribe living in what is now
Switzerland, who had been under pressure for half a century from the
expansion of the Germani. They determined to migrate en masse and
conquer the rest of Gaul. As it chanced, their plan matured in Caesar’s
first year as governor of Roman Gaul. The Helvetii were joined by several
neighbouring Celtic tribes, including a remnant of the Boii who had
migrated westwards across the Rhine not long before. The leaders of this
swollen horde decided to take the broadest track, which was through the
Roman province in southern Gaul. Learning of their intentions, Caesar
raced from Rome to confront them at Geneva.22
Caesar’s first action was to destroy the bridge across the Rhône at
Geneva. The Romans have so often been credited with bringing a
transport infrastructure to the Celtic lands that it needs to be said that a
system of roads and bridges already existed in Gaul. Caesar
encountered roads in Britain too. In Ireland, which was never engulfed by
the Roman empire, the early medieval law texts made provisions for
public roads.23 What Roman civil engineers achieved was a system of
stone bridges and straight, metalled roads built to withstand heavy traffic.
Legions on the march could travel at speed along these imperial arteries.
In the face of Caesar’s opposition, the Helvetii abandoned their plan to
pass through the Roman province and took another route, but Caesar
pursued them anyway and forced them to return to their former territory.
There they would continue to be a useful buffer between the Germani
and Roman Gaul.24
This demonstration of the power of Rome was not lost on other Gallic
tribes. Their conquest by the Helvetii had been averted. Could Rome
resolve another pressing problem? For many years the Aedui had been
locked in a struggle for supremacy with their neighbours the Arverni and
Sequani. The Aedui lived in what is now central France, between the
Loire and the Saône rivers, with their capital at Bibracte. They had long
been in the ascendant and in alliance with Rome. To their southwest
lived the Arverni in the Auvergne mountains. To their east beyond the
Saône lived the Sequani, with whom they were in conflict over tolls on
traffic along the Saône. In the 60s BC the Arverni and Sequani had taken
the reckless step of hiring thousands of Germanic mercenaries. This
certainly achieved their aim of defeating the Aedui, who were crushed in
61 BC. Yet a worse fate befell the victorious Sequani. The Germanic
mercenaries acquired a taste for the Gallic standard of living. The
productive land of Gaul supported a way of life unknown in the Germania
of their day. The chieftain Ariovistus settled in the territory of the Sequani
and seized a third of their land. He soon invited many thousands of
Germani to Gaul and demanded that more Sequani evacuate their land
to make room for these new settlers.25
71 Caesar’s campaigns in Gaul. The southern part of the lands of the Gauls had been taken by
the Romans before Caesar’s day. As governor of Cisalpine Gaul and Gallia Narbonensis, Caesar
was drawn into the politics of the warring Gaulish tribes and ended by taking the rest of Gaul into
the Roman empire.

So Caesar found himself in an odd position. He received a request for

aid in 58 BC from both the Aedui and their former enemies. This gave him
one reason for war. The underlying threat of the encroaching Germani
was a more powerful one. The Gauls themselves feared complete
takeover. That would leave the way clear for the Germani to fall upon
Roman Gaul and enter Italy. Thus Caesar felt justified in his campaign
that same year to expel Ariovistus from Gaul.26
It was bound to occur to some Gauls that the price of Roman aid could
be Roman domination. Reaction was not slow in coming. Rumours
reached Caesar that winter that the Belgae were conspiring against him.
Caesar tells us that ‘They were afraid that if all the rest of Gaul was
subdued, our troops would advance against them.’27 It may be that they
expected to be ejected from Gaul along with those Germani who had
settled there, since they were all intruders from east of the Rhine. The
Belgae were certainly willing to ally themselves in this campaign against
Rome with those Germani who had managed to cross the Rhine.28
Caesar lost no time in responding. He raised another two legions in
Italy and sent them into Gaul in the spring of 57 BC. This brought
Caesar’s forces up to eight legions, each nominally of 6,000 men. So
rapid was their march that they surprised the Remi, the first Belgic tribe
encountered, who promptly sided with Rome. The Remi provided Caesar
with information on the Belgic fighting forces. The total number of men in
the Belgic and Germanic coalition opposing Caesar was calculated at
298,000, compared to his 48,000.29
72 This larger than life-size figure of a Gaulish warrior (1st century BC) was found at Vachères in
southeastern France, part of the territory held by the Romans before Caesar’s conquests. His
accoutrements, including the expensive coat of mail, are those common among auxiliaries in the
Roman army.

There followed a master class by Caesar in military tactics. Even if the

estimate of the opposing forces was exaggerated, the Romans were
clearly outnumbered. On the other hand, neither the Celts nor the
Germani had standing armies of full-time soldiers. The massive coalition
army was a temporary alliance. Caesar kept the Belgae and their allies
busy with daily skirmishes, but avoided a full-scale battle. The Belgae
could not maintain a massive, hungry army in the field indefinitely. Corn
supplies ran low. The decision was made for every man to return home
and await events. This retreat allowed Caesar to conquer the Belgic
tribes piecemeal over the course of that year. Meanwhile, the Veneti and
other Gallic tribes of the Atlantic seaboard had been subjected to Roman
rule by a commander sent by Caesar with just one legion. So Caesar
considered Gaul to be pacified at the end of 57 BC.30 [72]
Not all Gauls tamely accepted this. Rebellion followed rebellion. This
restless mood threw up a leader in 52 BC whose name has echoed down
the centuries – Vercingetorix. Caesar describes him as ‘a powerful young
Arvernian, whose father Celtillus had held suzerainty over all Gaul, and
who had been put to death by his compatriots for seeking to make
himself king’.31 Here was a man bred to leadership and politics.
Vercingetorix was determined to free Gaul. He had no trouble rousing his
own retainers to action, but the Arvernian elders refused to take the risk
of rebellion and expelled him from their oppidum at Gergovia. Undaunted,
he raised a rag-tag band from the countryside around, expelled his
opponents and was proclaimed king of the Arverni. Roman rule was
unpopular enough to secure him the immediate support of other tribes of
central and western Gaul.32
In the Asterix series of comic books set in Roman Gaul, a running joke
is that all Gauls recall the battle of Gergovia (won by Vercingetorix), while
no one has a clue about the siege of Alesia (where Vercingetorix
capitulated). The reality is more complex. Vercingetorix has been
celebrated as much in defeat as victory. As he saw his men slaughtered
by the Romans at Alesia and defeat was inevitable, Vercingetorix
addressed an assembly:

I did not undertake the war for private ends, but in the cause of
national liberty. And since I must now accept my fate, I place
myself at your disposal. Make amends to the Romans by killing
me or surrender me alive as you think best.33
73 Vercingetorix Throws Down His Arms at the Feet of Julius Caesar. This painting by Lionel Noel
Royer (1899) shows the scene described by Plutarch.

Consultation with Caesar brought the request back to Alesia for the
arms and chiefs of the Gauls. As Plutarch tells the tale, Vercingetorix put
on his most beautiful armour and had his horse carefully groomed. He
then rode out through the gates to where Caesar was seated, where he
leapt from his horse, stripped off his armour and sat at Caesar’s feet.34
This was the sacrifice that turned him into a tragic hero. [73]
Vercingetorix no doubt knew the Roman habit of parading captured
enemies through the streets of Rome in triumphal processions, before
executing them. This was indeed to be his fate. It was the end of
resistance to Rome in Gaul.

Britannia
The Romans carved a province for themselves out of most of Britain,
which they called Britannia. In a sense the Romans created what would
become Scotland, for until their arrival there had been no division of
Britain in this way. One could scarcely call it undivided though. Celtic
tribes jostled each other for territory. Indeed these divisions aided the
Romans in taking over southern Britain.
The powerful and pugnacious Catuvellauni were a danger to their
neighbours. Their king Cassivellaunos had killed the king of the
Trinovantes, whose son Mandubracios fled to Gaul and placed himself
under Caesar’s protection. Caesar did not use this as his excuse for the
invasion of Britain, however. Instead he claimed that in almost all his
Gallic campaigns, the Gauls had received reinforcements from Britain.
Though that is more than likely, it scarcely warranted such a response.
Caesar was thriving on the glory of conquest. In the late summer of 55 BC
he set out to reconnoitre, knowing that the fighting season was too
advanced for a full campaign. He was defeated as much by the weather
as by the ferocity of the Britons. The following spring he set sail for Britain
again. The Britons were prepared. They had appointed Cassivellaunos
their war leader. Though he had previously been continually at war with
other British tribes, the threat of Roman invasion had created a common
cause. The Trinovantes had not forgotten their injury though. They
preferred to seek the support of Caesar. Seeing that they were well
treated by the Romans, several other tribes chose the same path. A
Roman assault on the stronghold of Cassivellaunos brought him to terms.
Caesar did not attempt to garrison the island. He was satisfied to take
hostages, fix an annual tribute to be paid and sail away.35 As Tacitus
puts it, Caesar pointed Britain out, rather than handed it over, to
posterity.36
74 The Roman campaigns under Aulus Plautius focused on the commercially valuable southeast
of Britain.

It was Caesar’s great-great-grand-nephew Claudius who conquered

Britain. He was no great military commander, but he did not need to be.
He selected competent men for that role. The most famous today is
Vespasian, who was to become emperor himself in AD 69. When
Claudius was raised to the purple in AD 41, he appointed Vespasian as
legate of the 2nd Augustan legion, at that time stationed near the Rhine.
It was well placed to participate in the invasion of Britain in AD 43, along
with three other legions. Overall command was given to Aulus Plautius.
[74]
The relentless expansionism of the Catuvellauni was still alienating
other Celtic tribes to Roman advantage. By AD 43 the Catuvellauni had
taken over the Trinovantes, the Atrebates and part of the Dobunni. Two
British kings had taken refuge with emperor Augustus. One may have
been Tincomaros, grandson of Commios of the Atrebates. His coins
cluster on the south coast around Chichester. So he would have been the
king who received support from a Roman military presence (arriving
between 10 BC and AD 10), remains of which have been found near
Fishbourne Roman palace.37 From around AD 10 Virica, another
grandson of Commios, began to issue coins. He was ousted from Calleva
(Silchester) in about AD 25 by a brother of Cunobelinos, king of the
Catuvellauni.38 [75]

75 Struck bronze coin of the British ruler Cunobelinos, AD 10–41/2. The obverse has the name
CVNOBELINVS with a beardless helmeted head looking right. The life story of Cunobelinos was
mangled by Geoffrey of Monmouth, then further fictionalized by Shakespeare in Cymbeline.

Then Caratacos (Caratacus), son of Cunobelinos, conquered the entire

kingdom of the Atrebates after AD 40, driving Virica to seek refuge and
allies in Rome. It was Virica who persuaded Claudius to send a force to
Britain. By this time Cunobelinos was dead, so it was his sons Caratacos
and Togodumnos who led the initial resistance to the Roman invasion.
Putting these kings to flight, Aulus Plautius advanced westwards. Those
of the Dobunni under the Catuvellaunian yoke were willing to welcome
the Romans. South of the River Avon, among the southern Dobunni and
the Durotriges, Vespasian faced hard fighting.39 This was a region worth
fighting for, with sources of lead and silver in the Mendips, which the
Romans went on to exploit. Further south again was the famed tin of
Cornwall. Tacitus counted the gold, silver and other metals of Britain as
the reward for victory.40
The conquest of the Celtic tribes of Britain was piecemeal. [76] To the
north of the initial border of the Roman province lay the extensive territory
of the Brigantes, stretching over most of what is now northern England.
At the time of the Roman invasion they were led by Queen Cartimandua,
who preferred the Romans to the Catuvellauni. When Caratacos sought
her protection, he was put in chains and delivered up to the Romans,
nine years after the start of the conquest. This was a major coup for
Claudius. The fame of Caratacos had spread even to Italy. Crowds lined
the streets of Rome to see the man who had defied imperial power for so
long. He was paraded in triumph with his wife and brothers, who had
been captured earlier. Such was the eloquence of Caratacos as he stood
before Claudius that he and his family were pardoned and released from
their chains.41
Thus Cartimandua purchased Roman favour and became a client
queen. After the removal of Caratacos, her husband Venutios was
regarded as the most competent British commander. So divorcing him
was Cartimandua’s gravest error. In his place she took Vellocatos, his
former armour-bearer, not only in marriage, but also to share in
governing the realm. This scandal rocked her household to its foundation.
The tribe favoured Venutios.42
War broke out between Cartimandua and Venutios. In desperation
Cartimandua called for Roman support. The Romans rescued her
person, but not her throne, which was lost to Venutios, now an enemy of
Rome. He was defeated by Petilius Cerialis, governor of Britain AD 71–
73, who took most of the territory of the Brigantes.43 It seems likely that
some of the Brigantes fled to Ireland at this time, for the tribe was
recorded by Ptolemy there. [see 67] Certainly, a group from what is now
northeast England settled briefly around this time on the small island of
Lambay, off the coast of Co. Dublin. In the next century a woman was
buried in classic Roman style at Stonyford, Co. Kilkenny.44 This lies not
far from the Bronze Age hill fort at Freestone Hill, which in the Iron Age
housed a Romano-British sanctuary.45 Here we have evidence of a
Romano-British presence in the area indicated by Ptolemy as the territory
of the Brigantes.
There may have been other refugees from the Romans. The Fir
Domnann appear in the Lebor Gabála Érenn as among the invaders of
Ireland. (Fir means ‘people’.) They were probably related to the Dumnonii
of southwest Britain and what is now the western Scottish Lowlands. The
name occurs in Inber Domnann (Malahide Bay, Co. Dublin), and in
northwest Mayo as Iorrais Domnann (Erris, Co. Mayo) and the nearby
Mag Domnann and Dun Domnann. An early Irish poem describes one of
their leaders as the over-king of Leinster.46
Agricola, governor of the province from AD 77 to 84, pushed the frontier
north to the Forth–Clyde line in AD 79–80. Indeed he drove deeper into
Caledonia with the aim of spreading Roman rule over the whole of
Britannia from shore to shore.47 The logic is understandable. To capture
the whole of an island means that nature provides its borders. On the
other hand the Caledonian highlands were inhospitable, held nothing that
the Romans wanted and would be a drain on manpower to keep pacified.
Subsequent Roman commanders preferred a divided island with a man-
made frontier. In AD 122 the emperor Hadrian ordered the construction of
a wall running for 120 km (75 miles) between the Solway and the Tyne.
Just 20 years later the emperor Antoninus Pius decided to move the
frontier up to the Forth–Clyde line first secured by Agricola. It was fortified
by a bank and ditch, known rather misleadingly as the Antonine Wall.
This was abandoned in favour of Hadrian’s Wall after Antoninus Pius
died.
The incoming Romans imposed their rule partly through existing tribal
organization. A tribe would become a Roman civitas. An existing
oppidum could be Romanized. For example Calleva became the
Romano-British town of Calleva Atrebatum (Silchester). There was some
reorganization. Part of the land of the Atrebates was hived off for a new
polity labelled the Regni (the people of the king). This evidently refers to
the status that Togidubnos (perhaps a relative of Virica) enjoyed as a
client-king after the Roman conquest, with his palace at what is now
Fishbourne. Tacitus cynically saw it as an example of the long-
established Roman custom of employing even kings to make others
slaves.48
76 The Roman conquest of Britain took several decades. In the end the Romans abandoned the
attempt to complete it by taking the Scottish Highlands.

Roman understanding of the Celtic tribes beyond Hadrian’s Wall was

more sketchy. Several tribal names were recorded there by Ptolemy
around AD 150, but we hear no more of most of them. One tribal name
that does recur is that of the Caledonii (Caledonians). If we translate
Ptolemy’s directions into modern terms, they lived in the Great Glen
around Loch Ness and the highlands to the south of it as far as the Firth
of Clyde. Their name is preserved in the place-name Dunkeld, which in
the 9th century was called Dun Chaillden (fort of the Caledonii).49 The
Caledonians became so notable in fighting the Romans that the latter
sometimes called the whole of north Britain beyond their border
Caledonia.
In AD 208 the ‘barbarians’ north of the Roman border were overrunning
Roman Britannia, looting and destroying. The emperor Septimius
Severus arrived in person to drive them back. These Britons were
described in Herodian as avoiding clothing, because it would obscure the
tattoos on their bodies consisting of coloured designs and drawings of all
kinds of animals.50 Cassius Dio, describing the same campaign,
remarked that ‘There are two principal races of the Britons, the
Caledonians and the Maeatians.… The Maeatians live near the cross
wall which cuts the island in two, and the Caledonians are behind
them.’51 In both sources the term ‘Britons’ was rather confusingly used to
mean only those outside the empire.
The Maeatians (Miathi in Brittonic) do not appear in Ptolemy’s
Geography and may represent a tribal regrouping in response to the
campaigns of Agricola. The name Dumyat at the western end of the Ochil
Hills, overlooking Stirling, is probably derived from Dun Miathi, ‘hill fort of
the Miathi’. It has the remains of a hill fort on its summit. Myothill, west of
Falkirk, also seems to preserve the tribal name.52

Pagan Picts and Scots

In AD 305 Constantius Chlorus claimed a victory over the ‘Caledones and
other Picts’.53 This is the first reference to Picti, a late Roman nickname
for all the northern British tribes beyond their borders. It means ‘painted’
in Latin. Isidore of Seville tells us that it refers to the use of plant dye to
create tattoos.54 When the Romans first encountered the Insular Celts,
they noted the British habit of dyeing their bodies with woad, which left a
blue colour.55 As the Britons within the Roman empire gradually adopted
Roman ways, those outside it would be easily distinguished by their
tattoos. In a similar way and at around the same time the Romans coined
the name Scoti for the people of Ireland.57 When British Christian
scholars began to write about the peoples of northern Britain and of
Ireland they used the names they found in Latin histories. The Irish only
really embraced the name Scot for those of their number who turned the
land of the Picts into Scotland.

Genetics: the Picts

Geneticist James Wilson noted in 2011 that a particular haplotype of Y-DNA R1b-L21 is not
only strongly Scottish in distribution, but appears most densely in the area with Pictish
symbols. Could this be a clue to Pictish ancestry?56 Now the cluster has been more
securely identified genetically by its own marker, L1065, so we can see how it fits on to the
L21 tree. [see 102] Its parent is found in Wales. That is indeed what we would expect if
L1065 reflects native Pictish ancestry, rather than the Irish who arrived in Scotland in the
post-Roman period (see Chapter 9).
 77 Meigle Museum in the Scottish Highlands houses 27 carved Pictish stones. Meigle 4,
shown here, depicts a horseman at the top of the stone, with two interlaced serpents behind
him. Under the horse’s hooves is another entwined serpent, a Pictish Beast and another
animal, with a second horseman to the right. The Pictish symbol of a crescent with V rod
occupies the lower left of the stone.

A barbarian conspiracy fell upon Britannia in AD 367. Picts, Saxons,

Scots and Attacotti attacked simultaneously from several directions, laid
waste to towns and drove off prisoners and cattle. [78, 79] (The Attacotti
were Irish vassal peoples known as aithechthuatha.) One Roman general
was killed and another captured. The emperor Valentinian I sent the
energetic commander Theodosius to restore order. He succeeded with
the aid of a strong army.58 This presaged the upheavals that were to
follow Britain’s departure from the empire in AD 410.
78 The Roman province of Britannia came under attack from three directions in the 4th century.
These attacks by the Picts, Scots and Germani continued after Britannia became independent of
the Roman empire in AD 410.

79 Roman dice-tower boasting of the downfall of the Picts. It was discovered near the villages of
Vettweiss and Froitzheim, Germany, close to the Rhine. The front face of the tower bears the
words: ‘Pictos victos; hostis deleta; lvdite secvri’ (‘the Picts are defeated; the enemy is destroyed;
play in safety’).

Overview
• Over time, the Roman empire swallowed all the Celts except those
of Ireland and northern Britain.
• The Cisalpine Gauls were defeated in the 2nd century BC. Their
territory became a Roman province in 81 BC.

• The Roman conquest of Iberia began in urgent self-defence in 218

BC and ended in 19 BC as part of a policy of imperial expansion.

• Julius Caesar became governor of that part of Gaul already Roman

in 58 BC and had conquered the rest of Gaul by 52 BC.
• Britain was conquered piecemeal AD 43–84, but the highlands and
islands of the far north were largely excluded from the Roman
province, a first step in the creation of what today is Scotland.

• Picti (Picts) was a late Roman nickname for all the northern British
tribes beyond their borders. It means ‘painted’ in Latin. At around
the same time the Romans coined the name Scoti (Scots) for the
people of Ireland.

• Picts, Saxons, Scots and Attacotti attacked Britannia simultaneously

in AD 367.
CHAPTER NINE

Christian Celts

My name is Patrick. I am a sinner, a simple country person, and

the least of all believers. I am looked down upon by many. My
father was Calpornius. He was a deacon; his father was Potitus,
a priest, who lived at Bannavem Taburniae [Bannaventa
Burniae]. His home was near there, and that is where I was
taken prisoner. I was about sixteen at the time. At that time, I did
not know the true God. I was taken into captivity in Ireland,
along with thousands of others. We deserved this, because we
had gone away from God, and did not keep his commandments.
We would not listen to our priests, who advised us about how
we could be saved. The Lord brought his strong anger upon us,
and scattered us among many nations even to the ends of the
earth.1

When western Europe emerged from the centuries within the Roman
empire, Latin had been so firmly entrenched among the literate that the
earliest written sources from the British Isles are in that language. St
Patrick, a Briton by birth writing from Ireland, does so in Latin. The Briton
Gildas, wailing that the attacks of Saxons, Picts and Scots were God’s
punishment upon the sinful kings of the Britons, does so in Latin. It was
the language of Christianity. Latin had long unified the empire.
Christianity spread through the empire from Rome. Yet the names of the
kings Gildas reviles are Celtic.2 Patrick tells us that Latin was not his first
language.3 Here are clues that Britain would turn out to be different from
the former Celtic-speaking regions on the Continent. In Iberia and Gaul
(except Armorica), Celtic languages were irretrievably lost. Romance
languages such as French and Spanish, which evolved from Latin, are
spoken there today.
 Christianity had been a minority religion within the Roman empire for
centuries after the life of Jesus of Nazareth. Initially it was an annoyance
to Rome. The monotheism of Christianity was in direct conflict with the
expectation that loyal Romans would sacrifice to the deified emperor.
Christians were periodically persecuted. St Alban is the most famous of
the British martyrs. His shrine at Romano-British Verulamion gave the
name St Alban’s to the modern city there. From 260 there was a respite.
Christianity became an approved cult until the Great Persecution of
Diocletian from 303. By then Christianity was too entrenched to be
uprooted, as the Edict of Toleration in 311 grudgingly recognized. In
August 314 Constantine summoned the bishops of the western church to
a council at Arles in Gaul. It was attended by the bishops of York, London
and Lincoln, which were the capitals of three of the provinces into which
Britannia was divided at that time.4 This is evidence of an organized
church in Britain in Late Roman times.
Sometime in the 4th century a Romano-Briton called Annianus tossed
into the sacred spring at Aquae Sulis (Bath) evidence of his divided
loyalties. Sulis Minerva was worshipped there. It was common for
appeals to be made to the goddess on lead tablets. Annianus asked the
Lady Goddess to retrieve six silver coins from whoever had stolen them,
‘whether pagan or Christian’. Only a Christian would use such
terminology. (Pagans did not refer to themselves as such.) Yet Annianus
did not shrink from invoking the power of Sulis.5 Judging by the coins still
thrown into fountains and wishing wells today, there is a human urge at
work here that neither Christianity nor modern science has managed to
eradicate.
Soon after 350 a remarkable man was born in Britain. Pelagius was a
Christian moralist, who preached the need for simple, virtuous living. He
emigrated to Rome soon after Christianity was made the state religion of
the Roman empire on 27 February 380. He left the city when it came
under threat by Alaric the Goth in 409 and died not long after 418. His
emphasis on salvation through good deeds appealed to many, but was
seen by key figures in the Church of his day, St Augustine of Hippo and
St Jerome, as diminishing the role of divine grace. This led to his
excommunication as a heretic.6 Pelagianism gained so much support in
post-Roman Britain that Pope Celestine sent St Germanus of Auxerre to
Britain in 429 to denounce it.7
 By the time Britannia became independent of Rome in 410, Christianity
was firmly established there. [80] In the quotation that begins this chapter
Patrick suggests that religious apathy was so common as to bring down
the wrath of God upon the Britons. Yet he was raised in a Christian
environment. Though Gildas found much to complain about in his sinful
countrymen, he never accused them of a return to paganism. Patrick tells
us that he and thousands of other Britons were captured by Irish raiders
and taken into slavery in Ireland. This makes it possible that Christianity
first arrived in Ireland unwillingly, though other Britons were migrating to
Ireland of their own free will, as we shall see (pp. 169–70). As early as
431 there were enough Christians in Ireland for Pope Celestine to send
Palladius to be the first bishop of the Irish (see below).8

80 This gold disc bears a monogram formed of the Greek letters chi (X) and rho (P), the first two
letters of Christ’s name, combined with alpha (A) and omega (W), the first and last letters of the
Greek alphabet, and an appellation of Christ. It is part of a hoard of 4th-century Christian objects
found at Water Newton, Cambridgeshire. It is the earliest group of Christian liturgical silver yet
found within the former Roman empire.

Ogham
The oldest recorded form of the Gaelic languages is Primitive Irish, which
is known only from ogham inscriptions. Given the centuries of contact
between Ireland and the Roman world, the Latin alphabet was probably
the first that the Irish encountered. Instead of adapting it to the writing of
Gaelic, the Irish simply took the concept and created their own alphabet.
The ogham symbols, formed of strokes along or across a line, are
reminiscent of tally notches. [81] The letters they represent were chosen
to fit the needs of Primitive Irish, which had no ‘p’ sound.9
Ogham inscriptions are found in Ireland and western Britain from the
4th to the 6th centuries. They are funerary memorial stones. The densest
concentration is in southwest Ireland in counties Kerry, Cork and
Waterford. Kerry alone has a third of the total number in Ireland. This is a
region of Ireland with no ecclesiastical foundation known before AD 600.
In the whole corpus of over 300 Irish ogham stones, hardly more than a
dozen show any sign of Christian affiliation.10 So for a body capable of
overseeing such memorials, we may look to the druids.11
Druidic control of the language of ogham might well explain its antique
character. It is strikingly similar to Latin. Near Slane in Co. Louth, there is
an ogham stone in memory of Mac Caírthinn Uí Enechglaiss (d. 446), to
use the modern Irish version of his name. It reads MAQI CAIRATINI AVI
INEQUAGLAS.12 Primitive Irish is markedly more ancient than Old Irish
(see below), which appears in manuscripts from the 6th century onwards.
The latter represents a giant leap towards Modern Irish, including the
presence of the letter ‘p’. Such a language change over so short a period
would be unusual. It has been argued therefore that Primitive Irish
represents the last bastion of the sacred language of the pagan Celts,
passed down from druid to druid, retaining features that had long
disappeared from the vernacular speech of non-druids.13
81 The ogham alphabet was written on the edge of stones, the edge marking the division between
strokes to the left and strokes to the right. This inscription from Ballyeightragh, Co. Kerry reads
MAQI LIAG MAQI ERCA, the name of the person commemorated.

Where ogham inscriptions are found in Britain we can be sure of Irish

settlement. The strongest concentration is in Dyfed, southwest Wales.
They are also found in Devon and Cornwall, northwest Wales and parts
of Scotland. The Isle of Man has six ogham inscriptions.14 Manx Gaelic,
descended from Old Irish, replaced a Brittonic language there.

Irish in Dyfed
In Roman times the people of what is now southwest Wales were the
Demetae.15 Their territory of Demetia appears to have passed
seamlessly to the medieval kingdom of Dyfed. Gildas, in his tirade
against the 6th-century kings of British tribes, does not spare ‘Vortipor,
tyrant of the Demetae’.16 A memorial stone to Voteporis has usually been
identified with the Vortipor of Gildas.17 [82] The fact that this stone is
inscribed both in Latin and in Irish (using ogham) lends support to the
claim in the 8th-century Irish epic The Expulsion of the Déisi that Eochaid
son of Artchorp settled in Demed (Demetia) and founded a dynasty.18
 Déisi simply means ‘vassal peoples’. But by the 8th century déisi
communities in the southeast of Ireland had formed the sub-kingdom of
Déisi Muman (déisi of Munster). The Expulsion of the Déisi promulgated
the notion of the Déisi as an ancient tribe expelled from Tara. The tale of
Eochaid son of Artchorp is an intriguing digression from the main story. It
preserves a pedigree for the kings of Dyfed which has similarities with the
Dyfed royal line in Welsh genealogies, at least as far back as Triffyn. He
appears in the Irish genealogy as Triphun, the great-grandson of
Eochaid, but in the Welsh version as a descendant of Maxen Wledic.
Both pedigrees include Vortipor (Gwrthefyr in Welsh) as the grandson of
Triffyn.19

82 A memorial stone inscribed MEMORIA VOTEPORIGIS PROTICTORIS (‘The monument of

Voteporix the Protector’), found at Castell Dwyran, Carmarthenshire, Wales. Ogham letters just
visible at the top and down the left side give the Gaelic version of the name: Votecorix.
 Maxen Wledic is the Welsh rendering of Magnus Maximus, a Galician-
born Roman general who was assigned as commander to Britain in 380.
In 383 Maximus was proclaimed Western emperor by his soldiers. He
made his son Victor his colleague in power, but their imperial ambitions
led to both their deaths in 388.20 Maximus appears as the improbable
progenitor of several Welsh royal lines. This could result from the
deduction of pedigrees from king-lists, starting with Maximus as overlord.
The Irish of southwest Wales could have been bands of landless
warriors (fianna), who had been recruited by the Roman authorities in the
latter half of the 4th century to protect Demetia from Irish raids. In that
case Triffyn would be the first Irish protector of Demetia, which title
passed down to Vortipor.21 Another Irish group which settled in south
Wales and also in Cornwall was the Uí Liatháin (descendants of Liatháin)
of east Cork.22 The Historia Brittonum tells us that the ‘sons of Liethan’
occupied the country of the Demetae, as well as Gower and Kidwelly.23

Early Christian Ireland

The Confession written by Patrick in old age, carefully copied by his
disciples and handed down through the ages, ensured him a fame denied
to his fellow missionaries in Ireland. The self-portrait that leapt from his
pen was of a passionate advocate of his faith, driven by dreams in which
he sensed the divine will. Having escaped from servitude in Ireland,
Patrick trained for the ministry in Britain. Then he returned to Ireland
intent on spreading Christianity.24 The stirring figure of Patrick has
attracted a huge body of tradition. For centuries he has been known as
the patron saint and apostle of Ireland. St Patrick’s Day is celebrated not
only in Ireland, but throughout the Irish diaspora, especially in the USA.
By comparison Palladius, the first bishop in Ireland, has been almost
lost to sight. He came from Auxerre in Gaul and presumably belonged to
the aristocratic Gallo-Roman family of the Paladii, which had controlled
that region for generations. Since he was sent in 431 to minister to those
in Ireland who were already Christian, he may not have sought
conversions as Patrick so plainly did. Two apparently Continental clerics,
bishops Auxilius and Secundinus, represented as disciples of Patrick in
later sources, are more likely to belong to the mission of Palladius. They
left a legacy in the place-names Killashee (Cell Auxili ‘the cell of Auxilius’)
in Co. Kildare, and Dunshaughlin (Dún Sechlainn ‘the fort of Secundinus’)
in Co. Meath. These two places lie in the centre of eastern Ireland.28

Genetics: Déisi Muman

Irish surnames such as Whalen, Phelan and O’Phelan are derived from the personal name
Faeláin (wolf). Faeláin is recorded a number of times among the Irish royal houses, so we
may guess that it would have been popular among lesser folk too. We would expect
therefore a number of different Y-DNA haplogroups among men of these surnames. That is
indeed the finding of the Family Tree DNA project for this group of surnames. The majority
of men within it fall into the common Irish haplogroup R1b-L21, but that does not
necessarily make them closely related.
One particular Faeláin is of interest for the story of an Irish dynasty in Dyfed, founded by
Eochaid son of Artchorp (see p. 166), for this same Artchorp appears in an early pedigree
as an ancestor of Faeláin, son of Cormac, king of the Déisi Muman.25 The Annals of Ulster
record the demise in 966 of this Faeláin. His descendants appear in the same annals as Ua
Faeláin by 1085, a step towards surname development. So it is quite possible that some
Whalens and Phelans descend from him.
A group of men of these surnames carry the R1b-L144 subclade of R1b-L21 and have
ancestry from counties Laois or Waterford in Ireland, overlapping with the territory of the
Déisi Muman (Munster). The genetic marker L144 has also been found in men of the Welsh
surnames Prosser and Griffith.26 The Welsh long retained the system of naming by
genealogy (see p. 207), so when they eventually adopted hereditary surnames, these
chiefly reflected paternal names. Griffith is derived from the Welsh personal name Gruffydd.
There will be many unrelated men with that surname. The Welsh equivalent of ‘son’ is map,
shortened to ap or ab, which in some cases was partly merged with the paternal name to
form a surname. Dafydd ap Rosser would become David Prosser.27 So here we have
surnames both Irish and Welsh linked by Y-DNA. It can only be a guess that the link might
have arisen from Irish settlement in southwest Wales. In the same haplogroup are men of
two surnames derived from places in the north of England: Kendall and Bracewell.

In Muirchú’s Life of Saint Patrick, the saint’s mission is connected to

the northeast of Ireland, rather than the centre.29 [83] This suggests that
the home from which he was taken was somewhere in the north of what
had been Roman Britain. In the surviving copies of Patrick’s Confession,
his birthplace appears as Bannavem Taburniae, which makes no sense
as a Romano-British place-name. The Latin of his period often omitted
breaks between words. If we place the break more logically, we get
Bannaventa Burniae. There was a Bannaventa in what is now
Northamptonshire. That can be ruled out as Patrick’s birthplace, which
was close to the Irish Sea, according to Muirchú. So the addition of
Burniae or Berniae would distinguish this unknown Bannaventa from the
one in Northamptonshire.30 There has been a mass of contradictory
speculation on its location, but no certainty. Even the dates for Patrick
are uncertain. The Annals of Ulster give no fewer than four alternative
dates for his death. A clue that tips the balance in favour of the last of
these, 493, is that a British disciple of Patrick, Mochta, who founded a
monastery at Louth in Ireland, died in 535.31
Patrick and Mochta were not the only Britons to enter northeast Ireland
of their own volition. The Irish annals refer to warbands of British people
rampaging around Ireland. The earliest reference tells us of the ‘killing of
Colman Mor, son of Diarmaid, in his chariot, by Dubhshlat Ua Treana,
one of the Cruithni’, in 552.32 Cruithni or Cruithin is the Irish version of the
Welsh Prydyn (Britons). From this we can reconstruct an earlier Brittonic
version, Pritani, meaning literally ‘figured folk’, usually interpreted as
‘tattooed people’. It seems that ancient Greek travellers, hearing Pritani,
wrote a Greek approximation which became Britanni in Latin.33 From this
ethnonym, Greeks and Romans created the island name Britannia, which
they preferred to Albion. They also coined terms which translate as ‘the
British Isles’ to include Ireland and the smaller islands of the group.34
This was simply a geographical convenience, identifying the archipelago
by the name of the largest island. It continues to this day.
83 Domhnall Ua Lochlainn, king of Ireland, commissioned this ornate bronze shrine in Armagh c.
1100 to encase a quadrangular bell reputed to have belonged to St Patrick. The front and sides of
the shrine are decorated with silver and gold filigree wound into curvilinear patterns; those on the
sides depict elongated beasts intertwined with ribbon-bodied snakes, characteristic of the final,
Viking-influenced, style of Irish Celtic art.

The fact that these Cruithin arrivals from Britain had no specific tribal
label (such as Brigantes) makes them part of a changing society. Tribal
society in Ireland gradually dissolved in the centuries after 400. The
earliest memorial stones found there frequently record the tribal affiliation
of the deceased. By the time the first true annals for Ireland start in the
middle of the 6th century, new dynastic kindreds were in evidence, which
came to dominate the island politically.35 Cruithin dynasties emerged in
Ulster: the Dál nAraidi of southern Antrim and their offshoot the Uí
Echach Cobha.
Both these dynastic names incorporate a personal name, that of the
founder of the kindred. Uí Echach Cobha means ‘descendants of Echu of
Cobha’. Magh Cobha lay within what became the Barony of Iveagh (Uíb
Echach) in Co. Down.36
A lineage can be pieced together for the Uí Echach Cobha, starting
with the death of Eochu in 553. The annal entry helpfully tells us that he
was the progenitor of the Uí Echach Ulad, another name for the Uí
Echach Cobha. This information could only have been added
retrospectively, so we cannot trust the genealogical detail that he was the
son of Conlaed, king of Ulaid.37 In the 8th century the Dál nAraidi were
anxious to cast off their ‘foreign’ label. The term Cruithin disappeared
from the annals after 773, and the Dál nAraidi claimed to be descended
from the older-established Ulaid.38
The deaths of successive chiefs of Uí Echach Cobha and Iveagh are
recorded in the annals. Surnames gradually developed within this
kindred. One Aonghusa (Angus) had a son who was described as Mac
(son of) Aonghusa in the usual Irish system of identification by
genealogy. It so chanced that he lived at a time when hereditary
surnames were developing, so MacAonghusa became the surname
McGuinness. This lineage became lords of the Barony of Iveagh in the
12th century, taking over from another branch of the family.39 It is
interesting then that several men of this surname carry a Y-DNA
haplogroup that seems to have arrived in Ireland from Britain.

Genetics: the Cruithin

A Y-DNA haplogroup that seems to have arisen in Britain has been found in some men who
can trace descent from the Uí Echach Cobha, a Cruithin (British) dynasty in Ireland. When
this link was first noticed, the phylogenetic tree of Y-DNA was a slender thing by
comparison with its bushy growth today, so the name of the haplogroup in question has
changed. Brian McEvoy and Dan Bradley of Trinity College Dublin identified it as simply a
subclade of I1c.40 I1c was the name in about 2005 of the present I2a2a (M223). The
subclade they had in mind is currently known as I2a2a1a1 (M284). It has now itself been
divided into subclades by newly discovered markers.
Haplogroup I2a2a1a1 (M284) is very rare outside the British Isles, except among those of
British and Irish origin. Trace amounts are found in France and Germany, and a slightly
higher percentage in Portugal, England’s oldest ally. To judge by the estimated date that it
burgeoned into sub-lineages, it was one of the I2 family that travelled with early farmers. So
it could pre-date the Celts in Britain. The bearers of I2a2a1a1 (M284) have a mixed bag of
surnames including English, Welsh, Scottish and Irish. Its descendant clade I2a2a1a1a1
(L126/S165) is more common in Scotland. Its offshoot I2a2a1a1a1a (S7753) includes men
of several surnames of Irish Gaelic origin, such as McGuinness, Callahan, McConville and
McManus, indicating that S7753 arrived in Ireland before the development of surnames.
The estimated date of the haplogroup is around AD 500, which makes a neat fit to the
earliest reference to the Cruithin in AD 552 (see p. 169).41
84 Tree of Y-DNA haplogroup I2a2a1a1 (M284).
85 Provinces and dynastic kindreds of early medieval Ireland, illustrating the expansion of the Uí
Néill in the north and the Eóganacht of Munster, and Irish migrations to parts of Britain.

If the ancestors of the Cruithin came to Ireland in the post-Roman

period, then we cannot be surprised at their absence from Ptolemy’s map
of Ireland around AD 150. [see 67] Yet we should not interpret every
population missing from this map as arrivals in Ireland after AD 150. The
Romans had no reason to survey Ireland in detail. As for the dynasties
that rose from obscurity into power in the 7th century, why should we
expect a clue to their existence before they were formed?
Under the hegemony of these new dynasties, the Uí Néill in the north
and the Eóganacht of Munster, there was a language shift in Ireland. [85]
When Christianity replaced druids with priests in Ireland, it also replaced
Primitive Irish with Latin as the sacred language. The Latin alphabet was
also used to create a new written form of Irish. [86] This was Old Irish,
taught in the monastic schools alongside Latin. It sprang from everyday
spoken Irish – the vernacular.42 The Old Irish period gave birth to a new
word for the Irish. That ethnonym was Féni. A clue to its origins lies in the
Old Irish view that in the past ‘there were three principal peoples in
Ireland, namely Féni, Ulaid, and Gaileōin, that is Laigin’. Thus the rulers
of Connacht and the Uí Néill dynasties were Féni, as opposed to the men
of Ulster and Leinster.43
86 The Kilnasaggart Stone in Co. Armagh marks the site of an early Christian cemetery. It has an
inscription in Irish, which records that Ternohc, son of little Ciaran, dedicated the place to the
Apostle Peter. Above the inscription is a Latin cross, and beneath is a cross within a circle. There
are other crosses on other faces of the pillar. Since Ternohc died in 716, the stone can be dated
to c. 700, which makes it the earliest datable Christian monument in Ireland.

Perhaps it is just a coincidence that the earliest Irish grammar primer

declares that the battle of Moira, Co. Down, in 637 was crucial in its own
creation. This battle marked the beginning of the end of Ulaid power and
the triumphal rise of the Uí Néill and their allies in Ulster. The primer
poetically declares that the battle dashed the ‘brain of oblivion’ out of the
head of its first author, who settled down to amass knowledge of poetry
and other writings. The man in question was Cenn Fáelad mac Ailill (d.
679) of an Ulster dynasty, the Cenél nÉogain. He wrote his Irish primer in
Derryloran, Co. Derry. This and his other writings in Gaelic helped to
establish the earliest vernacular literature in Europe.44
Anglo-Saxons and Britons
Then all the councillors, together with that proud tyrant
[Vortigern], 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 fierce and impious Saxons, a race hateful
both to God and men, to repel the invasions of the northern
nations. Nothing was ever so pernicious to our country, nothing
was ever so unlucky.45

So Gildas rants at the folly that brought Saxons down upon the Britons.
The choice of Saxon mercenaries to guard Britain does seem perverse.
The Gallic Chronicle tells us that the Saxons laid waste the British
provinces in 408.46 The result of hiring warriors from the enemy could
have been predicted by those who knew the history of Gaul (see p. 149).
The mercenaries grew restive, complained that their remuneration was
inadequate, broke their treaty and began a campaign of plunder, calling
support from fellow Saxons.47 Or so Gildas says, and it is a reasonable
explanation of how the Saxons first gained a toe-hold in Britain. Bede,
writing in the monastery of Jarrow in the 7th century, gives us a more
detailed picture. The incoming Germani were Angles, Saxons and
Jutes.48 He was himself of Anglian descent, and gave preference to the
Angles in the title of his history, The Ecclesiastical History of the English
People, which probably did much to ensure that the name England
(Angle-land) was adopted. People might be speaking a Romance
language in England today, had these newcomers not burst upon the
scene, bringing the dialects that developed into English. The most
Romanized population of Britain was in the rich lowlands, dotted with
bustling towns. Latin was probably widely spoken in southern Britain by
late Roman times. This was the very area that turned English. The good
agricultural land of the south attracted Anglo-Saxon settlement. Towns
were abandoned.49
What happened to the Latin-speakers? Some may have perished on
Saxon swords. Some may have adopted the language of the incomers.
Yet it is intriguing that the Celtic which survived in the British highland
zone developed a Latin accent at around this time, as though a rush of
Romance refugees had arrived.50 That would fit the picture that Gildas
painted of the ‘miserable remnant’ of the Romano-British making for the
mountains, fleeing overseas or surrendering themselves to be slaves to
their foes, as the only alternative to famine.51
By 441 the Saxons ruled ‘the British provinces’, according to the Gallic
Chronicle.52 Britannia had been divided into several provinces by Late
Roman times. So had all former Romano-British territory fallen to the
Saxons? It might have seemed that way to an observer across the
Channel, but other sources show how complex the political map of Britain
actually was in these changing times. Wales, the southwest and the Yr
Hen Ogledd were ruled by British kings. Yr Hen Ogledd (The Old North)
was the Welsh term for the region that is now southern Scotland and
northern England where a language akin to Welsh was spoken.

Genetics: Britons and Anglo-Saxons

There have been several attempts to work out from the DNA of the living the influence of
the Anglo-Saxons on the gene pool of England. Most recently the large People of the British
Isles study suggested a substantial contribution to the English population spreading in from
the east, putatively the Anglo-Saxons.53
No certainty is possible without directly comparing the DNA of people in Britain before the
Anglo-Saxon advent and after it. This was the aim of a study that has generated the first
ancient genome sequences from Britain. The study sampled five individuals from Hinxton,
Cambridgeshire, eastern England. Two men lived just before the Roman period and three
women lived in the middle of the Anglo-Saxon period. All five samples are broadly similar to
modern northern European peoples. Yet there are enough differences between the Britons
and Anglo-Saxons to distinguish them.54
87 The kingdoms and principal places of southern Britain c. 600. British kingdoms remained in the
west, but a huge swathe of what is now England had been taken by Angles and Saxons.

When the Saxons began to advance into the southwest and further
north from around 550, there had been time for generations of inter-
marriage between Briton and Saxon in the borderlands between the two.
[87] The early kings of the Hwicce in the Severn Valley had Germanic
names, yet they were Christian and probably of the British Church.55
Cerdic, founder of the royal line of Wessex, has a name that seems like a
Germanization of the British name Caraticos.56 The laws of Ine of
Wessex (ruled 688–726) recognize the existence within his domain of
Welsh (as the Saxons called the British), though according them a lower
status.57 Saxon settlement in the far southwest was so slow that Cornwall
retained a Brittonic language for centuries afterwards (see below pp. 180,
187.) So the impact of the Anglo-Saxons varied by region.

Kingdoms of Wales
Just as the Romans can be seen as creating Scotland, so in a way the
Anglo-Saxons created Wales. Before their arrival, what is now Wales was
part of Britannia, united to the rest of the Roman province by culture and
language. Once the Angles and Saxons had overrun much of lowland
Britain, Wales emerged as a separate country. The Welsh names for
Wales (Cymru) and its people (Cymry) were unknown in Roman times.
The term Cymry (people of the same home region) first appears in a
poem addressed to the king of Gwynedd about 632/4.58 It was the Anglo-
Saxons who christened them ‘Welsh’, from the Old English word walh or
wealh, meaning ‘foreign’. The great earthwork known as Offa’s Dyke was
designed by King Offa of Mercia (d. 796) to keep the Welsh at bay.59 It
has roughly marked the border between England and Wales ever since.
Medieval Wales was a patchwork of often contending kingdoms. [87]
The continuity between the Romano-British Demetia and medieval Dyfed
in southwest Wales has already been mentioned (pp. 165–67). It is a
similar picture in southeast Wales, where the territory of the Silures
became the medieval kingdom of Gwent. This was the most Romanized
region of Wales. Imported Mediterranean pottery shows that trade with
Constantinople (Byzantium) continued following the collapse of the
Roman empire. The name Gwent derives from Venta Silurum, the capital
of the civitas of the Silures. By contrast, the large medieval kingdom of
Powys in east Wales was not a direct successor to a post-Roman polity.
The relentless pressure of the encroaching Angles created its eastern
border, as the Britons were pushed into the upper valleys.60
In north Wales Gwynedd had natural advantages. The flatland of the
Isle of Anglesey (Ynys Môn) was ideal for arable farming. The island was
protected by the sea and the soaring bastion of Snowdonia. In Roman
times northwest Wales seems to have been at least part of the territory of
the Ordovices (‘hammer-fighters’).61 A 5th-century inscription at Penbryn
in southwest Wales commemorates a man of the tribe, who died far from
home. [88] So how did the name Gwynedd arise? A memorial stone
found at Ffestiniog, from around 500, records the new name. It is
inscribed in Latin: CANTIORIX HIC IACIT. VENEDOTIS CIVES FUIT
CONSOBRINOS MAGLI MAGISTRATI (‘Cantiorix lies here. He was a
citizen of Venedota [Gwynedd], a cousin of Maglos the magistrate’).62
The use of Late Latin and the terminology of citizenship and magistracy
is powerful evidence that a Romano-British polity had survived for a
century in this corner of Wales. Four former district names in Gwynedd
are derived from Roman personal names.63
Yet several scholars have favoured the idea that the name Venedota is
derived from the Old Irish Féni (see p. 173). One argument is that the
Féni of Connacht had broken through the barrier of the Laigin of Leinster
to reach the Irish Sea c. 500 and so theoretically could have crossed that
sea to north Wales.64 It is an ingenious idea, but strained. The root word
weni-is found in Brittonic and Gaulish, with the probable sense of ‘family,
kindred’, preserved in the Breton word gwenn, meaning ‘race’.65 This is
the type of word, signifying ‘us’ as opposed to ‘them’, that often becomes
the basis of ethnonyms, as we have seen with Cymry. The Breton name
of the present-day town of Vannes is Gwened [see 92], a parallel to
Gwynedd. Vannes takes its name from the Veneti tribe of Armorica,
whose territory became an early Christian diocese centred at Vannes.66
In the case of Gwynedd, the distinction of ‘us’ versus ‘them’ probably
began as a way to exclude the Irish. There is evidence of Irish settlement
in Gwynedd, though by the Laigin rather than the Féni. The name of the
Lly^n Peninsula here is derived from the Laigin, and the same root
appears in Porthdinllaen, a coastal village on the peninsula. The power
base of the first dynasty of Gwynedd was Anglesey, not Lly^n. Gildas
calls Maelgwyn, an early 6th-century king of Gwynedd, the ‘dragon of the
island’.67 Maelgwyn’s descendant Cadfan (d. c. 625) has a memorial
stone at Llangadwaladr on Anglesey to the ‘most wise and renowned of
all kings’. [see 6] It is built into the fabric of the church, and was probably
moved there when his grandson had the church built. Llangadwaladr
means ‘church of Cadwaladr’. It lies close to Aberffraw, known to be a
royal site at a later date.68
Later kings of Gwynedd, of a second dynasty, had a tradition that
Maelgwyn’s great-great-grandfather, Cunedda, came south from Manaw
Gododdin, on the Firth of Forth, and expelled the Irish from Wales.69 That
would explain why a 5th-century elegy on the death of the northern
Cunedda son of Edern was preserved in Wales.70 Modern historians tend
to be dismissive.71 Genealogies can be tampered with. One argument is
that the head of the second dynasty of Gwynedd, Merfyn Frych (d. 844),
may have been from the Isle of Man, and could have aimed to legitimize
his foreign origin by selecting an immigrant as the founder of the first
dynasty.72 This reasoning seems positively tortuous. We may doubt that
Cunedda himself came south, but knowledge of him did. The most
probable source of that knowledge would be his descendants.
88 A Latin inscription at Penbryn, Ceredigion: CORBALENGI IACIT ORDOVS. (‘Here lies
Corbalengos the Ordovix.’) His tribe was no doubt mentioned because he was buried outside its
territory in north Wales.

Britons abroad
The Roman civitas of the Dumnonii, with its capital at Isca (Exeter),
became the kingdom of Dumnonia after Britain separated from the
Roman empire. It remained British long after the Anglo-Saxons had taken
over southeastern Britain. In about 540 Gildas ranted at its king,
Constantine, ‘the tyrant whelp of the filthy lioness of Dumnonia’.73 The
learned Aldhelm, abbot of Malmesbury, addressed a letter around 675 to
King Geraint of Dumnonia pressing him to adopt Catholic (Roman)
church practice.74 This Geraint (or perhaps a son of the same name) was
attacked by Ine of Wessex around 710.75 By about 813 the inroads of the
kings of Wessex had reduced Dumnonia to Cerniw (in Welsh) or Kernow
(in Cornish). The name is derived from the Celtic root corn, meaning
‘horn’, referring to the southwestern peninsula of Britain. The Anglo-
Saxons added their word walh for ‘Britons’ to create the present county
name Cornwall.
Dumnonia had absorbed immigrants from Ireland earlier. This may
have been the initial impetus for British migration to the province of
Armorica in Gaul, followed by another wave as the Saxons of Wessex
penetrated Dumnonia. [89] There was so strong a flow from Britain to
Armorica that the latter gained the name Brittany. For the French,
Brittany is Bretagne, while Britain is Grand Bretagne (Great Britain). Most
of the incomers appear to have come from the southwest of Britain, for
the early medieval kingdoms of Dumnonea and Kernev appeared in
Brittany.76
Tangible evidence of British settlement are the many Breton churches
named after their local founders, following the same practice as in
Cornwall and Wales. Some of these Brittonic saints were venerated in
both Cornwall and Brittany, such as Samson and Petroc, but the great
majority were honoured in just a single church dedication. In the Middle
Ages St Petroc was the most revered Cornish saint. The place-name
Padstow (holy place of St Petroc) records his ministry there. Petroc
appears to have died and been buried at Padstow, but his relics were
moved to Bodmin. His cult spread to Brittany, as is manifest in the place-
names formed with his name (Perec in Breton).77
89 The migration of Britons to Armorica and Galicia in the Post-Roman period.

The Life of St Samson of Dol is the earliest written narrative from

Brittany. It tells us that the saint was born in south Wales and dedicated
to God while still an infant. As an adult he travelled through Cornwall to
Brittany. A Samson was present at the Council of Paris in 562, who could
be the St Samson who founded the monastery at Dol-de-Bretagne.78
Mobility within the Brittonic-speaking region is the thread running through
these lives.
A separate Brittonic settlement in the former Roman province of
Gallaecia in Spain was known as Britonia. Gallaecia was larger than the
present region of Galicia, encompassing Asturias and Leon (Spain) and
northern Portugal. A British diocese there is first mentioned in 572, with
its see probably at the monastery of Santa Maria de Bretoña near
Mondoñedo. The parish churches belonging to this diocese extended
from Mondoñedo north to the sea, and east across the River Eo into what
is now Asturias, suggesting a substantial migration of Brittonic
speakers.79

Christian Picts and Scots

We know and from the chronicles and books of the ancients we
find that among other famous nations our own, the Scots, has
been graced with widespread renown.… The Britons they first
drove out, the Picts they utterly destroyed.

Thus the Declaration of Arbroath in 1320 boosted a claim to Scottish

independence. It suited the Scottish nobility of the time to claim the utter
destruction of the native Celtic tribes north of the Antonine Wall by the
incoming Irish (Scoti). Such a genocide is highly unlikely, but the Picts
were no longer an independent people after 900.80 How had this come
about? Until recently the explanation seemed simple: Gaels from Ireland
had invaded Argyll. There are certainly some ogham inscriptions in
Argyll. Yet there is no sudden cultural shift. Archaeologist Ewan
Campbell suggested that the people west of the mountain spine of
Scotland, isolated by geography from the linguistic changes elsewhere in
Britain, had simply retained the ancient Gaelic language of the British
Isles.81 That idea founders on the fact that Scottish Gaelic is a
descendant of Old Irish. Geography does seem crucial though. Kintyre
reaches out to the northeast tip of Ireland. [90] Around it are islands, so
transport by boat would be commonplace. Trade, marriages and
alliances could weave an invisible web across the water.
In the 8th century Bede had heard that the Irish ‘came from Ireland
under their leader Reuda and won lands from the Picts either by friendly
treaty or by the sword.… They are still called Dalreudini after this
leader.’82 ‘Dalreudini’ is recognizable as the Gaelic kingdom of Dál Riata
or Dalriada, which emerged around 700 from the union of several
kindreds. Though mainly in Scotland, Dál Riata during part of its
existence encompassed a small part of northeast Ireland. As for ‘their
leader Reuda’, it seems that two of these kindreds, those of Kintyre and
Cowel, promoted the idea of their joint descent from a Domangart Réte
who died around 507. A separate kin-group that held sway in Islay seems
an offshoot of the Dál Fiatach, ruling dynasty of the Ulaid. By contrast,
the dominant kindred on Skye begins with Pictish names, yet spent three
years in exile in Ireland in the 660s. Another three kindreds ruled Lorn.
After all these regions were welded together as the Gaelic kingdom of
Dál Riata, a genealogy was concocted descending all the kindreds from
an Irish Eochaid.83

90 The early medieval kingdom of Dál Riata was composed of territories on the west coast of
Scotland and the Inner Hebrides that had earlier belonged to separate kindreds. Most of the
territory of Dál Riata became the historic county of Argyll.

At the same time that Bede absorbed the approved history of Dál
Riata, he was fed an origin story by the Picts that involved a cunning
piece of propaganda. In this fiction, when the Picts settled in northern
Britain they had no wives and asked the Irish for some. ‘The latter
consented to give them women only on condition that, in all cases of
doubt, they should elect their kings from the female royal line.’ Bede
innocently reported that the custom had been observed among the Picts
to his day.84 The supposed matrilineal succession of the Picts was taken
seriously until modern times, when historians realized that a
contemporary of Bede was the Gaelo-Pictish Bridei, who had claimed a
Pictish throne through his Pictish mother. The origin story conveniently
legitimized this.85
Even more startling is the modern unpicking of the traditional view of
Kenneth MacAlpin (d. 878), long heralded as the Scot who vanquished
the Picts. Critical scrutiny of the sources reveals that he was in fact more
probably Pictish himself, and was tagged on to the lineages of Dál Riata
by later genealogists. He was the first to rule both Pictavia and Dál Riata.
The crucial shift to a Gaelic-speaking court seems to have occurred in
the time of his grandsons. They may have spent their youth sheltering
with their aunt Máel Muire (d. 913), who married two high kings of Ireland
in succession. It is in their time that the concept emerges of the kingdom
of Alba, which encompassed both Picts and Scots.86
The closer one looks at this complex picture, the less it seems like one
huge Irish army rushing up the beaches around 500, establishing a
kingdom in Argyll that would eventually vanquish the native Picts. Gaelic
could have been introduced to Scotland partly through a web of alliances,
with threads of religion, politics and marriage. This would not rule out the
actual movement of people from Ireland to Scotland. Far from it. It would
just make the process more diffuse and drawn out. Among the Irish
welcomed into the Gaelic enclave in western Britain was St Columba,
who founded his famous abbey on the little Hebridean island of Iona in
563.87 He is seen as the apostle of the Picts, completing the conversion
to Christianity of all the peoples of the British Isles.

91 The Book of Kells contains the four Gospels in Latin, written in a type of uncial script devised in
Ireland. In addition to the ten full-page illuminations, the text is enlivened by many illuminated
initials, such as this one from Matthew.

Irish art came with Columba and the community he founded. The
swirling La Tène style had continued to develop in Ireland after the
Continental heartlands of La Tène and most of Britain were absorbed into
the Roman sphere. As Ireland embraced Christianity, Irish art blossomed
in such masterpieces as the famed Book of Kells, a gospel book created
around 800 in a Columban monastery in Britain or Ireland. In addition to
the well-known full-page illuminations, almost all the folios of the Book of
Kells contain initial letters decorated with intertwining art with bird or
animal features. [91]

Overview
• The Roman province of Britannia was Christian when it left the
Roman empire in AD 410.

• Latin was the language of Christianity, yet it was not the first
language of all the Romano-British. Celtic had survived there, as
well as in those parts of the British Isles outside the Roman empire:
Ireland and northern Britain.

• The Irish developed their own form of writing using ogham signs.
Irish settlement in parts of Britain can be identified by ogham
inscriptions.

• The Romano-British civitas of the Demetae became the medieval

kingdom of Dyfed in southwest Wales. There is an ogham and Latin
memorial there to Vortipor, identified by genealogies as one of an
Irish royal dynasty of Dyfed.

• St Patrick was among thousands of Britons captured and enslaved

by Irish raiders in the 5th century. St Patrick escaped and later
returned to take Christianity to the pagan Irish.

• Genetic evidence supports the arrival of Cruithin (Britons) in

northeastern Ireland in the post-Roman period.
• Tribal society in Ireland gradually dissolved in the centuries after AD
400. Dynastic kindreds emerged which dominated Ireland by the 7th
century.

• Angles, Saxons and Jutes gained control of the southeast of Britain

in the 5th century. Genetic evidence suggests that this involved
mass movement. Celtic polities remained in the rest of Britain, but
were reduced by later Anglo-Saxon expansions which penned
Brittonic speakers into Wales and Cornwall.

• The concept of Wales as a separate country emerged from the

process of Anglo-Saxon expansion. Poetry created in the Old North
was preserved in Wales. That survival supports the tradition that the
first royal dynasty of Gwynedd came south from Manaw Gododdin.

• The Roman civitas of the Dumnonii, with its capital at Isca (Exeter)
became the kingdom of Dumnonia in southwest England. A stream
of migration left Dumnonia for Armorica, which changed its name to
Brittany. A second Brittonic settlement in northwestern Iberia was
known as Britonia.

• The medieval claim that Scots from Ireland had utterly destroyed the
native Picts is not supported by the evidence, including from
genetics. Even the traditional concept of Irish invasion now appears
simplistic. Irish movement to Scotland seems to have come about in
complex ways.

• The Irish St Columba founded an abbey on the Hebridean island of

Iona in 563. He is seen as the apostle of the Picts.
CHAPTER TEN

Loss and Revival

Great nations, I mean such as have been famous, and made a

considerable figure in the world, are almost like great rivers, that
are never thoroughly known, unless you ascend to their very
spring and original: it is with some measure of justice that the
Celtae, a people better known by the name of Gauls, should be
reputed great, either upon account of the number of their
people, valiant actions, or the antiquity of descent.1

In Brittany, Ireland, Scotland and Wales live at least some people for
whom a Celtic language is a mother-tongue, passed down a chain of
ancestors from prehistory to the present. They grow fewer every day.
English and French have almost swallowed up those Celtic languages
that survived into the Middle Ages. In Continental Europe, Celtic
languages were lost to Latin during the Roman empire. Romance
languages derived from Latin had sprung up in Armorica and Galicia
before the arrival of colonies of Britons there in the post-Roman period.
Only in Armorica did the influx of Brittonic-speakers return the region to a
Celtic language.
Cumbric, the Brittonic language of the Old North, vanished so long ago
that few have even heard of it. It sank under the tide of English washing
north and becoming tinted into Scots, the dialect of the Scottish
Lowlands.2 [see 94] In the 18th century Lowland Gaelic too gave way to
Scots. Cornish died out in the 19th century.3 The last native Manx
speaker, Ned Maddrell, died on 27 December 1974, but the shift to
English was already under way on Man when he was born in 1877.4
Of the living Celtic languages, Welsh is in the healthiest condition.
Though UNESCO sees Welsh as vulnerable, Irish and Scottish Gaelic
are classified as definitely endangered, and Breton as severely
endangered. Almost two million people spoke Breton at the beginning of
the 20th century. That number has now declined to around 250,000.5 [92]
That Breton has survived into the 21st century at all is remarkable on
several counts. Breton has never been an elite language. Latin was the
language of the Church, and French that of the Breton nobility from as
early as the 10th century. Breton was the speech of the illiterate majority.
This left Breton almost without a written literature. Then in the white heat
of revolutionary fervour, the Republic of France in 1794 sought to stamp
out regional languages and dialects in favour of Standard French. Breton
remains unrecognized today by national government as an official or
regional language.6
By contrast Irish is the national language of the Republic of Ireland, yet
the pattern of survival is similar. It is spoken as a community language
only in a few rural areas mostly in the west of the country, collectively
known as the Gaeltacht. [see 1] The measures taken for its support since
Ireland’s independence from Britain have not reversed its decline as a
first language, though it is growing as a second language. Similarly in
Scotland, Gaelic is found as a mother tongue furthest away from the
main population centres. Its main stronghold is the Outer Hebrides,
where over half the inhabitants speak it. [see 1]
92 Percentage of Breton-speakers in each Breton county in 2004. Rennes and Nantes, the main
towns of Brittany, lay outside the Breton sphere even in the 9th century, to judge from place-name
evidence.
93 The proportion of respondents in the 2011 census who said they could speak Welsh.

In Wales the decline in Welsh-speaking was inexorable over the

course of the 20th century, from nearly one million speakers down to half
that. [93] By 1991 there were none left speaking Welsh alone. Yet for the
first time ever, the 2001 census showed that the percentage of people
speaking Welsh had increased, thanks to the teaching of Welsh as a
second language in schools. It is the decline in Welsh-speaking
households that causes most concern.7

Immigrants and emigrants medieval to modern

Part of the reason for the decline of Celtic-speaking is that people do not
always choose to live in the land of their parents. Populations have not
remained static since the early medieval period. Indeed there has been
far more moving to and from the remaining Celtic-speaking regions than
can be encompassed by one short section of this book, but we must
touch on some crucial trends.
 Viking settlement left a legacy of Norse speech in Orkney and
Shetland, which survived the transfer of these islands from the Danish
crown to Scotland in 1468. This language, known as Norn, had also been
spoken in Caithness, the northeastern tip of mainland Scotland. Then
Scots-speaking settlers took their language to Caithness and the
Northern Isles [94]. In the islands it gradually overtook Norn. Pictish
areas were first Gaelicized and then Anglicized by the creation of
ethnically diverse burghs, with early Scots probably serving as a lingua
franca and replacing other languages by about 1350.8
The Norman conquest of England added a great deal of French
vocabulary to English, but left no other linguistic legacy. It was the
language of the elite. However, the subsequent Anglo-Norman incursions
into Wales and Ireland began a process of language shift to English. By
the end of the Tudor period southern Pembrokeshire was known as ‘Little
England beyond Wales’. The linguistic frontier between English and
Welsh in the county, known as the Landsker, has been remarkably stable
ever since. It is also a cultural divide, revealed in place-names and
church-types, which take us back to the period when Norman barons
seized southern Pembrokeshire.9
It was from Pembrokeshire that an exiled Irish king set sail for home in
1167, triggering the Anglo-Norman invasion of Ireland. Gerald of Wales,
a contemporary witness, blamed a fickle woman. It put him in mind of
Helen of Troy. At the time Ireland was still divided into often embattled
kingdoms. In 1152 Dermot MacMurrough, king of Leinster, abducted
Derbforgaill, the wife of the king of Breifne. According to Gerald, she was
nothing loath, having conceived a passion for Dermot. Modern historians
see politics rather than romance in the episode. Either way, the king of
Breifne held a grudge and in 1166 persuaded the high king of Ireland to
banish Dermot MacMurrough. The deposed Dermot then made the
classic error of seeking foreign aid to regain his throne. Henry II of
England did no more than grant him permission to seek allies in Henry’s
domains, on condition of Dermot’s oath of allegiance. It was Richard
FitzGilbert, Earl of Pembroke, who actually pledged troops to Dermot, in
return for Dermot’s eldest daughter to wife and the succession to his
kingdom. The result in the short term was victory for Dermot
MacMurrough in 1170. He had little time to enjoy it; he died in May 1171.
The consequences were profound. Henry II arrived in Ireland in October
1171, intent on ensuring that the gains of his vassals should come under
his lordship.10 It was the start of a long struggle over Ireland.
English settlers in Cornwall, Wales and Ireland inevitably reduced the
percentage of people there speaking Celtic languages. Elizabeth I
encouraged English settlement in Ireland. Her successor James I of
England and VI of Scotland fostered the Plantation of Ulster by colonists
from Britain. Most of these colonists came from the Scottish lowlands and
brought with them their Scots language. [94] No less important was traffic
out of Brittany, Ireland, Scotland and Wales, taking fluent Celtic-speakers
away from their linguistic community. Over the centuries many Bretons
moved into other parts of France, particularly the Paris area, in search of
wider opportunities. Similarly there is a long history of immigration from
Ireland, Scotland and Wales to England, with London attracting large
numbers. The industrial cities of the Midlands were another draw after
the Industrial Revolution. The European discovery of the Americas,
Australia and New Zealand opened up vistas for the more adventurous.
Only rarely did they form a Celtic-speaking colony, such as that of the
Patagonian Welsh in South America and the enclaves of Scottish Gaelic
in Canada. Both immigration and emigration resulted in many marriages
across linguistic lines. Where one partner is English-speaking and the
other bilingual, English is commonly the language of the home, and
becomes the mother-tongue of any children reared in it.
94 History of Scots in Scotland and Ulster. Scots developed from the Old English spoken in
Northumbria by Anglian settlers. It became so different from standard English that it has gained
official status as a regional language.

The dominance of English

There is no doubt that over the centuries there have been deliberate
attempts by government and other authorities to suppress the Celtic
languages within the British Isles. The irony is that suppression was
scarcely necessary, if the aim was an anglophone society. The seeds of
its development were already in place.
The major factors in favour of language replacement are time and
numbers. The longer two languages are in contact, the more time there is
for the speakers of one or both groups to become bilingual. If one of the
two groups is much larger than the other, the members of the smaller
group are more likely to become bilingual, which is the most common
route to the death of the minor language.11
 English-speakers have long been in the overwhelming majority within
Britain. The first census of Britain was taken in 1801. Even then, before
the great industrial boom sucked workers from the Celtic fringe into
expanding English cities, England was well ahead with over eight million
people, while Scotland had one and a half million, and Wales only half a
million. If the highlands of Britain had stunning mountain scenery, the
lowlands had fields of wheat, which could feed more mouths. London
dominated Britain’s trade. It was also the seat of political power and hub
of the judiciary. Knowledge of English was a prerequisite for participation
in education, the professions and country-wide commerce. This is a
typical pattern in which minority languages are lost.
This fundamental problem explains the limited success of modern
revivals. Gaelic has been taught in the schools of the Republic of Ireland
for decades. Yet few speak it at home. Ardent supporters of the revival of
Cornish have made an effort to learn the language, but they have to
communicate in English most of the time, since that is the majority
language. [see 100] Indeed, English has become the lingua franca of
international communications. Only determined efforts can prevent the
loss of the living Celtic languages.

Rediscovering a forgotten family

One reaction to English dominance over the Celts of the British Isles was
a defiant enthusiasm for all things Celtic. First there needed to be an
understanding that Welsh, Gaelic and Cornish are Celtic languages. This
began with the intellectual quickening of the Renaissance, with its revival
of interest in Classical sources. Any scholar familiar with the works of
Caesar and Tacitus could fit together the crucial facts: the names Celt
and Gaul were synonymous, and the British spoke a similar language to
Gaulish.12 The finest scholars did more than simply absorb Classical
knowledge though. They fell in love with logical deduction.
This blast of fresh air generated a healthy scepticism about the
pseudo-histories hitherto accepted as genuine (see Chapter 1). The
Scottish scholar George Buchanan (1506–1582) poured scorn on such
fables. [95] If Caesar and Tacitus had failed to find out the origins of the
British even after diligent enquiry, whence came this tale of Brutus as the
founder of Britain?13 The English antiquary William Camden (1551–1623)
gave an answer as good as any since. He set Brutus in the context of
other stories of eponymous founders, such as Scota for the Scots and
Danus for the Danes. These characters were invented to fill a gap in
knowledge.14

95 The Scottish scholar George Buchanan at the age of 76, painted by Arnold Bronckorst in 1581.

Buchanan’s history of the Scots was published in the year of his death,
four years before the first edition of Camden’s Britannia. So to Buchanan
goes the credit for the first detailed work of scholarship to place the
British Isles linguistically within the Celtic sphere. Using Classical sources
which recorded the Late Iron Age expansions of the Gauls, he reasoned
that they also spread into Britain. He not only recognized that place-
names incorporating -brig-and -dunum were Celtic in origin, but argued
for them being spread from Gaul.15 It was the start of a still flourishing
debate (see pp. 133–35). Buchanan’s book was a glorious polemic.
Camden described himself as ‘a plain honest and diligent searcher after
the truth’. He drew deep on Roman writers to demonstrate the similarities
of culture, religion and language between Britannia and Gaul.16
Old ideas were not completely abandoned. Buchanan could not resist
the strange tale of Pictish origin found in Bede (see p. 28). Camden drew
on Genesis, just like medieval authors. So did the next scholar to come to
our attention, which is scarcely surprising, given his vocation. Paul-Yves
Pezron (1639–1706) was a Doctor in Divinity and Abbot of La Charmoy in
France. Yet his scholarship was also marked by intensive use of
Classical sources. His interest was in the origins of the Bretons.17

96 The title-page of the first volume of Edward Lhuyd’s Archaeologia Britannica, published in
1707.
 Welsh scholar Edward Lhuyd (1660–1709), at the time keeper of the
Ashmolean Museum in Oxford, commissioned a translation into English
of Pezron’s book, which appeared in 1706.18 Lhuyd was deeply
interested in the relationship between the living languages of the Celtic
group (not that he so labelled them). In the absence of published
dictionaries for them all, he travelled through Cornwall, Scotland, Ireland
and Brittany taking notes. In 1707 the first volume of his Archaeologia
Britannica appeared. [96] It provided the first detailed account of the
Insular Celtic languages, exploring their affinity with each other and with
Gaulish. The concept of the Celtic language family was thus set on a
solid footing.

A passion for Celtic

It was a concept with a considerable romantic appeal. Other countries
might be richer in Greek and Roman antiquities, but these islands at the
end of Europe had preserved linguistic treasures, remnants of a tongue
elsewhere lost. British physician James Parsons (1705–1770) was born
in Devon but educated in Dublin and Paris. His early years in Ireland had
enabled him to learn Irish and he later learnt Welsh. His book The
Remains of Japhet (1767) is based almost entirely on Genesis and
pseudo-history. He deserves mention, though, for his fulminations
against the English arrogance that could overlook the cultural
achievements of their Celtic neighbours:

It is too much the disposition of some among us to asperse, and

set at nought, the natives of Ireland, Scotland and Wales.…
Both the Irish and the Welsh were ever well versed in the arts of
music, poetry, government and war.… In music no nation was
equal to Ireland.19

Others too were discovering what lay behind the language barrier. By
1757 the English poet Thomas Gray had penned an ode The Bard. He
based it on ‘a tradition current in Wales, that Edward the First, when he
completed the conquest of that country, ordered all the bards, that fell
into his hands, to be put to death’. It is a dramatic device to rage against
the careless obliteration of so much lyrical eloquence by the encroaching
English tongue. In the words of historian Eoin MacNeill, ‘Its weird
rhapsodical spirit contained the germ of the Celtic literary revival.’20
Travel to Wales and Scotland became popular among the English
gentry from the latter half of the 18th century. Romantic scenery and
ruined castles were eagerly sketched. Queen Victoria’s love of her
Highland hide-away at Balmoral, acquired in 1848, generated an even
greater enthusiasm in England for all things Scottish.
In Victoria’s reign there was also a revival of interest in Celtic art. It
dovetailed with renewed appreciation of the exuberance of Gothic,
despised in the previous century in favour of Neo-Classical purity. At the
end of the 19th century, Art Nouveau rebelled altogether against the
straitjacket of symmetry. There was a great attraction to undulating forms
inspired by nature, harking back to Celtic art.21 [97]
Happily, art knows no language barriers; so much in literature is lost in
translation. Organizations sprang up to battle against the submergence of
the Celtic languages. The Honourable Society of Cymmrodorion was
officially founded in London in 1751 by Richard Morris and his brother
Lewis, natives of Anglesey who had settled in London. It aimed to restore
the literary heritage of the nation.22 The Gwyneddigion Society was
another London-based Welsh literary and cultural society, founded in
1770.
The Welsh tradition of eisteddfodau, festivals of poetry and music, in
which bards and minstrels competed to win awards, was rescued almost
from the brink of the grave. In medieval times, such grand events needed
the patronage of princes. The bardic tradition petered out without the
courts that had supported it. By the 18th century it had shrunk to a few
devotees meeting in taverns. In 1789 one such devotee thought to
petition the Gwyneddigion Society for patronage. A new era began. In
September of that year an eisteddfod at Bala in Gwynedd paved the way
for the modern institution. Even so, the first National Eisteddfod series in
the 1860s [98] lost the will to promote the Welsh language, which was
seen as having no utilitarian value. There could be no greater proof of the
vulnerability of any language where money is minted in another tongue.
However, Welsh became the official language of the National Eisteddfod
in 1937.23 The result was a more confidently Welsh institution. Today this
annual festival embraces Welsh culture at its widest, including the visual
and performing arts, and a science and technology pavilion. The
message is clear. Welsh is a living language, not a fossil.

97 Silver box designed by Archibald Knox in 1903/4 for Liberty & Co., in a series inspired by the
Celtic crosses of his native Isle of Man.

98 The National Eisteddfod at Carnarvon Castle in 1862. An engraving from The Illustrated
London News.

The Welsh National Eisteddfod became the inspiration for similar

annual celebrations in Ireland, the Highlands of Scotland, Cornwall and
the Isle of Man. A Pan-Celtic spirit is reflected in the InterCeltic Festival
held at Lorient, Brittany, designed to foster contacts between the six
modern ‘Celtic nations’ of Brittany, Cornwall, Ireland, the Isle of Man,
Scotland and Wales. [99] ‘Celtic nation’ here means a people with a
Celtic language that is either still spoken or was spoken into modern
times.

99 Celtic dancers perform on 5 August 2007 in Lorient, Brittany, during the 37th Interceltic Music
Festival of Lorient.

In Ireland the Feiseanna and Oireachtas, respectively local and

national festivals for the promotion of Gaelic language and culture, were
consciously modelled on their Welsh predecessors.24 They were licensed
by the Gaelic League, founded in 1893 to preserve Irish language and
culture. The new society was the brainchild of Eoin MacNeill (1867–
1945), whom we have already met as an historian. He was also a
revolutionary. His passion for the preservation of the Irish language and
culture led him inexorably into Nationalist politics. He established the Irish
Volunteers in 1913, a step towards the battle for independence. Though
MacNeill personally opposed the idea of an armed rebellion, he was
imprisoned for a year after the Easter Rising of 1916.25 He went on to
become a politician in the Irish Free State, created in 1922 as a dominion
of the British Commonwealth. (Northern Ireland excluded itself.) This was
followed by a fully independent Éire in 1937.

Political pressures
When Celtic pride pushed into politics, alarm bells began to sound. In the
20th century the power of nation states was threatened from above and
below by pressures towards federalism and devolution. The
determination on Irish independence from Britain was an example of the
desire of a minority within a state to become a majority in its own state.
The counter urge towards federalism within Europe was driven partly
by the desire to create a body large enough to compete economically
with the massive power blocs of the US and USSR. More urgent though
was the need to end the cycle of bloodshed that had taken Europe
through two world wars. Closer co-operation between the countries of
Europe held out hope of peace. On 18 April 1951 representatives of six
nations signed the Treaty of Paris to create the European Coal and Steel
Community. The founder nations were Belgium, France, Italy,
Luxembourg, the Netherlands and West Germany. In 1957 the Treaty of
Rome created the European Economic Community (EEC) for the same
‘inner six’ nations. Other European nations could see economic
advantages in a common market. The ‘outer seven’ nations (Austria,
Denmark, Norway, Portugal, Sweden, Switzerland and the United
Kingdom) formed their own European Free Trade Association in 1960,
but most were interested in links to the EEC. Denmark, Ireland and the
United Kingdom joined the EEC on 1 January 1973. Fear of loss of
sovereignty to Brussels has been a constant theme in British politics ever
since.
At the same time the pressure for devolution was growing in Wales and
Scotland. Plaid Cymru returned its first MP in 1966 and the Scottish
National Party won its first seat at Westminster in 1967. Far more
alarming, however, was the violence in Northern Ireland. The Provisional
Irish Republican Army conducted an armed paramilitary campaign from
1969 until 1997, aimed at the forced birth of a united Ireland. Protestant
paramilitary organizations fervently against a united Ireland contributed to
the terror and lawlessness that blighted Ulster.
 Against this background the I Celti exhibition held in the Palazzo
Grassi in Venice in 1991 put a match to a powder keg of English anxiety.
It was created by French and Italian learned societies, not a political
body. Yet it was consciously designed to support the EEC with a vision of
a pan-European culture in the deep past. The Berlin Wall fell as work
began on the exhibition. So it was with hope of a reunited Europe that the
exhibition opened.26 Notably, the pan-European culture that the Palazzo
Grassi proudly displayed was not the Roman empire, which imposed
European unity by armed might. Federalism was not to be equated with
empire. On view instead were the tribal Celts. Here was a society in
which power was local, but trade was widespread.
In Spain under General Franco (dictator 1939–75) there was a strong
urge to be seen as fully European, rather than a detached part of North
Africa. (Spain’s links to North Africa began with the Phoenicians, but
were strongest in the centuries of Moorish control.) Celtic heritage was
stressed, with its links to other parts of Europe. Ideologically this was
bound up with an ultraconservative viewpoint.27 The authoritarian Franco
wanted to establish national homogeneity. He promoted the use of
Castilian Spanish and suppressed other languages such as Catalan,
Galician and Basque. He was emphatically a supporter of the nation state
against devolution. The concept of the Celts, it seems, could serve a
variety of political purposes, some of which could leave archaeologists
distinctly uneasy.28
To state what will be obvious to the reader by now, the ancient Celts
were never united in a federal system or any other. Their political
organization when they emerged into history was tribal. Those who
prepared the magnificent I Celti exhibition never pretended otherwise.
The optimistic idea was to show that modern political boundaries may not
be deeply rooted in the past. The real problem with using the Celts to
carry that message was that they were just one of a number of peoples in
Iron Age Europe. Millions of people in modern Europe do not identify
themselves as descendants of Celts.

Celtoscepticism
The backlash can be seen in a rash of works in the 1990s, which not only
(rightly) denied that the Celts had ever united Europe. They also argued
that the word ‘Celt’ should not be used at all for any peoples of the British
Isles.29 A common theme was that only those peoples who had been
firmly labelled as Celts by ancient authors should be so described. From
the scholarly perspective, this was simplistic, betraying uncritical use of
documentary sources and unfamiliarity with the fluidity of ethnonyms.
Identification labels are used to distinguish between ‘them’ and ‘us’, as
we have seen. Different labels may be used as the focus shifts to a more
inclusive ‘us’. Today a person from the north of England could see
himself as a Geordie, English, British or European, depending on the
context. The concept ‘European’ requires an understanding that there are
other continents on the planet. So Europeans in antiquity seldom
identified themselves as European. Nonetheless in our modern eyes they
were.
A familiar modern ethnonym for an ancient people may not have been
used in antiquity at all (see the ‘Italics’ below), or may have arisen in the
Roman period as a collective name for a group of tribes. The Germani
had no collective name for themselves when they were first encountered
by the Romans. They were a tribal people, for whom the important
identification was the tribal name. When Tacitus enquired of Germani the
origin of their name, he was informed that it just happened to be the
name of the tribe who first crossed the Rhine and pushed into Gaul.
While the tribe had since renamed themselves the Tungri, the name
Germani had stuck in the minds of their enemies, and been recently
adopted by the Germani themselves as the collective name for all their
tribes.30 By that time a collective name would have been useful to
distinguish themselves from the non-Germanic-speaking peoples that
they were encountering in their expansion.
The Celts of Britain and Ireland were likewise tribal. In day-to-day
affairs, a person was identified by parentage and tribe, as we find on their
memorials. Until Caesar paid his visit, the Celtic-speakers of Britain can
very seldom have encountered anyone who did not speak a Celtic
language. So there was no need for a collective name except in
distinguishing themselves from those outside their particular island. The
collective names were geographical (Iwerni, Albiones, Pritani/Cruithin
etc). By contrast the Gauls had come in contact with Greeks long before,
who labelled them Keltoi (‘the tall ones’).31 Since Britain had waves of
arrivals from Gaul before the Roman conquest, some notion of belonging
to a Celtic-speaking ‘us’ versus ‘non-Celts’ may have penetrated Britain
from Gaul, perhaps of interest to a few exceptionally well-travelled or
learned individuals, but not in common use.
Greek and Roman writers were not reliable in their grasp of the origins
even of peoples they knew well. ‘Italic’ was an obvious modern label for a
language group spoken in Italy, but it was not used in antiquity. The
Romans called themselves Roman and their language Latin. They were
familiar with neighbouring peoples such as the Oscans and Umbrians,
but without understanding their relationship to the Latins. Indeed one
minor author, himself of Gaulish origin, claimed that the Umbrians were
descended from the ancient Gauls.32 It took modern scholarship to
decipher inscriptions in the Umbrian language and classify it as related to
Latin and Oscan.33 Too much time had elapsed before the adoption of
writing for any recollection of a common origin to be preserved by Roman
authors.
So the modern discovery of the relationship between Celtic-speakers in
the British Isles and on the Continent was part of a pattern. Linguistics
was being used to shed light on prehistory and early history. In a similar
way, archaeology in the modern era has enabled scholars to delve
deeper than the written word. It has become an established principle that
archaeologists should not be led or limited by documentary sources. We
are no longer restricted to the works of ancient Greeks and Romans for
our understanding of the European past.
No doubt many archaeologists accepted Celtoscepticism as
appropriate scholarly caution, fitting the framework of an anti-migrationist
mood that had dominated Anglophone archaeology for decades. In the
post-Colonial era there was an understandable distaste for invasionist
models in archaeology. It seemed preferable to think that ideas, styles,
even languages, could move with minimal human agency. The
questioning of invasionist assumptions served a useful purpose. Invasion
is certainly not the only way that people have moved around. In recent
years a fresh look at human mobility has begun to reveal its complexity.
We have entered a new phase in the debate now that genetics can be
used to test theories about population movement or the relationship
between peoples.34
Back to the future
The most enlightening results have come from ancient DNA. What is
emerging is a story that goes beyond Celtic roots to the origins of the
entire Indo-European language family. Joining the dots of evidence, we
can draw a line from a Siberian mammoth-hunting family (box p. 70) to
many a modern-day speaker of Celtic (box pp. 26–27). It seems that the
European gene pool was stirred vigorously in the Copper Age by
migrants from the European steppe, long regarded as the most likely
homeland of the Indo-European languages. Yet the answers generate
more questions. That the Insular Celts were related to their Continental
cousins seems clear enough, but scarcely satisfies the thirst for
knowledge. Deductions about the details have been made in this book
that require testing by ancient DNA. There is much more work to be
done.

100 In 2008 new signs welcoming people to Cornwall in both English and Cornish were erected
on all ten entry roads and the Tamar Bridge (shown here). Cornish died out in the 19th century,
but there is interest in its revival.
 Neither can the preservation of living Celtic languages be taken for
granted, though there has been progress in this cause. Travel through
Wales and you will see bilingual public signage everywhere. In Scotland
that policy was first implemented in the Gaelic-speaking areas, but is
gradually spreading. In Ireland road signs are bilingual, except in the
Gaeltacht, where Irish alone is used. To stay alive languages need to be
spoken and heard. Radio and television programmes in Welsh and
Gaelic from the BBC and in Irish from RTÉ and TG4 have a crucial role
therefore. Even more important has been the steady increase in the
number of primary schools in Brittany, Wales and Ireland (including
Northern Ireland), and the opening of one on the Isle of Man, offering
education through the medium of a Celtic language. This immersion in
the language produces confident and fluent speakers. Democracy in the
1990s delivered devolution within the UK in the form of the Welsh
Assembly, Scottish Parliament and Northern Ireland Assembly, all of
which have measures in place to support the Celtic languages. So we
end on a note of hope that the story of the Celts is far from over. [100,
101]
It will be clear from this book that today’s Celtic-speakers are not stuck
in a time warp. The Celtic languages have such deep roots in the past
that constant linguistic evolution and adaptation is as much a part of their
story as their common origin. Material cultures too are always changing.
Whether they generate new phases by colliding with another culture, or
by sending shoots out into a new environment, or by inventing a better
mousetrap, cultures do not stand still. They mutate almost like living
things. Two identical cultural ‘capsules’ placed in two different locations
will immediately start to accumulate different experiences. The modern
Bretons, Cornish, Irish, Manx, Scots and Welsh cherish their own
particular identities. Though they may acknowledge some commonality,
each has its own history, partly reflected in their genes. For example a
characteristic genetic cluster can be detected in the Cornish today which
is distinct from those found in the modern Welsh.35 In short the human
past is a mixture of continuity and change. Today’s Celt has a life very
different from that of his or her ancestors 4,000 years ago. A singer then
could be heard only as far as the voice can carry. Today a song can wing
its way across the world.
101 A druidic ceremony at Névez, Finistère, Brittany, marking the springtime Beltaine festival,
symbol of renewal and fecundity, celebrated each year on 1 May.

Overview
• The percentage of fluent Celtic-speakers has declined in modern
times. Causes include the weak position of a minority language
where the majority language is required for educational, social and
economic functions, immigration and emigration, and mixed-
language households.
• Renaissance scholarship rediscovered the relationship between the
Insular and Continental Celtic languages.
• In the 18th and 19th centuries there was a resurgence of interest in
the Celtic languages, literature, art and music within the British Isles.

• Organizations were formed to support the Insular Celtic languages

and culture. Some campaigners saw a need for political change.
• In the 20th century the power of nation states was threatened from
above and below by pressures towards federalism and devolution.
• Celtoscepticism was one reaction. Anglophone archaeologists
argued that the word Celtic should not be used about Insular Celts.

• Genetics has entered the debate. DNA from human remains can
track the movement of people. It is beginning to answer some of the
burning questions about the Celts, but more work is needed to fill in
the details.

• Measures are in place to support the surviving Celtic languages.

APPENDIX

Surnames and DNA

Y-DNA is handed down from father to son, as have surnames generally

been, so can links be found between the two? This has proved a fruitful
area of research. It is not always practicable though. Certain British and
Irish surnames, such as Brown, Davies, Evans, Jones, Kelly, Murphy,
Roberts, Smith, Taylor, Thomas, Walker, Williams and Wilson, are so
common that there will be hundreds if not thousands of unrelated
lineages with the same name.
This is not surprising when we consider the origins of surnames.
Names such as Smith and Taylor are occupational, and there were many
men with the same occupation who were not closely related. Jones (son
of John) is the most common name in Wales and found so widely in
England too that just over 1 per cent of British people are so named.1
Murphy (descendant of Murchadha) is the commonest surname in Ireland
and has been carried into England and Scotland by Irish immigration.
Another common Irish name is Kelly (descendant of Ceallaigh).2 Murphy
and Kelly are found in 1.2 per cent of the Irish population each, and
genetically men of these surnames have numerous patrilineal lineages,
none of which overwhelmingly predominates. That is what we would
expect, since the surnames derive from personal names which were
common in the past.3 Rarer surnames might still have more than one
origin, but strong clusters can emerge from DNA testing that help to
distinguish between the possible origins (see box p. 208).

Origins of surnames
At the time of the Norman conquest of England in 1066, surnames in the
modern sense were unknown. A person was simply identified by a
personal name, bestowed at the font. Now and then it might be
necessary to distinguish one person from others of the same name. This
was done by a descriptive addition, known as a by-name, referring to
some striking personal quality (Robert le Gros), or occupation (Alfred the
Steward), or father’s name (Roger FitzRalph), or place of origin (John the
Dane), or place of residence (Alstan of Boscombe). There was no
consistency in this and for centuries afterwards the same person might
appear in different records with a different appellation. Only very
gradually did hereditary surnames develop from such descriptors. The
knightly class began to adopt dynastic names in the 12th century.
Surnames had filtered down to most English families by 1400, though
their form was still evolving.4
In Wales most people only began to adopt hereditary surnames under
the Tudors and even in the 19th century some men were still taking their
father’s Christian name as their surname. In the Scottish Highlands that
custom was abandoned in the 18th century, but the clan system resulted
in large numbers of people with the same surname. Chiefs of clans
increased the number of their followers by attaching men of other
descents, who took the clan name.5 In both cases we can expect many
unrelated lineages with the same surnames.
Some Irish surnames can be traced back further than any others in the
British Isles, with a few appearing in the early 10th century AD, though
most were created during the 11th and 12th centuries. Previously,
standard Gaelic naming was predominantly genealogical. A man would
be identified as mac (son of), or ua (grandson or descendant of), which
became simplified to Ó. For example O’Brien meant
grandson/descendant of Brian. Surnames in the form O’Brien could
easily be passed down to the next generation, becoming inherited
surnames. In time the Mac forms were also handed down, both in Ireland
and Scotland.

Descendants of Brian Boru

The most famous Brian in Irish history was Brian Boru, High King of
Ireland. As I write, the millennial celebration is under way of his victory at
the Battle of Clontarf on 23 April 1014, often inaccurately portrayed as
the decisive defeat of the Vikings by the Gaels. In reality the battle was a
dynastic struggle in which Irish and Vikings were allied. Brian Boru was
supported by the Limerick Vikings against the Norse of Dublin and the
men of Leinster. Brian Boru came of a minor déisi lineage in Munster who
lived around the head of the Shannon estuary. They were in the line of
devastation as Viking long-ships snaked up the Shannon in 836. At the
time, Brian’s ancestors were known as the Déis Tuaiscirt. Later they took
the name Dál Cais.6 [see 85] This lineage rocketed to power in two
generations. Brian’s father Cennétig mac Lorcán (Kennedy son of
Lorcan) was described on his death in 951 as the king of Tuadmumu
(north Munster),9 later anglicized as Thomond. Two of his sons managed
to unseat an Eóghanacht king to become king of Munster. Brian was the
second of these, and went on to take the throne of Tara from the Uí Neill.
Sadly his triumph at Clontarf was a disaster for his family. Brian and
many of his relatives were killed.11 Descendants of Brian Boru continued
to hold the kingdom of Thomond, with some interruptions, until in 1543
Murchadh Ó Briain (Murrough O’Brien) was created Baron Inchiquin and
Earl of Thomond by Henry VIII of England, in return for abandoning his
native title.12

Genetics: Irvine surname

A surname dictionary may give several possible origins for surnames such as Erwin, Irwin
or Irvine. In fact one well-researched dictionary warns against confusion between Erwin
(which can also appear as Everwin, Irwin, Irwine, Irwing and Urwin) and Irvin (with its
variants Irvine, Irving, Ervin, Erving and Urvine). The first is derived from an Old English
personal name Eoforwine (‘boar-friend’) as shown by its appearance in England in 1185 in
the form ‘William son of Irwine’. The second group of names derives from the place-name
Irvine in Ayrshire. It is first recorded in Scotland in 1226 in the form ‘Robert of Hirewyn’.7
The Clan Irwin Surname DNA Study has succeeded in identifying 30 separate genetic
families for men of these two clusters of surnames. The largest by far is a group with origins
in the Scottish Borders, who carry the rare Y-DNA haplogroup R1b1a2a1a2c1j1
(L555/S393).8 [102]
 102 A selective phylotree of Y-DNA haplogroup L21, showing those haplogroups mentioned
in the text.

Genetics: descendants of Brian Boru

The current Baron Inchiquin is Sir Conor Myles John O’Brien, whose Y-DNA has been
tested. His haplogroup was revealed to be R1b1a2a1a2c1f2a (L226/S168) [see 102], which
confirmed previous deduction that this rare marker was linked to Dál Cais kindred.10

Royal Stewart line

The movements between different parts of the British Isles over the
centuries have been so complex that the lines of descent of any one
family could zigzag across the Isles like a cat’s-cradle. Brittany must be
included in that weave. The army that William the Conqueror brought to
Britain included Bretons. Most probably some were returning to the land
of their ancestors. One particular Breton line was to rise to the Crowns of
both Scotland and England. David I of Scotland was exiled in England for
some years before taking the throne of Scotland in 1124. He spent time
at the court of Henry I (1100–1135). There he was influenced by Norman
culture and gained Norman and Breton allies.13 Walter Fitzalan (d. 1177)
entered David’s service and rose to become Steward of Scotland, which
became an hereditary office. Walter was the third son of a Breton knight,
Alan fitz Flaad, lord of Oswestry, descended from the hereditary stewards
of Dol in Brittany. Walter was granted the district which later became
Renfrewshire and extensive other lands in Scotland, where he and his
descendants settled knights and other men recruited from the Welsh
border country. Walter’s descendants eventually acquired the surname
Stewart from their office. Walter, 6th High Steward of Scotland, married
Marjorie Bruce, daughter of Robert I of Scotland. On the death without
heirs in 1371 of Marjorie’s brother, David II, his nephew Robert Stewart
took the throne of Scotland.14
The House of Stewart ended with Mary, Queen of Scots, but since she
married her cousin Henry Stewart or Stuart, Lord Darnley (descendant of
Alexander, 4th High Steward of Scotland), their son continued the line,
and brought it to England, as James VI of Scotland and I of England. One
of his descendants in the male line is the present Duke of Buccleuch. The
title was created on 20 April 1663 for James, Duke of Monmouth, the
illegitimate son of Charles II, on his marriage to Anne Scott, sister and
heiress of the 2nd Earl of Buccleuch.15
Other descendants of Alexander, 4th High Steward of Scotland,
survive, which has made it possible to cross-check the royal Stewart
DNA (see box below). They include members of the Appin Stewart
branch, descended from a grandson of Alexander.
James II of England and VII of Scotland was deposed in 1688 after a
son was born to his Catholic queen. Parliament feared the return of
Catholicism. In this they reflected the views of the majority in England,
Wales and the strongly Presbyterian Scotland. James had two Protestant
daughters, Mary and Anne, who reigned in turn after him, followed by the
Protestant descendants of James I of England and VI of Scotland. The
deposed king found some of his strongest support in the Scottish
highlands among his own kin. The Stewarts of Appin loyally supported
his cause. They were involved in the Jacobite risings claiming the throne
for the former king’s son James Francis (‘The Old Pretender’) and
grandson Charles Edward (‘The Young Pretender’ or ‘Bonny Prince
Charlie’). Charles Stewart of Ardsheal, head of a cadet branch of Appin,
led the Appin Regiment at the Battle of Culloden in 1746. It ended in a
rout for the Young Pretender which effectively finished the Jacobite
cause. The Appin Regiment suffered massive losses and Charles
Stewart of Ardsheal had to flee the country to escape execution. A
descendant of his is now clan chief of the Appin Stewarts.

Genetics: Royal Stewart

In 2012 the DNA of the 10th Duke of Buccleuch was tested. His Y-DNA haplogroup was
found to be an exact match to that of a descendant of Charles Stewart of Ardsheal. The
haplogroup is R1b1a2a1a2c1i1a (L744/S388).16 It fits on the end of a chain downwards
from the widespread British and Breton marker R-L21. [see 102]

Genetics: MacFarlane
The great majority of men who carry R1b-L21 fall into the subclade defined by the marker
DF13. The brother marker DF63 is found much more rarely. Within the latter branch is a
subclade defined by CTS6919. [102] A number of MacFarlane and MacFarland men carry
this haplogroup, some of whom have a documented lineage from the Arrochar line.17

Clan MacFarlane
The region around lovely Loch Lomond is now Dunbartonshire. It was
once the district of Lennox. The old Scottish title Mormaer of Lennox was
held by a family with origins before the era of surname development.
Their personal names were Gaelic, apart from that of the 1st and 2nd
Earls of Lennox, both named Alwyn. This obviously Germanic name has
caused confusion. Scholars battled over whether Alwyn I was a Celt or
an Anglo-Saxon. It seems he was both. Paternally Alwyn I was a son of
Murdac, Mormaer of Lennox. Maternally he was a grandson of an Alwyn,
most likely Alwyn MacArkyl, who was prominent at the court of King
David I (1124–1153). A poem probably written to celebrate his coming of
age describes him as Alwyn the younger, which could cause further
confusion, since he became Alwyn the elder after his son Alwyn was
born. In the poem he is being distinguished from his maternal
grandfather.18
In the 13th century Gilchrist, a younger son of Alwyn II, was granted
the feudal barony of Arrochar between Loch Long and Loch Lomond by
his eldest brother, the 3rd Earl of Lennox. Gilchrist became the ancestor
of Clan MacFarlane. The surname derives from Gilchrist’s great-
grandson Phàrlan, 4th Baron of Arrochar. His son was Malcolm Mac
Phàrlan. MacPhàrlan, later spelled MacFarlane, then became an
hereditary surname. Since Phàrlan was an uncommon personal name,
there is less likelihood of unrelated MacFarlanes than is the case with
names such as Donald and Gregor.
In the 16th century, the MacFarlanes were caught up in the drama of
the royal succession. Protestants and Catholics battled over the choice of
marriage partner for the infant Mary, Queen of Scots. Duncan
MacFarlane of Arrochar supported the Earl of Lennox against the
Catholic party in the Battle of Glasgow Muir in 1544. He was a staunch
supporter of the Reformation. However, when the Protestant Henry VIII
sought to gain by force the marriage of Mary to his infant son in 1547,
Duncan fought and died for Scotland at the Battle of Pinkie Cleugh.
His descendant William MacFarlane sold Arrochar in 1784. Other
MacFarlanes had earlier settled in Ulster, and moved from there to the
US, while another branch chose to make their home in Co. Dublin.19 This
scattered clan provides a test case for the usefulness of DNA testing (see
box on p. 211.)
Notes
Prologue

1. Tacitus, Germania, 4.
2. Tacitus, Agricola, 11.
3. MacNeill 1920, 1–3.

Chapter 1: The Voices of the Celts

1. Fled Bricrend 27–29.

2. Koch and Carey 2003, 324.
3 Ptolemy, II.2; Rivet and Smith 1979, 508–09; Koch 2006, 824.
4 Driscoll and Yeoman 1997, 220–26.
5. Koch and Carey 2003, 326.
6. Rendered in Welsh as Golistan.
7. Koch and Carey 2003, 323.
8. Lapidge, Blair, Keynes and Scragg 1999, 163, 503.
9. Fraser 2009, 131.
10. Koch and Carey 2003, 322, stanza B2.26.
11. Koch and Carey 2003, 53.
12. Koch 2006, 997–98.
13. Meyer 1911, 99, from St Gall MS 904.
14. Skene 1868, I, 523
15. Caesar, VI.13–14.
16. The Light Fantastic (1986).
17. Geoffrey of Monmouth, VIII. 10–12.
18. Koch and Carey 2003, 59.
19. Koch and Carey 2003, 59–63, 108, 110–13.
20. Koch 2006, 352–53.
21. Bieler 1979, Muirchú, Life of Saint Patrick.
22. Carney 2005, 451, 454.
23. Adamnan.
24. Clancy and Márkus 1995, 104–15.
25. Lives of Saints Declan and Mochuda, 93.
Adamnan, chapter 35.
 Adamnan, chapter 35.
26.
27. Táin Bó Cúailnge.
28. Annals of Tigernach, 404, 407.
29. Mac Carron and Kenna 2012.
30. Walsh 1991.
31. Lynn 1989.
32. Hurl, McSparren and Moore 2002.
33. Koch 2006, 261–64.
34. Ó hUiginn 2006.
35. Jackson 1964.
36. Mallory 1992.
37. Koch and Carey 2003, 158.
38. Ó Corráin 2006.
39. Baillie 1995, 65–67.
40. Warner 1990.
41. Mallory 1993.
42. Gildas, II.4.
43. Historia Brittonum.
44. Ó Cróinín 2005, 182.
45. Crick 2004.
46. Geoffrey of Monmouth, dedication.
47 Historia Regum Britanniae, ed. Wright, xvi.
48. William of Newburgh, 29–37; Gerald of Wales, The Journey Through Wales, 1.5.
49. Historia Regum Britanniae, ed. Wright, xviii.
50. Thorpe 1966, 50. The manuscript is BL Harleian MS 3859, ff. 174v–198r.
51. Rivet and Smith 1979, 396–98, 443–44.
52. Isidore, IX.ii.102.
53. Summerfield 2011.
54. Genesis 6–10.
55. Ross 1981.
56. Flavius Josephus, I.6.
57. Reynolds 1983.
58. Ross 1981, 23.
59. Lebor Gabála Érenn, I.10–11.
60. Ross 1981.
61. Isidore, IX.ii.26.
62. Koch 2006, 775.
63. Flavius Josephus, I.6.
64. Novembre 2008; Tian 2008; Tian 2009; O’Dushlaine 2010.
65. Schiffels 2014; Felix Jeyareuben Chandrakumar for the Y-DNA haplogroup allocation from
raw data accompanying Schiffels.
Bede, chapter 1.
66. Bede, chapter 1.

67. Historia Brittonum III, 13, 15; Carey 1994.

68. Ó Cróinín 2005, 185–86; Carey 1994; Koch 2006, 739
69. Lebor Gabála Érenn, vol. 2 (Irish Texts Society 35), 252.

Chapter 2: The Gauls and Celtic

1. Koch and Carey 2003, 12–13: translation by Philip Freeman of Diodorus Siculus, Library of
History, 5.28–30.
2. Koch and Carey 2003, 8–9, translation of Polybius, History, 2.28.3–10.
3. Caesar, I.1; and see Pausanias, 1.4.1.
4. Sims-Williams 2006, 180–82.
5. Stephani Byzantini, 539, 581. For a translation see Koch and Carey 2003, no. 3.
6. Herodotus, II.33, IV, 49.
7. Aristotle, Meteorology, I.13.350.
8. Smith 1854, I, 106.
9. Herodotus, IV, 49.
10. Cunliffe 1997, 28–31.
11. Koch (ed.) 2006, 887.
12. Hummler 2007; Cunliffe 1997, 31.
13. Livy, V.34–35.
14. Cunliffe 1997, 32; Frey 1995, 520.
15. Koch 2006, 832–33.
16. Ritchie 1995.
17. Koch 2007, §83.
18. Moore and Chiriotti 2010.
19. Ó hUiginn 2006.
20. Caesar, V.15–19.
21. Tacitus, Annals, 14.35, trans. J. T. Koch and J. Carey in Koch and Carey 2003, 43.
22. Koch and Carey 2003, 322, stanza B2.37.
23. Adamnan, VII.
24. Greene 1972; Mallory 1998.
25. Nutt 1900.
26. Koch and Carey 2003, 90, 139–43. MacKillop 2004, under Liath Macha.
27. Homer, Iliad, XVI.
28. Walters Art Museum, Maryland.
29. Isidore, XVIII.xxxvi.
30. de Leeuw 2008.
31. Greene 1972; Mallory 1998; Karl 2003; Karl 2006.
32. Berggren and Jones 2000.
Talbert 2000, 118, 155; Rivet and Smith 1979, 437–38.
 Talbert 2000, 118, 155; Rivet and Smith 1979, 437–38.
33.
34. Lejars 2003.
35. Koch and Carey 2003, 13–14: translation by Philip Freeman of Diodorus Siculus, Library of
History, 5.31.
36. Morley 2008.
37. Jay 2013.
38. Jay 2012.
39. Jay 2012.
40. Boyle 2008.
41. Carter and Hunter 2003; Carter 2010.
42. Polybius II, 29.
43. Caesar, VII.81.
44. Information from National Museums Scotland, where both the original and the replica are
displayed.
45. Gilbert, Brasseur, Dalmont and Maniquet 2012.
46. Ptolemy, II.6.
47. Gilbert, Brasseur, Dalmont and Maniquet 2012.
48. Caesar, VI.17.
49. Gilbert, Brasseur, Dalmont and Maniquet 2012.
50. Caesar, V.12.
51. Koch 2007, 19; Mallory 2013, 163.
52. Herodotus, I.163; IV.152; Roller 2006, 34–36.
53. Aubet 2001, 33–34, 162, 259–62; Mata 2001; Aubet 2008; Deamos 2009.
54. Freeman 2010, 312.
55. Herodotus, I.163. I use the Greek version of his name here.
56. Mallory and Adams 2006, 242.
57. Prósper 2010/2011, 58 and fn. 5.
58. Herodotus, I.163–65 and note.
59. Manco 2015, 26–27: table 1.
60. Scozzari 2012. The mutation V20 defined C7 at that point, which was renamed C6 and
then C1a2 in 2014.
61. Koch 2010 and 2011.
62. Valério 2014.
63. Roller 2006, 9–11, 74–77; Koch 2006, 1270.
64. Avienus, trans. Murphy, lines 90–100 and p. 86. Strabo, I. 4.5 and IV, 4.1 records the
names used by Pytheas.
65. Freeman 2001, Kindle locations 706–11.
66. Koch 2006, 38–39, 709; Rivet and Smith 1979, 40.
67. Rivet and Smith 1979, 247–48; Koch 2006, 38–39.
68. Hubert 1934, 186; Dillon and Chadwick 1967, 4; Corcoran 1970, 24; Anthony 2007, 367;
Cunliffe 2010, 34.

Chapter 3: Bell Beakers and Language

Chapter 3: Bell Beakers and Language

1. Childe 1930, 4–5.

2. Hubert 1934, 186; Dillon and Chadwick 1967, 4; Corcoran 1970, 24; Anthony 2007, 367;
Cunliffe 2010, 34.
3. Manco 2015, chapter 10.
4. Sherratt 1987.
5. Childe 1958, 223.
6. Dietrich 2012.
7. Dineley 2004.
8. Rojo-Guerra 2006.
9. Guerra-Doce 2006.
10. Czebreszuk and Szmyt 2003.
11. Shepherd 2012.
12. Sarauw 2007 and 2008.
13. Østmo 2012; Prescott 2012.
14. Identification by Roger Taylor, Royal Albert Memorial Museum & Art Gallery, Exeter.
15. Parker Pearson 1995.
16. Querré and Convertini 1998.
17. Salanova 2011.
18. Vander Linden 2007.
19. O’Brien 2004.
20. Northover, O’Brien and Stos 2001.
21. Warner, Moles and Chapman 2010.
22. Pare 2000.
23. Senna-Martinez 2011.
24. Sofia Martinez, O Primeiro Alquimista: A Idade do Bronze em Portugal (2012).
25. Carlin and Brück 2012.
26. Jeunesse 2014.
27. Turek 2006; Fitzpatrick 2013; Peška 2013; Vergnaud 2013.
28. Peška 2013; Prieto-Martínez 2012.
29. Fitzpatrick 2009; Fitzpatrick 2013, 50–53; Darvill 2006, 151.
30. Maryon 1936; Taylor 1980, 22; Independent 5 August 2014.
31. Nocete 2014.
32. Chapman 2006; Warner, Moles and Chapman 2010.
33. Taylor 1994.
34. Armbruster 2013.
35. Heath 2012, 10–12, 39.
36. Fitzpatrick 2013, 53.
37. Jay, Parker Pearson 2012.
38. Price, Knipper, Grupe and Smrcka, 2004.
Menk 1979; Cox and Mays 2000, 281–83; Nicolis 2001, 2, 403; Budziszewski, Haduch and
 Menk 1979; Cox and Mays 2000, 281–83; Nicolis 2001, 2, 403; Budziszewski, Haduch and
39. Włodarczak 2003.
40. Lee 2012.
41. 1000 Genomes Project Consortium 2010; Myres 2011; Wei 2013; Sikora, Colonna, Xue
and Tyler-Smith 2013.
42. Sims-Williams 2006, 17.
43. Talbert 2000, 18.
44. Ptolemy, II.2; Koch 2007, maps 15.2 and 21.2.
45. Rivet and Smith 1979, 129.
46. Fraser 2009, 20.
47. Sims-Williams 2006, 179.
48. Koch and Carey 2003, 8.
49. Fortson 2010, 276–77.
50. Watkins 1966.
51. Kortlandt 1981.
52. Cowgill 1970; Ringe, Warnow and Taylor 2002; Nakhleh, Ringe and Warnow 2005;
Schrijver 2006; Hamp and Adams 2013.
53. Manco 2015, 166–67, revised.
54. De Alarcão 2001.
55. Krahe 1963.
56. Mees 2003.
57. Villar 2000, 2004.
58. Kitson 1996.

Chapter 4: The Indo-European Family

1. Jones, Works, III, 34. Spellings modernized by the present author.

2. ODNB; Canon and Franklin 2004.
3. Muller 1986; Canon and Franklin 2004; Mallory and Adams 2006, 4–6.
4. Mallory 1989, 143.
5. Renfrew 1987.
6. Diamond and Bellwood 2003.
7. Mallory and Adams 2006, 101–03, 166, 241, 260–62.
8. Gray and Atkinson 2003; Bouckaert 2012.
9. Anthony 2013; Anthony and Ringe 2015; Pereltsvaig and Lewis 2015.
10. Chang, Cathcart, Hall and Garrett 2015.
11. Bryce 2005, 11–12, 21–39.
12. Watkins 2001; Goedegebuure 2008; Josephson 2012.
13. Mallory 1989; Anthony 2007; Anthony and Ringe 2015.
14. Anthony 2007, 72–75; Anthony and Ringe 2015.
15. Carpelan, Parpola and Koskikalio 2001.
16. Huehnergard 2011; Militarev 2005.
Nichols 1997, 125–28; Mallory and Adams 2006, 82–83.
 Nichols 1997, 125–28; Mallory and Adams 2006, 82–83.
17.
18. Anthony 2007, 287–97.
19. Anthony 2007, 94–95; Häkkinen 2012.
20. Vasil’ev 1999.
21. Inizan 2012.
22. Clark 1982.
23. Inizan 2012.
24. Hartz, Terberger and Zhilin 2010.
25. Stupak 2006; Smyntyna 2007.
26. Altinbilek, Astruc, Binder and Pelegrin 2012.
27. Raghavan 2014.
28. Gatsov and Nedelcheva 2011.
29. Manco 2015, 96–98; Anthony 2007, 171 fig. 9.3.
30. Anthony 2007, 148–49.
31. Anthony 2007, chapters 8–9.
32. Nikitin 2012.
33. Mallory and Adams 2006.
34. Anthony 2007, chapter 13.
35. Anthony and Chi 2009.
36. Haas 1998; O’Brien 1995.
37. Manco 2015, 113.
38. Lazaridis 2014.
39. Haak 2015.
40. Nikitin 2012; Wilde 2014.
41. Haak 2015.
42. Lazarovici 2010; Anthony 2007, 164–74.
43. Sherratt 1981; Bogucki 1993; Greenfield 2010; Marciniak 2011.
44. Manzura 2005a, 327.
45. Parpola 2008.
46. Kohl 2007, 45–46; Korvin-Piotrovskiy 2012.
47. Kirtcho 2009.
48. Outram 2009; Anthony 2007, chapter 10.
49. Anthony 2007, 287–93; Chernykh 2008.
50. Morgunova and Khokhlova 2013.
51. Anthony 2007, chapter 13.
52. Anthony 2007, 328–39.
53. Harrison and Heyd 2007, chapter 9; Kristiansen 2005.
54. Homer, Iliad, 17.51–52.
55. Harrison and Heyd 2007, chapter 9; Kristiansen 2005.
56. Anthony 2007, 43–48, 75, 249–59, 260–62, fig. 13.11; Anthony 2013.
57. Thissen 1993; Bauer 2006.
58. Kremenetski, Chichagova and Shishlina 1999; Kremenetski 2003.

59. Anthony 2007, 64–65, 307–11; Anthony 2013.

60. Mallory and Mair 2000.
61. Anthony 2007, chapter 14; Anthony 2008; Wlodarczak 2009; Manco 2015, 131–32.
62. Haak 2015.
63. Kroonen 2012.
64. Anthony 2007, 361–67; Heyd 2011.
65. Martirosyan 2013; Mallory and Adams 1997, 26–30.
66. Sherratt 1986.
67. Manzura 2005b; Sherratt 1986; Andreu 2010, 646; Koukouli-Chrysanthaki and
Papadopoulos 2009.
68. Anthony 2007, 343–48.
69. Anthony 2007, 371–82, 389–411, 452–57.
70. Kuzmina 2007, 379–413.
71. Keyser 2009.

Chapter 5: Stelae to Bell Beaker

1. Tertullian, 57, citing Nicander of Colophon, a Greek author of the 2nd century BC.
2. Labaune 2013, 181, fig. 3.
3. Robb 2009.
4. Harrison and Heyd 2007.
5. Mallory 1989, 203–05; Telegin and Mallory 1994; Mallory and Adams 1997, 544–46;
Anthony 2007, 268–71, 291, 320–21, 338–39, fig 13.11.
6. Mallory 1989, 203–05; Telegin and Mallory 1994, 23.
7. Schloen and Fink 2009; Pardee 2009; Struble and Herrmann 2009.
8. Telegin and Mallory 1994.
9. Zavaroni 2009.
10. Heyd 2011, 541.
11. Gogâltan 2013.
12. Höppner 2005.
13. Mazzieri and Dal Santo 2007.
14. Maggi and Pearce 2005.
15. Mallory and Adams 1997, 217–18, 317–18, 482–83, 485–86.
16. Manco 2015, 113.
17. Dolfini 2010.
18. Dolfini 2013.
19. De Marinis 1998.
20. De Saulieu 2013.
21. Velušček 2004;Čufar, Kromer, Tolar and Velušček 2010.
22. Carozza and Mille 2007; Roberts 2009.
Bernard, Carles, Picavet and Morin 2005.
23. Bernard, Carles, Picavet and Morin 2005.
24. Nocete 2006; Nocete 2011, 3278–95; Hanning, Gauß and Goldenberg 2010; Roberts 2008.
25. Lira 2010.
26. Bendrey 2012.
27. Kunst 2007.
28. Cardoso 2000.
29. Müller 2007.
30. Diaz-Guardamino Uribe 2010.
31. Gallay 1978 and 1995.
32. Harrison and Heyd 2007.
33. Harrison and Heyd 2007, 134–35, 142, 146, 151, 161–64.
34. Harrison and Heyd 2007, 147, 149, 160.
35. Telegin and Mallory 1994, 40–41.
36. Harrison and Heyd 2007, 158–59, 170–71.
37. Harrison and Heyd 2007, 154–55, 171–72.
38. Vander Linden 2013; Beckerman 2012.
39. Müller and van Willigen 2001; Wlodarczak 2009; Cardoso 2014.
40. Heyd, Husty and Kreiner 2004; Lechterbeck 2014.
41. Salanova 2008.
42. Bettencourt and Luz 2013.
43. Kunst 2001; Carvalho-Amaro 2013.
44. Kunst 2001; Ferreira 2003; Carvalho-Amaro 2013.
45. Turek 2012.
46. Among Cucuteni and Svobodnoe types: Anthony 2007, figs. 11.4, 12.6, 12.9.
47. Odriozola and Hurtado Pérez 2007; Curtis, Popovic, Wilson and Wright 2010; Všianský,
Kolář and Petříka 2014.
48. Parkinson 2010; Roberts, Sofaer and Kiss 2008.
49. Cardoso 2014.
50. Cunliffe and Koch 2010; Koch and Cunliffe 2013.
51. Hyllested 2010.
52. Manco 2015, 183.
53. Hickey 2002; Shisha-Halevy 2003.
54. Matasović 2008 and 2012.
55. Lemercier 2012; Salanova 2004.
56. Gibson 2013, 76.
57. Endro˝di and Horváth 2006.
58. Endro˝di 2013.
59. Price, Knipper, Grupe and Smrcka 2004.
60. Desideri and Besse 2010.
61. Anthony 2007, 183, 250: fig. 11.10, 256, 298; Ruzickova 2009.
62. Piguet and Besse 2009; Kulcsár and Szeverényi 2013.
63. Sheridan 2008.
64. Heyd 2007.
65. Harrison and Heyd 2007, 185–87, 192.
66. Chiaradia, Gallay and Todt 2003; Menk 1979.
67. Lemercier 2012.
68. Garrido Pena 1997; Rios 2013; Kulcsár and Szeverényi 2013.
69. Koch 2006, 364–65, 374.
70. Gibson 2014.
71. Fitzpatrick 2013, 56–58 and fig. 2.8; Gibson 2013, 78.
72. Fokkens, Achterkamp and Kuijpers 2008; Woodward and Hunter 2011.
73. De Bernardo Stempel 2007; Nicolaisen 1982.
74. Mallory 2013, chapter 9, particularly p. 261.
75. Stevens and Fuller 2012; Whitehouse 2014.
76. Rocca 2012.
77. Burgess and O’Connor 2008.
78. Kristiansen 1998, 144; Henderson 2007, chapter 3; Cunliffe 2008, 254–58.
79. Tinsley 1981; Turner 1981.
80. Lorrio and Zapatero 2005, 221–27; De Alarcão 2001; Burgess and O’Connor 2008.
81. Manco 2015, 166–67.
82. De Alarcão 2001.
83. Prósper 2014; García Quintela 2005.

Chapter 6: The Iron Sword

1. Aristotle, Nicomachean Ethics, 3.7.

2. Manti and Watkinson 2008.
3. Mödlinger 2013.
4. Bouzek 2001; Makhortyk 2008.
5. Makhortyk 2008.
6. Kristiansen 1998, 137.
7. Isaac 2010.
8. Kristiansen 1998, 161, 233.
9. Pare 1991.
10. Kristiansen 1998, 211–16; Manning 1995.
11. Kristiansen 1998, 254–63.
12. Fernández-Götz and Krausse 2013.
13. Fernández-Götz and Krausse 2013.
14. Olivier 1999; Verger 2006.
15. Piggott 1995.
16. Green 1996, 73.
Eluère 1991.
 Eluère 1991.
17.
18. Diodorus Siculus, V.22.
19. Rolley 2003.
20. Aratus, lines 206ff.
21. Olivier 1999; Verger 2006.
22. Tacitus, Agricola, 16; Tacitus, Annals, 14.31.
23. Chaume and Reinhard 2011.
24. Chaume and Reinhard 2011.
25. Minerva 7 (5), 6.
26. Moscati 1991, 34, 136, 499–500.
27. Knipper 2014.
28. Knipper 2014.
29. De Marinis 1991; Uhlich 2007.
30. Caesar, IV.10.
31. Pliny, III.20.
32. Pliny, III.1.
33. De Marinis 1991.
34. De Marinis 1991.
35. Uhlich 2007.
36. Koch 2006, 463–64.
37. Koch 2006, 708–08.
38. Ptolemy, II.2, 7–8.
39. Ellis Evans 1995.
40. Koch 1992; Sims-Williams 2007; Matasović 2007.
41. Frey 1991, 129.
42. Fernández-Götz and Krausse 2013.
43. Cunliffe 1997, 63–64.
44. Koch 2007, §17.3, §17.4, §18.
45. Furger-Gunti 1991.
46. Karl 2003.
47. Megaw and Megaw 1995; Harding 2007, 63.
48. Megaw and Megaw 1995.
49. Karl 2006.
50. Maier 1991; Collis 1995, 161–62, 172.

Chapter 7: On the Move

1. Justin, XXIV.4.
2. Ammianus Marcellinus, 15.9.4.
3. Caesar, I.1; II.4.
4. Koch 2006, 195–99.
Koch 2007, maps § 17.3 and § 84.
 5. Koch 2007, maps § 17.3 and § 84.
6. Caesar, VI.44.
7. Halkon 2011.
8. Justin, XX.5.
9. Livy, V.34
10. Tinner 2003.
11. Livy, V.33.
12. Dionysius of Halicarnassus, XIII.11
13. Pliny the Elder, 12.2.
14. Diodorus Siculus, V. 26.
15. Caesar, II.15.
16. Diodorus Siculus, XV.70.
17. Polybius, II.17.
18. Dionysius of Halicarnassus, XIII.12; Livy V.34–35.
19. Frey 1995.
20. Polybius, II.17.
21. Livy V.34–35; Sims-Williams 2006, 91, 199.
22. Grassi 2011.
23. Vitali 1991; Frey 1995.
24. Uhlich 2007.
25. Karl 2006;Čižmář 1991; Bujnar and Szabó 1991.
26. Božič 1991; Guštin 2011; Jovanović 1991.
27. Polyaenus, XII.42; Theopompus, F40 – see Shrimpton 1991, appendix, 221.
28. Strabo, 7.3; Arrian, I.4.
29. Božič 1991; Guštin 2011.
30. Arrian, VII.15.
31. Pausanias, I.4.1; Diodorus Siculus, XX.19.1; Justin, XV.2.
32. Justin, XXIV.4–5.
33. Diodorus Siculus, XXII.9.1-3: Justin, XXIV.6-8; Pausanias, X.19.8–23.13.
34. Athenaeus, VI.234a–c.
35. Polybius, IV.46; Pausanias, X.19.7; Livy, XXXVIII.16.
36. Livy, XXXVIII.16.
37. Ammianus Marcellinus, XV.9.
38. Livy, XXXVIII.16.
39. Rankin 1987, chapter 9; Koch 2007, §19.2.
40. Strabo, IV.1.13.
41. Strabo, XII.5.1.
42. Koch 2006, 1350–51.
43. Acts 16:6.
44. NPNF2:6, 497.
45. Kelly 1975, 25–26, 37.
46. Eska 2006.
47. Lorrio and Zapatero 2005.
48. García Alonso 2006.
49. Lorrio and Zapatero 2005.
50. Villar 2004, 248.
51. Pliny, III.3.
52. Strabo, 3.1.6, 3.3.5; Talbert 2000, map 26.
53. Luján Martínez 2006.
54. Villar 2004, 256, 267–68; Talbert 2000; http://pleiades.stoa.org.
55. Turner 1981.
56. Cruciani, Trombetta and Antonelli 2011.
57. Chiaroni, Underhill and Cavelli-Sforza 2009.
58. Bertoncini 2012; Capocasa 2014.
59. Armit, Swindles and Becker 2013.
60. Chapman 2006.
61. Armit, Swindles and Becker 2013.
62. Mallory 2013, 185.
63. Caulfield 1977.
64. De Bernardo Stempel 2007.
65. MacNeill 1920, chap. 4.
66. Toner 2000; Darcy and Flynn 2008.
67. Busby 2012.
68. Moore 2006.
69. Lacey 2006.
70. Busby 2012.
71. Personal communication David Powell.
72. Caesar, V.12.
73. Koch 2007, 21, 109–10 and maps 50, 76, 252–67.
74. Ptolemy, II.2; II.8; Koch 2006, 357–58; Koch 2007, map 15.6; Cunliffe 2005.

Chapter 8: Celts vs Romans

1. Polybius, I.1.
2. Polybius, II.19–35.
3. Markoe 2000, 181–82.
4. Polybius, III, 9–11.
5. Polybius, III.34, 44, 82–118.
6. Arslan 1991.
7. Manco 2015, 182–83.
8. Polybius, III.15, 17, 20; Livy, XXI, 1–15.
9. Polybius, III.97–99.
Polybius, X.2, 6.
10. Polybius, X.2, 6.
11. Polybius, X.6–15, 18, 34–38.
12. Livy, XXVIII, 1–2; Polybius, XI. 20–24a, 33.
13. Livy, XXVI. 50.
14. Polybius, X.17.
15. Silva 2013.
16. Livy, XLIII.2.
17. Diodorus Siculus, XXXIII.1; Cassius Dio, XXII.73; Justin, XLIV.2.
18. Plutarch, Caesar, 12.
19. Caesar, I.2–4, VII.4.
20. Lintott 1999, 109–13.
21. Plutarch, Caesar, 3–6, 14.
22. Caesar, I.2–8.
23. Karl 2001.
24. Caesar, I.9–29.
25. Caesar, I.30–32; VI.11–12; Strabo IV.3.2.
26. Caesar, I.34–54.
27. Caesar, II.1.
28. Caesar, II.3.
29. Caesar, II.3–4
30. Caesar, II.8–35.
31. Caesar, VII.4.
32. Caesar, VII.
33. Caesar, VII.89.
34. Plutarch, Caesar, 27.
35. Caesar, IV.20–38; V.1–23.
36. Tacitus, Agricola, 13.
37. Res Gestae Divi Augusti, 32; Manley and Rudkin 2005.
38. Koch 2006, 520.
39. Cassius Dio, 60.19–20 refers to Berikos, generally taken to be Vericos.
40. Tacitus, Agricola, 12.
41. Tacitus, Annals, XII, 36-8.
42. Tacitus, Histories, III.45.
43. Tacitus, Histories, III.45; Tacitus, Agricola 17; Tacitus, Annals XII.40.
44. Raftery 2005, 174–76.
45. Ó Floinn 2000.
46. Koch 2006, 750.
47. Tacitus, Agricola, 22–38.
48. Tacitus, Agricola, 14, and note 48; Ptolemy, II.2; Koch 2007, map 15.6; Koch 2006, 1362.
49. Ptolemy, II.2; Koch 2007, maps §15.2 and §21.2; Fraser 2009, 20.
50. Herodian, III.14.
51. Cassius Dio, LXXVII.12.
52. Fraser 2009, 15–17.
53. Panegyrici Latini VI: Panegyric of Constantine 226–27 and note 27.
54. Isidore, XIX.xxiii.7.
55. Caesar, V.14; Pomponius Mela, III, 51.
56. Moffat and Wilson 2011, 159.
57. Freeman 2001.
58. Ammianus Marcellinus, XXVI.4.5; XXVII.8; Rance 2001.

Chapter 9: Christian Celts

1. Patrick, Confession, 1.
2. Gildas, 27–33.
3. Patrick, Confession, 9.
4. Thomas 1981, 46–48, 197.
5. Tomlin 1992, 16–17.
6. Rees 1998.
7. Murray 2003, IV.16: Chronicle of Prosper of Aquitaine.
8. Murray 2003, IV.16: Chronicle of Prosper of Aquitaine.
9. McManus 1996.
10. Ó Cróinín 1995, 33–34, 305.
11. Mallory 2013, 263.
12. Ó Cróinín 1995, 53.
13. Koch 1995.
14. Koch 2007, 18, §20, §391; Edwards 2007.
15. Ptolemy, II.2.
16. Gildas, III.31.
17. Edwards 2007; Koch 2007, §391, no. 145; Charles-Edwards 2013, 174–75.
18. Expulsion of the Dessi 112-13.
19. Archaeologia Cambrensis, 5th series, 9 (1892), 64–65.
20. Murray 2003, IV.16: Chronicle of Prosper of Aquitaine.
21. Rance 2001. See Thornton 2003, chapter 5 for a more critical review of the evidence.
22. Ó Cróinín 1995, 19.
23. Historia Brittonum, III.14.
24. Patrick, Confession.
25. O’ Brien 1962, 253.
26. Family Tree DNA: R1b L21 and Subclades Project.
27. Hey 2000, 93.
28. Ó Cróinín 1995, 20–23.
29. Bieler 1979 includes Muirchú’s Life of Saint Patrick.
30. Rivet and Smith 1979, 511–12.
31. Ó Cróinín 1995, 23-27.
32. Annals of the Four Masters.
33. Rivet and Smith 1979, 280–82.
34. Pliny, IV, 30; Ptolemy, II, 1–2.
35. Ó Cróinín 1995, chapter 2.
36. Muhr 1996, 1–7, 151–53.
37. Annals of Ulster.
38. Ó Cróinín 1995, 48.
39. O’Laverty 1878, lvii, 35.
40. McEvoy and Bradley 2010, 117.
41. Family Tree DNA I-M223 Y-Haplogroup Project; Moffat and Wilson 2011, 24–25; estimated
dates supplied by Kenneth Nordtvedt.
42. Koch 1995.
43. Koch 2006, 738–39.
44. Annals of Ulster; Auraicept Na N-Éces, lines 63-78; Lacey 2006, 229–32.
45. Gildas, 23.
46. Murray 2003, IV.17: Gallic Chronicle.
47. Gildas, 23.
48. Bede, 15.
49. Lane 2014.
50. Schrijver 2007.
51. Gildas, 25.
52. Murray 2003, IV.17: Gallic Chronicle.
53. Winney 2012; Leslie 2015.
54. Schiffels 2014.
55. Bede, 193; Manco 1998, 31.
56. Lapidge, Blair, Keynes and Scragg 1999, 93.
57. Laws of the Earliest English Kings 36–61.
58. Koch 2006, 532.
59. Asser, 14.
60. Charles-Edwards 2013, 14–17.
61. Charles-Edwards 2013, 20–21; Rivet and Smith 1979, 434.
62. Edwards 2013.
63. Koch 2006, 518–19.
64. Charles-Edwards 2013, 176–79.
65. Rivet and Smith, 491.
66. Koch 2007, §24.
67. Gildas, 33.
68. Koch 2007, §391, no. 90; Edwards 2013; Charles-Edwards 2013, 476.
69. Historia Brittonum, section 62; Koch 2006, 518–20.
70. Koch 2006, 1261–62.
Fraser 2009, 153; Charles-Edwards 2013, 190.
71. Fraser 2009, 153; Charles-Edwards 2013, 190.
72. Thornton 2003, chapter 4.
73. Gildas, III.28.
74. Aldhelm: the Prose Works, Letter IV.
75. Anglo-Saxon Chronicle.
76. Koch 2006, 275–78, 750.
77. Orme 2000, 1, 214–19.
78. Koch 2006, 1558.
79. Koch 2006, 291.
80. Woolf 2007, chapter 8.
81. Campbell 2001.
82. Bede, 11.
83. Fraser 2009, 145–46, 156–60, 203–06.
84. Bede, 11.
85. Fraser 2009, 54, 239.
86. Woolf 2007, 93–98, 110–25, 220, 320–21.
87. Fraser 2009, chapter 4.

Chapter 10: Loss and Revival

1. Pezron trans. Jones 1706, 1–2.

2. Jones 1997.
3. Jenner 1904, 19–22.
4. Broderick 1999, 5, 75.
5. Moseley 2010.
6. Ternes 1992, 372–77.
7. Jones 2012.
8. Johnston 1997.
9. John 1972.
10. Gerald of Wales, Conquest of Ireland; Ó Cróinín 1995, 285–88.
11. Thomason 2001, 66 and chapter 9.
12. Caesar, I.1; Tacitus, Agricola, 11.
13. Buchanan 1582, II, 1–2, 6–8.
14. Camden 1607, I.
15. Buchanan 1582, II.
16. Camden 1607, I.
17. Pezron trans. Jones 1706, i.
18. Pezron trans. Jones 1706.
19. Parsons 1767, ix–x.
20. MacNeill 1920, 7.
21. Moscati 1991, 29–30.
Koch 2006, 527–29.
22. Koch 2006, 527–29.
23. Edwards 2006.
24. Koch 2006, 737–38.
25. Ó Croidheáin 2006, 138, 151–53.
26. Moscati 1991, foreword.
27. Burillo Mozota 2005.
28. Dietler 1994.
29. Chapman 1992; Dietler 1994; Collis 1997 and 2003; James 1999.
30. Tacitus, Germania, chapter 2.
31. De Bernardo Stempel 2008.
32. Solinus, 2.11, citing Marcus Antonius [Gnipho], a grammarian of Gaulish origin who taught
in Rome in the 1st century BC; repeated by Isidore, IX.ii.87.
33. Buck 1904.
34. Manco 2015.
35. Leslie 2015.

Appendix: Surnames and DNA

1. McKie 2006, 174.

2. Reaney and Wilson 1997.
3. McEvoy and Bradley 2006.
4. Hey, 2000, 31, 51–53.
5. Reaney and Wilson 1997, Introduction; Redmonds, King and Hey 2011, 2–3.
6. Ó Cróinín 2005, 237, 266–67; Duffy 2013.
7. Reaney and Wilson 1997, 157, 249.
8. http://dnastudy.clanirwin.org
9. Annals of Ulster.
10. Wright 2009; Family Tree DNA: R-L226 Project; Dennis Wright personal communication.
11. Ó Cróinín 2005, 276; Duffy 2013.
12. Complete Peerage, XII (1), 702–03.
13. Chibnall 2000, 69.
14. Barrow 2004.
15. Complete Peerage, II, 366.
16. Family Tree DNA: Stewart Stuart DNA Project.
17. Family Tree DNA projects MacFarlane and DF63 and subclades.
18. Complete Peerage, VII, 585–93.
19. Burke and Burke 1847, II, 800–01; MacFarlane 1922.
Bibliography
Abbreviation: PNAS = Proceedings of the National Academy of Sciences of the United States of
America
1000 Genomes Project Consortium. 2010. A map of human genome variation from population-
scale sequencing, Nature, 467, 1061–73.
Adamnan. Life of Saint Columba by Adomnán of Iona, ed. and trans. R. Sharpe. 1995. London:
Penguin.
Aldhelm: the Prose Works, ed. and trans. M. Lapidge and M. Herren. 1979. Ipswich: D.S. Brewer.
Allen, M. J., Gardiner J. and Sheridan, A. (eds). 2012. Is There a British Chalcolithic? People,
Place and Polity in the Later 3rd Millennium. Oxford and Oakville, CT: The Prehistoric
Society/Oxbow Books.
Altinbilek-Algül, C., Astruc, L., Binder, D., Pelegrin, J. 2012. Pressure blade production with a
lever in the Early and Late Neolithic of the Near East, chapter 5 in Desrosiers (ed.) 2012, 157–
79.
Ammianus Marcellinus. Roman History, trans. C. D. Yonge. 1862. London: Bohn.
Andreu, S. 2010. The Northern Aegean, in E. H. Cline (ed.), The Oxford Handbook of the Bronze
Age Aegean, 643–59. New York: Oxford University Press.
Anglo-Saxon Chronicle, ed. and trans. M. Swanton. 1996. London: J. M. Dent. Annals of the Four
Masters, ed. and trans. John O’Donovan, Annala Rioghachta Eireann: Annals of the kingdom of
Ireland by the Four Masters, from the earliest period to the year 1616. Edited from MSS in the
Library of the Royal Irish Academy and of Trinity College Dublin with a translation and copious
notes, 7 vols. 1848–51. Dublin: Hodges and Smith.
Annals of Tigernach, trans. W. Stokes, Revue Celtique, ed. H. D’Arbois de Jubainville, 16–18
(1895–97). Digital edition: CELT: Corpus of Electronic Texts. Annals of Ulster. The Annals of
Ulster (to A.D. 1131), ed. and trans. S. Mac Airt and G. Mac Niocaill. 1983. Dublin: Dublin
Institute for Advanced Studies.
Anthony, D. W. 2007. The Horse, the Wheel and Language: How Bronze Age Riders from the
Eurasian Steppes Shaped the Modern World. Princeton and Oxford: Princeton University Press.
Anthony, D. W. 2008. A new approach to language and archaeology: the Usatovo Culture and the
separation of Pre-Germanic, Journal of Indo-European Studies, 36 (1–2), 1–51.
Anthony, D. W. 2013. Two IE phylogenies, three PIE migrations, and four kinds of steppe
pastoralism, Journal of Language Relationship, 9, 1–22.
Anthony, D. W. and Chi, J. Y. (eds). 2009. The Lost World of Old Europe: The Danube Valley,
5000–3500 BC. New York and Princeton: Institute for the Study of the Ancient World/Princeton
University Press.
Anthony, D. W. and Ringe, D. 2015. The Indo-European homeland from linguistic and
archaeological perspectives, Annual Review of Linguistics, 1, 199–219.
Aratus. Phaenomena. Included in A. W. Mair and G. R. Mair (trans.), Callimachus, Hymns and
Epigrams. Lycophron. Aratus. 1921 (Loeb Classical Library). London: William Heinemann.
Aristotle. Meteorology, in Works of Aristotle, trans. E. W. Webster, vol. 3. 1931. Oxford:
 Aristotle. Meteorology, in Works of Aristotle, trans. E. W. Webster, vol. 3. 1931. Oxford:
Clarendon Press.
Aristotle. Nicomachean Ethics, vol. 19 of Aristotle in 23 Volumes, trans. H. Rackham. 1934.
Cambridge, MA: Harvard University Press; London: William Heinemann Ltd.
Armbruster, B. 2013. Gold and gold-working of the Bronze Age, chapter 25 in The Oxford
Handbook of the European Bronze Age, A. Harding and H. Fokkens (eds), 454–68. Oxford:
Oxford University Press.
Armit, I., Swindles, G. T. and Becker, K. 2013. From dates to demography in later prehistoric
Ireland? Experimental approaches to the meta-analysis of large 14C data-sets, Journal of
Archaeological Science, 40 (1), 433–38.
Arrian. The Anabasis of Alexander, trans. E. J. Chinnock. 1884. London: Hodder and Stoughton.
Arslan, E. 1991. The Transpadane Celts, in Moscati et al. (eds) 1991, 461–70.
Asser. Alfred the Great: Asser’s Life of King Alfred and Other Contemporary Sources. S. Keynes
and M. Lapidge (trans.). 1983. Harmondworth: Penguin Books.
Athenaeus. Deipnosophistae, trans. Charles Burton Gulick. 1927–41. 7 vols (Loeb Classical
Library). Cambridge, MA: Harvard University Press.
Aubet, M. E. 2001. The Phoenicians and the West: Politics, Colonies and Trade (2nd ed.).
Cambridge: Cambridge University Press.
Aubet, M. E. 2008. Political and economic implications of the new Phoenician chronologies, in
Beyond the Homeland: Markers in Phoenician Chronology (Ancient Near Eastern Studies,
Supplement 28), C. Sagona (ed.), 247–59. Leuven and Dudley, MA: Peeters.
Auraicept Na N-Éces, ed. and trans. G. Calder. 1917. Edinburgh: John Grant.
Avienus, Rufus Festus. Ora Maritima: A Description of the Seacoast from Brittany to Marseilles
[Massilia], trans. J. P. Murphy. 1977. Chicago: Ares Publishing.
Baillie, M. G. L. 1995. A Slice Through Time: Dendrochronology and Precision Dating. London:
Batsford.
Barrow, G. W. S. 2004. Stewart family (per. c. 1110–c. 1350), Oxford Dictionary of National
Biography.
Bauer, A. A. 2006. Between the steppe and the sown: prehistoric Sinop and interregional
interaction along the Black Sea coast, chapter 12 in Beyond the Steppe and the Sown, D. L.
Peterson, L. M. Popova and A.T. Smith (eds), 225–46. Leiden and Boston: Brill.
Beckerman, S. 2012. Dutch beaker chronology re-examined, Palaeohistoria 53/54, 25–64.
Bede. The Ecclesiastical History of the English People, The Greater Chronicle, Bede’s Letter to
Egbert, ed. and trans. J. McClure and R. Collins. 1994. Oxford and New York: Oxford University
Press.
Bendrey, R. 2012. From wild horses to domestic horses: a European perspective, World
Archaeology, 44 (1), 135–57.
Berggren, J. L. and Jones, A. 2000. Ptolemy’s Geography: An Annotated Translation of the
Theoretical Chapters. Princeton and Woodstock: Princeton University Press.
Bernard, C., Carles, J., Picavet, R. and Morin, A. 2005. Etude préliminaire sur des poignards
gravés de type Remedello découverts dans les Préalpes du Sud (Chastel-Arnaud, Drôme,
France) et réflexions sur leur insertion dans le Néolithique final régional, Bulletin de la Société
préhistorique française, 102 (2), 345–59.
Bertoncini, S. et al. 2012. A Y variant which traces the genetic heritage of Ligures tribes, Journal
of Biological Research, 84 (1), 143–46.
Bettencourt, A. M. S. and Luz, S. 2013. A corded-mixed Bell Beaker vase at the monumental
enclosure of Forca, Maia, North of Portugal, chapter 1 in Prieto-Martínez and Salanova (eds)
2013, 15–20.
Bieler, L. (ed. and trans.). 1979. The Patrician Texts in the Book of Armagh. Scriptores Latini
Hiberniae, 10. Dublin: Dublin Institute for Advanced Studies.
 Hiberniae, 10. Dublin: Dublin Institute for Advanced Studies.
Bogucki, P. 1993. Animal traction and household economies in Neolithic Europe, Antiquity, 67
(256), 492–503.
Bouckaert, R. et al. 2012. Mapping the origins and expansion of the Indo-European language
family, Science, 337, 957–60.
Bouzek, J. 2001. Cimmerians and early Scythians: the transition from geometric to early
orientalising style in the North Pontic area, in North Pontic Archaeology: Recent Discoveries
and Studies (Colloquia Pontica 6), G. R. Tsetskhladze (ed.), 33–44. Leiden, Boston and Köln:
Brill.
Boyle, A. et al. 2008. Site D (Ferry Fryston) in the Iron Age and Romano-British periods, in The
Archaeology of the A1(M) Darrington to Dishforth DBFO Road Scheme, F. Brown et al., 121–
59. Oxford: Lancaster Imprints, Oxford Archaeology North.
Božič, D. 1991. The Taurisci, in Moscati et al. (eds) 1991, 471–77.
Broderick, G. 1999. Language Death in the Isle of Man: An Investigation into the Decline and
Extinction of Manx Gaelic as a Community Language in the Isle of Man. Tubingen: Niemeyer.
Bryce, T. 2005. The Kingdom of the Hittites. Oxford and New York: Oxford University Press.
Buchanan, G. 1582. Rerum Scoticarum Historia. Translated as The History of Scotland, 1690.
Buck, C. D. 1904. A Grammar of Oscan and Umbrian. Boston: Ginn.
Budziszewski, J., Haduch, E. and Włodarczak, P. 2003. Bell Beaker Culture in South-Eastern
Poland, in The Northeast Frontier of Bell Beakers: Proceedings of the symposium held at the
Adam Mickiewicz University, Poznañ (Poland), May 26–29 2002, J. Czebreszuk and M. Szmyt
(eds). BAR International Series 1155, 155–81. Oxford: Archaeopress.
Bujnar, J. and Szabó, M. 1991. The Carpathian Basin, in Moscati et al. (eds) 1991, 277–85.
Burgess, C. and O’Connor, B. 2008. Iberia, the Atlantic Bronze Age and the Mediterranean, in
Contacto cultural entre el Mediterráneo y el Atlántico (siglos XII-VIII ANE). La precolonización a
debate (Serie Arqueológica 11), S. Celestino, N. Rafel, and X.-L. Armada (eds), 41–58. Escuela
Española de Historia y Arqueología en Roma-CSIC. Madrid.
Burillo Mozota, F. 2005. Celtiberians: Problems and Debates, e-Keltoi: the Journal of
Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 411–80.
Burke, J. and Burke, J. B. 1847. Genealogical and Heraldic Dictionary of the Landed Gentry of
Great Britain and Ireland. London: Henry Colburn.
Busby, G. B. J. et al. 2012. The peopling of Europe and the cautionary tale of Y chromosome
lineage R-M269, Proceedings of the Royal Society B: Biological Sciences, 279 (1730), 884–92.
Caesar. The Conquest of Gaul, trans. S. A. Handford, rev. J. F. Gardner. 1982. London and New
York: Penguin Books.
Camden, W. 1607. Britannia, trans. P. Holland. London.
Campbell, E. 2001. Were the Scots Irish?, Antiquity, 75, 285–92.
Campbell, J., John, E. and Wormwald, P. 1982. The Anglo-Saxons. London and New York:
Penguin Books.
Canon, G. and Franklin, M. J. 2004. A Cymmrodor claims kin in Calcutta: an assessment of Sir
William Jones as philologer, polymath, and pluralist, Transactions of the Honourable Society of
Cymmrodorion, New Series, 11, 50–69.
Capocasa, M. et al. 2014. Linguistic, geographic and genetic isolation: a collaborative study of
Italian populations, Journal of Anthropological Sciences, 92, 201–31.
Cardoso, J. L. 2000. The fortified site of Leceia (Oeiras) in the context of the Chalcolithic in
Portuguese Estramadura, Oxford Journal of Archaeology, 19 (1), 37–55.
Cardoso, J. L. 2014. Absolute chronology of the Beaker phenomenon north of the Tagus estuary:
demographic and social implications, Trabajos De Prehistoria, 71 (1), 56–75.
Carey, J. 1994. The Irish National Origin-Legend: Synthetic Pseudohistory. Cambridge: University
Carey, J. 1994. The Irish National Origin-Legend: Synthetic Pseudohistory. Cambridge: University
of Cambridge.
Carlin, N. and Brück, J. 2012. Searching for the Chalcolithic: continuity and change in the Irish
Final Neolithic/Early Bronze Age, in Allen, Gardiner and Sheridan (eds) 2012, 193–210.
Carney, J. 2005. Language and literature to 1169, chapter 13 in Ó Cróinín (ed.) 2005, 451–510.
Carozza, L. and Mille, B. 2007. Chalcolithique et complexification sociale: quelle place pour le
métal dans la définition du processus de mutation des sociétés de la fin du Néolithique en
France? in Le Chalcolithique et la construction desinégalités, 1, le continent européen, J.
Guilaine (ed.), 195–232. Paris: Editions Errance.
Carpelan, C., Parpola, A. and Koskikalio, P. (eds). 2001. Early Contacts between Uralic and Indo-
European: Linguistic and Archaeological Considerations: Papers Presented at an International
Symposium Held at the Tvärminne Research Station of the University of Helsinki, 8–10 January
1999 (Mémoires de la Société Finno-Ougrienne 242). Helsinki: Finno-Ugrian Society.
Carter, S. and Hunter, F. 2003. An Iron Age chariot burial from Scotland, Antiquity, 77 (297), 531–
35.
Carter, S. et al. 2010. A 5th century BC Iron Age chariot burial from Newbridge, Edinburgh,
Proceedings of the Prehistoric Society, 76, 31–74.
Carvalho-Amaro, G. 2013. Pre-Bell Beaker ware from Estremadura, Portugal, and its likely
influence on the appearance of Maritime Bell Beaker ware, chapter 18 in Prieto Martínez and
Salanova, L. (eds) 2013, 197–208.
Cassius Dio. Roman History, trans. E. Cary. 1914–27. 9 vols (Loeb Classical Library) Cambridge,
MA: Harvard University Press.
Caulfield, S. 1977. The Beehive Quern in Ireland, The Journal of the Royal Society of Antiquaries
of Ireland, 107, 104–38.
Celtic Inscribed Stones: Online database created by University College London, 1999.
Chang, W., Cathcart, C., Hall, D. and Garrett, A. 2015. Ancestry-constrained phylogenetic
analysis supports the Indo-European steppe hypothesis, Language, 91 (1) Chapman, M. 1992.
The Celts. The Construction of a Myth. London and New York: St Martin’s Press.
Chapman, R. J. et al. 2006. Microchemical characterisation of natural gold and artefact gold as a
tool for provenancing prehistoric gold artefacts: a case study in Ireland, Applied Geochemistry,
21 (6), 904–18.
Charles-Edwards, T. M. 2013. Wales and the Britons 350–1064. Oxford: Oxford University Press.
Chaume, B. and Reinhard, W. 2011. Les statues du sanctuaire de Vix-Les Herbues dans le
contexte de la statuaire anthropomorphe hallstattienne, in Gruat and Garcia (eds) 2011, 293–
310.
Chernykh, E. 2008. The ‘steppe belt’ of stock-breeding cultures in Eurasia during the Early Metal
Age, Trabajos de Prehistoria, 65 (2), 73–93.
Chiaradia, M., Gallay, A. and Todt, W. 2003. Different contamination styles of prehistoric human
teeth at a Swiss necropolis (Sion, Valais) inferred from lead and strontium isotopes, Applied
Geochemistry, 18, 353–70.
Chiaroni, J., Underhill, P. and Cavalli-Sforza, L. L. 2009. Y chromosome diversity, human
expansion, drift and cultural evolution, PNAS, 106 (48), 20174–79.
Chibnall, M. 2000. The Normans. Malden, MA: Wiley-Blackwell.
Childe, V. G. 1930. The Bronze Age. London and New York: Cambridge University Press.
Childe, V. G. 1958. The Dawn of European Civilization (6th ed.). New York: Alfred A. Knopf.
Čižmář, M. 1991. The Celtic population of Moravia in the fourth century B.C., in Moscati et al.
(eds) 1991, 273–76.
Clancy, T. O. and Márkus, G. 1995. Iona: The Earliest Poetry of a Celtic Monastery. Edinburgh:
Edinburgh University Press.
 Edinburgh University Press.
Clark, J. E. 1982. Manufacture of Mesoamerican prismatic blades: an alternative technique,
American Antiquity, 47 (2), 355–76.
Collis, J. 1995. The first towns, chapter 10 in Green 1995, 159–75.
Collis, J. 1997. Celtic myths, Antiquity, 71 (271), 195–201.
Collis, J. 2003. The Celts: Origins, Myths & Inventions. Stroud: Tempus Publishing.
Complete Peerage = The Complete Peerage of England, Scotland, Ireland, Great Britain and the
United Kingdom by G. E. C [okayne], ed. V. Gibbs et al. (2nd ed.) 12 vols. 1910–1959. London:
The St Catherine Press.
Corcoran, J. X. W. P. 1970. The origins of the Celts: the archaeological evidence, chapter 1 in N.
Chadwick, The Celts. Harmondsworth and New York: Penguin.
Cowgill, W. 1970. Italic and Celtic superlatives and the dialects of Indo-European, in G. Cardona
et al. (eds), Indo-European and the Indo-Europeans, 113–53. Philadelphia: University of
Pennsylvania Press. Reprinted in J. Klein (ed.), 2006. The Collected Writings of Warren Cowgill.
New York: Beech Stave Press.
Cox, M. and Mays, S. 2000. Human Osteology in Archaeology and Forensic Science. London:
Greenwich Medical Media.
Crick, J. C. 2004. Monmouth, Geoffrey of (d. 1154/5), Oxford Dictionary of National Biography.
Oxford: Oxford University Press.
Cruciani, F., Trombetta, B., Antonelli, C. et al. 2011. Strong intra-and inter-continental
differentiation revealed by Y chromosome SNPs M269, U106 and U152, Forensic Science
International: Genetics, 5(3), e49–52.
Čufar, K., Kromer, B., Tolar, T. and Velušček, A. 2010. Dating of 4th millennium BC pile-dwellings
on Ljubljansko barje, Slovenia, Journal of Archaeological Science, 37, 2031–39.
Cunliffe, B. 1997. The Ancient Celts. Oxford and New York: Oxford University Press.
Cunliffe, B. 2005. Iron Age Communities in Britain: An Account of England, Scotland and Wales
from the Seventh Century BC until the Roman Conquest (4th ed.). Abingdon: Routledge.
Cunliffe, B. 2008. Europe Between the Oceans: Themes and Variations: 9000 BC–AD 1000. New
Haven and London: Yale University Press.
Cunliffe, B. 2010. Celticization from the West, chapter 1 in Cunliffe and Koch (eds) 2010, 13–38.
Cunliffe, B. and Koch, J. T. (eds). 2010. Celtic from the West: Alternative Perspectives from
Archaeology, Genetics, Languages and Literature (Celtic Studies Publications 15). Oxford:
Oxbow Books.
Curtis, N., Popovic, L., Wilson, N. and Wright, M. 2010. The moon, the bonfire and the beaker?
Analysing white inlay from Beaker pottery in Aberdeenshire, Past: The Newsletter of the
Prehistoric Society, 65, 1–3.
Czebreszuk, J. (ed.). 2004. Similar but Different: Bell Beakers in Europe. Poznan: Adam
Mickiewicz University. Reprinted in 2014. Leiden: Sidestone Press.
Czebreszuk, J. and Szmyt, M. (eds). 2003. The Northeast Frontier of Bell Beakers: Proceedings
of the symposium held at the Adam Mickiewicz University, Poznañ (Poland), May 26–29 2002
(BAR International Series 1155). Oxford: Archaeopress.
Darcy, R. and Flynn, W. 2008. Ptolemy’s map of Ireland: a modern decoding, Irish Geography, 41
(1), 49–69.
Darvill, T. 2006. Stonehenge: The Biography of a Landscape. Stroud: Tempus. De Alarcão, J.
2001. Novas perspectivas sobre os Lusitanos (eoutrosmundos), Revista Portuguesa De
Arqueologia, 4, 2.
Deamos, M. B. 2009. Phoenicians in Tartessos, chapter 8 in Dietler and López-Ruiz (eds) 2009,
193–228.
De Bernardo Stempel, P. 2007. Pre-Celtic, Old Celtic layers, Brittonic and Goidelic in ancient
De Bernardo Stempel, P. 2007. Pre-Celtic, Old Celtic layers, Brittonic and Goidelic in ancient
Ireland, in Language Contact in the Place-names of Britain and Ireland, P. Cavill and G.
Broderick (eds), 137–63. Nottingham: English Place-Name Society.
De Bernado Stempel, P. 2008. Linguistically Celtic ethnonyms: towards a classification, in Celtic
and Other Languages in Ancient Europe, J. L. García Alonso (ed.), 101–18. Ediciones
Universidad Salamanca.
de Leeuw, H. 2008. Chariots on High Crosses: Celtic or Christian?, The Journal of the Royal
Society of Antiquaries of Ireland, 138, 5–25.
De Marinis, R. 1991. Golasecca Culture and its links with Celts beyond the Alps, in Moscati et al.
(eds) 1991, 93–102.
De Marinis R. C. 1998. The eneolithic cemetery of Remedello Sotto (BS) and the relative and
absolute chronology of the Copper Age in Northern Italy, Notizie Archeologiche Bergomensi, 5,
33–51.
De Saulieu, G. 2013. Rock carvings and Alpine statue-menhirs, from the Chalcolithic to the Middle
Bronze Age, chapter 16 in The Oxford Handbook of the European Bronze Age, A. Harding and
H. Fokkens (eds), 291–310. Oxford: Oxford University Press.
Desideri, J. and Besse, M. 2010. Swiss Bell Beaker population dynamics: eastern or southern
influences?, Archaeological and Anthropological Sciences, 2 (3), 157–73.
Desrosiers, P. M. (ed.). 2012. The Emergence of Pressure Blade Making: From Origin to Modern
Experimentation. New York and London: Springer.
Diamond, J. and Bellwood, P. 2003. Farmers and their languages: the first expansions, Science,
300 (5619), 597–603.
Díaz-Guardamino Uribe, M. 2010. Las Estelas Decoradas en la Prehistoria de la Península
Ibérica, PhD Thesis, Madrid.
Dietler, M. 1994. ‘Our Ancestors the Gauls’: archaeology, ethnic nationalism, and the
manipulation of Celtic identity in modern Europe, American Anthropologist, 96 (3), 584–605.
Dietler, M. and López-Ruiz, C. (eds). 2009. Colonial Encounters in Ancient Iberia: Phoenician,
Greek and Indigenous Relations. Chicago and London: University of Chicago Press.
Dietrich, O. et al. 2012. The role of cult and feasting in the emergence of Neolithic communities.
New evidence from Göbekli Tepe, south-eastern Turkey, Antiquity, 86, 674–95.
Dillon, M. and Chadwick, N. 1967. The Celtic Realms. London: Weidenfeld & Nicolson.
Dineley, M. 2004. Barley, Malt and Ale in the Neolithic (BAR S1213). Oxford: Archaeopress.
Diodorus Siculus. The Library of History, trans. C. H. Oldfather. 1933–67. 12 vols (Loeb Classical
Library). Cambridge, MA: Harvard University Press.
Dionysius of Halicarnassus. Roman Antiquities, trans. E. Cary. 7 vols. 1937–50. (Loeb Classical
Library) Cambridge, MA: Harvard University Press.
Dolfini, A. 2010. The origins of metallurgy in central Italy: new radiometric evidence, Antiquity, 84
(325), 707–23.
Dolfini, A. 2013. Early Metallurgy in the Central Mediterranean, chapter 15 in Archaeometallurgy
in Global Perspective, B. W Roberts and C. P. Thornton (eds), 473–506. New York: Springer.
Driscoll, S. T. and Yeoman, P. 1997. Excavations within Edinburgh Castle in 1988–91
(Monograph Series no. 12). Edinburgh: Society of Antiquaries of Scotland.
Duffy, S. 2013. Brian Boru and the Battle of Clontarf. Dublin: Gill & Macmillan.
Edwards, H. T. 2006. Eisteddfod and Eisteddfod Genedlaethol Cymru, in Koch (ed.) 2006, 664–
68.
Edwards, N. 2007. A Corpus of Early Medieval Inscribed Stones and Stone Sculpture in Wales, 2:
SouthWest Wales. Cardiff: University of Wales Press.
Edwards, N. 2013. A Corpus of Early Medieval Inscribed Stones and Stone Sculpture in Wales, 3:
North Wales. Cardiff: University of Wales Press.
 North Wales. Cardiff: University of Wales Press.
Ellis Evans, D. 1995. The early Celts: the evidence of language, chapter 2 in Green (ed.) 1995, 8–
20.
Eluère, C. 1991. The Celts and their gold: origins, production and social role, in Moscati et al.
(eds) 1991, 349–55.
Endrὅdi, A. 2013. Recent data on the settlement history and contact system of the Bell Beaker–
Csepel Group, in Moments in Time: Papers Presented to Pál Raczky on his 60th Birthday, A.
Anders and G. Kulcsár (eds), 693–705. Ősrégészeti Tanulmányok / Prehistoric Studies, I.
Endrὅdi, A. and Horváth, A. 2006. Kora Bronzkori arany korong Csepel–Szigetrὅl, Budapest
Régiségei, 40, 21–30.
Eska, J. 2006. The Galatian language, in Koch (ed.) 2006, 788.
Expulsion of the Dessi, The, ed. K. Meyer, Y Cymmrodor, 14 (1901), 101–35.
Fernández-Götz, M. and Krausse, D. 2013. Rethinking Early Iron Age urbanisation in Central
Europe: the Heuneburg site and its archaeological environment, Antiquity 87, 473–87.
Ferreira, S. D. 2003. Os Copos no Povoado Calcolítico de Vila Nova de São Pedro, Revista
Portuguesa de Arqueologia, 6 (2), 181–228.
Fitzpatrick, A. 2009. In his hands and in his head, The Amesbury Archer as a metalworker, in
Bronze Age Connections: Cultural contact in prehistoric Europe, P. Clark (ed.), 176–88. Oxford:
Oxbow Books.
Fitzpatrick, A. P. 2013. The arrival of the Bell Beaker set in Britain and Ireland, chapter 2 in Koch
and Cunliffe (eds) 2013, 41–70.
Flavius Josephus. Antiquities of the Jews, trans. William Whiston. 1895. London.
Fled Bricrend [The Feast of Bricriu], trans. G. Henderson. 1899. London: David Nutt for The Irish
Texts Society.
Fokkens, H., Achterkamp, Y. and Kuijpers, M. 2008. Bracers or bracelets? About the functionality
and meaning of Bell Beaker wrist-guards, Proceedings of the Prehistoric Society, 74, 109–40.
Fortson, B. W. 2010. Indo-European Language and Culture: An Introduction (2nd ed.).
Chichester: John Wiley & Sons.
Fourdringier, E. 1878. Double sépulture gauloise de la Gorge Meillet, territoire de Somme-Tourbe
(Marne). Paris and Châlons sur Marne: privately published.
Fraser, J. E. 2009. From Caledonia to Pictland: Scotland to 795. (The New Edinburgh History of
Scotland 1) Edinburgh: Edinburgh University Press.
Freeman, P. 2001. Ireland and the Classical World. Austin, TX: University of Texas Press. Kindle
edition.
Freeman, P. 2010. Ancient references to Tartessos, chapter 10 in Cunliffe and Koch (eds) 2010.
Frey, O.-H. 1991. The formation of the La Tène culture in the fifth century BC, in Moscati et al.
(eds) 1991, 127–45.
Frey, O.-H. 1995. The Celts in Italy, chapter 27 in Green (ed.) 1995.
Furger-Gunti, A. 1991. The Celtic war chariot: The experimental reconstruction in the
Schweizerisches Landesmuseum, in Moscati et al. (eds) 1991, 356–59.
Gallay, A. 1978. Stèles néolithiques et problématique archéologique, Archives Suisses
d’Anthropologie Générale 42 (2), 75–103.
Gallay, A. 1995. Les stèles anthropomorpes du site mégalithique du Petit-Chasseur à Sion
(Valais, Suisse), in Statue-stelee massi incisi nell’Europa dell’Età del Rame, S. Casini, R. C. De
Marinis and A. Pedrotti (eds), 167–94, Notizie Archeologiche Bergomensi 3.
García Alonso, J. L. 2006. -Briga Toponyms in the Iberian Peninsula, e-Keltoi: the Journal of
Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 689–714.
García Quintela, M. V. 2005. Celtic Elements in Northwestern Spain in Pre-Roman times, e-Keltoi:
the Journal of Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 497–569.
 the Journal of Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 497–569.
Garrido Pena, R. 1997. Bell Beakers in the southern Meseta of the Iberian Peninsula:
socioeconomic context and new data, Oxford Journal of Archaeology, 16, 187–209.
Gatsov, I. and Nedelcheva, P. 2011. Neolithic chipped stone assemblages in North Western
Anatolia, Turkey, Eurasian Prehistory, 8 (1–2), 89–95.
Geoffrey of Monmouth. History of the Kings of Britain, trans. A. Thompson, revised J. A. Giles.
1999. Cambridge, Ontario: In Parentheses Publications.
Gerald of Wales. The Journey Through Wales and the Description of Wales, trans. L. Thorpe.
1978. London and New York: Penguin.
Gerald of Wales. The Conquest of Ireland, in The Historical Works of Giraldus Cambriensis, trans.
F. Forester, ed. T. Wright. 1863. London: Henry G. Bohn.
Gibson, C. 2013. Beakers into Bronze: tracing connections between western Iberia and the British
Isles 2800–800 BC, chapter 3 in Koch and Cunliffe (eds) 2013.
Gibson, C. 2014. Closed for business or cultural change? Tracing the re-use and final blocking of
megalithic tombs during the Beaker period throughout Atlantic Europe. Paper presented at the
conference Atlantic Europe in the Metal Ages, Cardiff, 12 April 2014.
Gilbert, J., Brasseur, E., Dalmont J.-P. and Maniquet, C. 2012. Acoustical evaluation of the
Carnyx of Tintignac, Proceedings of the Acoustics 2012 Nantes Conference, 3955–59.
Gildas. The Ruin of Britain, and other works, trans. M. Winterbottom. 1978. London: Phillimore.
Goedegebuure, P. M. 2008. Central Anatolian languages and language communities in the colony
period: a Luwian-Hattian symbiosis and the independent Hittites, in Anatolia and the Jazira
during the Old Assyrian period, J. G. Dercksen (ed.) (Old Assyrian Archives, Studies, 3.
PIHANS 111), 137–80. Leiden: The Netherlands Institute for the Near East.
Gogâltan, F. 2013. Transilvaniași spațiul Nord-Pontic. relații interculturale între sfârșitul Epocii
Cupruluiși începutul Epocii Bronzului (CCA. 3500–2500 A. Chr.) Terra Sebus. Acta Musei
Sabesiensis, 5, 31–76.
Grassi, E. 2011. Attività metallurgiche a Mediolanum tra archeologia e archeometria: Metallurgy in
Mediolanum between archaeology and archaeometry, Archeologia Uomo Territorio, 30.
Gray, R. and Atkinson, Q. 2003. Language-tree divergence times support the Anatolian theory of
Indo-European origin, Nature, 426, 435–39.
Green, M. (ed.). 1995. The Celtic World. Abingdon: Routledge.
Green, M. A. 1996. Celtic Art: Reading the Messages. London: Weidenfeld & Nicolson.
Greene, D. 1972. The chariot as described in Irish literature, in Iron Age in the Irish Sea Province,
C. Thomas (ed.), CBA Research Report 9, 59–73. London: Council for British Archaeology.
Greenfield, H. J. 2010. The Secondary Products Revolution: the past, the present and the future,
World Archaeology, 42 (1), 29–54.
Gruat, P. and Garcia, D. (eds). 2011. Stèles et statues du début de l’âge du Fer dans le Midi de la
France (VIIIe-IVe s. av. J.-C.): chronologies, fonctions et comparaisons. Documents
d’Archéologie Méridionale, 34.
Guerra-Doce, E. 2006. Exploring the significance of Beaker pottery through residue analyses,
Oxford Journal of Archaeology, 25 (3), 247–59.
Guštin, M. 2011. On the Celtic tribe of Taurisci: local identity and regional contacts in the ancient
world, in The Eastern Celts: The Communities Between the Alps and the Black Sea (Annales
Mediterranei), M. Guštin and M. Jevtić (eds), 119–28. Koper: Univerza na Primoerskem.
Haak, W. et al. 2008. Ancient DNA, Strontium isotopes, and osteological analyses shed light on
social and kinship organization of the Later Stone Age, PNAS, 105 (47), 18226–31.
Haak, W. et al. 2010. Ancient DNA from European early Neolithic farmers reveals their Near
Eastern affinities, PloS Biology, 8 (11), e1000536.
Haak, W. et al. 2015. Massive migration from the steppe is a source of Indo-European languages,
Nature, published online 2 March 2015.
Haas, J. et al. 1998. Synchronous Holocene climatic oscillations recorded on the Swiss Plateau
and at timberline in the Alps, The Holocene, 8 (3), 301–09.
Häkkinen, J. 2012. Early contacts between Uralic and Yukaghir, in Per Urales AD Orientem Iter
polyphonicum multilingue Festkrift tillägnad Juha Janhunen på hans sextioårsdag den 12
februari 2012 (Mémoires de la Société Finno-Ourgrienne 264), T. Hyytiäinen, L. Jalava, J.
Saarikivi and E. Sandman (eds), 91–101. Helsinki: Finno-Ugrian Society.
Halkon, P. 2011. Iron, landscape and power in Iron Age East Yorkshire, The Archaeological
Journal, 168, 133–65.
Hamp, E. P. and Adams, D. Q. 2013. The Expansion of the Indo-European Languages: An Indo-
Europeanist’s Evolving View. Sino-Platonic Papers 239.
Hanning, E., Gauß, R. and Goldenberg, G. 2010. Metal for Zambujal: experimentally
reconstructing a 5000-year-old technology; Metal para Zambujal: reconstrucción experimental
de una tecnología de 5.000 años, Trabajos de Prehistoria, 67 (2), 287–304.
Harding, D. W. 2007. The Archaeology of Celtic Art. Abingdon and New York: Routledge.
Harrison, R. and Heyd, V. 2007. The Transformation of Europe in the Third Millennium BC: the
example of ‘Le Petit-Chasseur I + III’ (Sion, Valais, Switzerland), Praehistorische Zeitschrift, 82
(2), 129–214.
Hartz, S., Terberger, T. and Zhilin, M. 2010. New AMS-dates for the Upper Volga Mesolithic and
the origin of microblade technology in Europe, Quartär 57, 155–69.
Heath, J. 2012. Life in Copper Age Britain. Stroud: Amberley Press.
Henderson, J. 2007. The Atlantic Iron Age: Settlement and Identity in the First Millennium BC.
Abingdon and New York: Routledge.
Herodian of Antioch’s History of the Roman Empire from the Death of Marcus Aurelius to the
Accession of Gordian III, trans. E. C. Echols. 1961. Berkeley and Los Angeles: University of
California Press.
Herodotus. The Histories, trans. R. Waterfield. 1998. Oxford and New York: Oxford University
Press.
Hey, D. 2000. Family Names and Family History. London and New York: Hambledon and London.
Heyd, V. 2007. When the West meets the East: The eastern periphery of the Bell Beaker
phenomenon and its relation with the Aegean Early Bronze Age, in Between the Aegean and
the Baltic Seas: Prehistory Across Borders: Proceedings of the International Conference,
Bronze and Early Iron Age Interconnections and Contemporary Developments Between the
Aegean and the Regions of the Balkan Peninsula, Central and Northern Europe University of
Zagreb, 11–14 April 2005 (Aegaeum: Annales d’archéologie égéenne de l’Université de Liège
et UT-PASP 27), I. Galanaki, H. Tomas, Y. Galanakis and R. Laffineur (eds), 91–104. Université
de Liege and University of Texas at Austin.
Heyd, V. 2011. Yamnaya groups and tumuli west of the Black Sea, in Ancestral Landscapes:
Burial Mounds in the Copper and Bronze Ages, E. Borgna and S. Müller Celka (eds), 536–55.
Travaux de la Maison de l’Orient et de la Méditerranée 61. Lyon.
Heyd, V., Husty, L. and Kreiner, L. 2004. Siedlungen der Glockenbecherkultur in Süddeutschland
und Mitteleuropa. Arbeiten zur Archäologie Süddeutschlands. Bd 17. Büchenbach: Verlag Dr.
Faustus.
Hickey, R. 2002. Internal and external forces again: changes in word order in Old English and Old
Irish, Language Sciences, 24 (3–4), 261–83.
Historia Brittonum, translated in Six Old English Chronicles, ed. J. A. Giles. 1848. London: Henry
G. Bohn.
Historia Regum Britanniae of Geoffrey of Monmouth, I: Bern, Burgerbibliothek, MS 568, ed. N.
Historia Regum Britanniae of Geoffrey of Monmouth, I: Bern, Burgerbibliothek, MS 568, ed. N.
Wright. 1985. Cambridge: D. S. Brewer.
Homer. The Iliad, trans. A. T. Murray (Loeb Classical Library). Cambridge, MA: Harvard University
Press; London: William Heinemann Ltd. 1924; or trans. E. V. Rieu, revised P. Jones. 2003.
London: Penguin Books.
Höppner, B. et al. 2005. Prehistoric copper production in the Inn Valley (Austria), and the earliest
copper in Central Europe, Archaeometry, 47 (2), 293–315.
Hubert, H. 1934. The Rise of the Celts. London: Kegan Paul & Co.
Huehnergard, J. 2011. Proto-Semitic language and culture, in The American Heritage Dictionary
of the English Language (5th ed.), 2066–69. Boston and New York: Houghton, Mifflin and
Harcourt.
Hummler, M. 2007. Bridging the gap at La Tène, Antiquity, 81, 1067–70.
Hurl, D., McSparren, C. and Moore, P. 2002. Excavations at the Dorsey, Co. Armargh. Centre for
Archaeological Fieldwork, Queen’s University, Belfast. Data Structure Report no. 4.
Hyllested, A. 2010. The precursors of Celtic and Germanic, in Proceedings of the 21st UCLA
Indo-European Conference, S. W. Jamison, H. C. Melchert and B. Vine (eds), 107–28. Bremen:
Hempen.
Inizan, M. L. 2012. Pressure débitage in the Old world: forerunners, researchers, geopolitics,
handing on the baton, chapter 2 in Desrosiers (ed.) 2012, 11–42.
Isaac, G. R. 2010. The origins of the Celtic languages: language spread from east to west,
chapter 7 in Cunliffe and Koch 2010, 162–65.
Isidore. The Etymologies of Isidore of Seville, ed. and trans. S. A. Barney, W. J. Lewis, J. A.
Beach and O. Bergho. 2006. Cambridge: Cambridge University Press.
Jackson, K. H. 1964. The Oldest Irish Tradition: A Window on the Iron Age. Cambridge and New
York: Cambridge University Press.
James, S. 1999. The Atlantic Celts: Ancient People or Modern Invention? London: British
Museum Press.
Jay, M. et al. 2012. Chariots and Context: new radiocarbon dates from Wetwang and the
chronology of Iron Age burials and brooches in East Yorkshire, Oxford Journal of Archaeology,
31 (2), 161–89.
Jay, M. et al. 2013. British Iron Age chariot burials of the Arras culture: a multi-isotope approach
to investigating mobility levels and subsistence practices, World Archaeology, 45 (3), 473–91.
Jay, M., Parker Pearson, M. et al. 2012. The Beaker People Project: an interim report on the
progress of the isotope analysis of the organic skeletal material, in Allen, Gardiner and Sheridan
(eds) 2012, 226–36.
Jenner, H. 1904. A Handbook of the Cornish Language, chiefly in its latest stages with some
account of its history and literature. London: David Nutt.
Jeunesse, C. 2014. Pratiques funéraires campaniformes en Europe – Faut-il remettre en cause la
dichotomie Nord-Sud? La question de la réutilisation des sépultures monumentales dans
l’Europe du 3e millénaire, in Données récentes sur les pratiques funéraires néolithiques de la
Plaine du Rhin supérieur, P. Lefranc, A. Denaire and C. Jeunesse (eds), BAR International
Series 2633, 211. Oxford: Archaeopress.
John, B. S. 1972. The linguistic significance of the Pembrokeshire landsker, The Pembrokeshire
Historian: Journal of the Pembrokeshire Local History Society, 4, 7–29.
Johnston, P. 1997. Older Scots phonology and its regional variation, chapter 3 in Jones (ed.)
1997, 47–111.
Jones, C. (ed.) 1997. The Edinburgh History of the Scots Language. Edinburgh: Edinburgh
University Press.
Jones, H. M. 2012. A Statistical Overview of the Welsh Language. Cardiff: Welsh Language
Board.
Jones, Works = The Works of Sir William Jones, ed. A. M. Jones, 13 vols (2nd ed.). 1807.
London: J. Stockdale.
Josephson, F. 2012. Transfer of morphemes and grammatical structure in ancient Anatolia, in
Copies Versus Cognates in Bound Morphology, L. Johanson and M. Robbeets (eds), 337–54.
Leiden: Brill.
Jovanović, B. 1991. The Scordisci, in Moscati et al. (eds) 1991, 337–47.
Justin. Marcus Junianus Justinus. Epitome of the Philippic History of Pompeius Trogus, trans.
with notes by the Rev. J. S. Watson. 1853. London: Henry G. Bohn.
Karl, R. 2001. … on a road to nowhere ..: Chariotry and the road systems in the Celtic world,
IRQUAS Online Project. New Perspectives in Irish Studies.
Karl, R. 2003. Iron Age chariots and medieval texts: a step too far in ‘breaking down boundaries’?,
e-Keltoi: the Journal of Interdisciplinary Celtic Studies, 5: Warfare, 1–30.
Karl, R. 2006. Boii and the Celts in Bohemia, in Koch (ed.) 2006, 222–26.
Kelly, J. N. D. 1975. Jerome: His Life, Writings and Controversies. London: Duckworth; New York:
Harper & Row.
Keyser, C. et al. 2009. Ancient DNA provides new insights into the history of south Siberian
Kurgan people, Human Genetics, 126 (3), 395–410.
Kirtcho, L. 2009. The earliest wheeled transport in Southwestern Central Asia: new finds from
Altyn-Depe, Archaeology, Ethnology and Anthropology of Eurasia, 37 (1), 25–33.
Kitson, P. R. 1996. Reconstruction, typology and the ‘original homeland’ of the Indo-Europeans, in
Linguistic Reconstruction and Typology, J. Fisiak (ed.). (Trends in Linguistics: Studies and
Monographs 96), 183–239. Berlin: Mouton de Gruyter.
Knipper, C. et al. 2014. Social differentiation and land use at an Early Iron Age ‘princely seat’:
bioarchaeological investigations at the Glauberg (Germany), Journal of Archaeological Science,
41, 818–35.
Koch, J. T. 1992. Gallo-Brittonic vs. Insular Celtic: The inter-relationships of the Celtic languages
reconsidered, in Bretagne et pays celtiques – langues, histoire, civilisation: Mélanges offerts à
la mémoire de Léon Fleuriot, Gw. Le Menn, J.-Y. Le Moing (eds), 471–95. Saint-Brieuc and
Rennes.
Koch, J. T. 1995. The Conversion and the transition from Primitive to Old Irish c. 367–c. 637,
Emania 13, 39–50.
Koch, J. T. (ed.) 2006. Celtic Culture: A Historical Encyclopedia, 5 vols. Santa Barbara, CA: ABC-
CLIO.
Koch, J. T. 2007. An Atlas for Celtic Studies: Archaeology and Names in Ancient Europe and
Early Medieval Ireland, Britain and Brittany (Celtic Studies Publications 12). Oxford: Oxbow
Books and Celtic Studies Publications.
Koch, J. T. 2010. Paradigm shift? Interpreting Tartessian as Celtic, chapter 9 in Cunliffe and Koch
(eds) 2010, 185–301.
Koch, J. T. 2011. Tartessian 2: The Inscription of Mesas do Castelinho, ro and the Verbal
Complex, Preliminaries to Historical Phonology. Aberystwyth: University of Wales Centre for
Advanced Welsh and Celtic Studies.
Koch, J. T. and Carey, J. (eds). 2003. The Celtic Heroic Age: Literary Sources for Ancient Celtic
Europe & Early Ireland & Wales (4th ed.). Aberystwyth: Celtic Studies Publications.
Koch, J. T. and Cunliffe, B. (eds). 2013. Celtic from the West 2: Rethinking the Bronze Age and
the Arrival of Indo-European in Atlantic Europe. Oxford and Oakville: Oxbow Books.
Kohl, P. L. 2007. The Making of Bronze Age Eurasia (Cambridge World Archaeology).
Cambridge, New York and Melbourne: Cambridge University Press.
 Cambridge, New York and Melbourne: Cambridge University Press.
Kortlandt, F. 1981. More evidence for Italo-Celtic, Ériu, 32, 1–22. Reprinted in Italo-Celtic Origins
and Prehistoric Development of the Irish Language, 25–50. 2007. Amsterdam and New York:
Editions Rodopi B.V.
Korvin-Piotrovskiy, A. G. 2012. Tripolye Culture in Ukraine, in The Tripolye Culture: Giant-
settlements in Ukraine: Formation, Development and Decline, F. Menotti and A. G. Korvin-
Piotrovskiy (eds), 6–18. Oxford and Oakville: Oxbow Books.
Koukouli-Chrysanthaki, Ch. and Papadopoulos, S. 2009. The Island of Thassos and the Aegean
in the prehistory, Asmosia VII: Proceedings of the 7th International Conference of Association
for the Study of Marble and Other Stones in Antiquity, Thassos, September 15–20, 2003,
Bulletin de Correspondance Hellénique Supplément 51.
Krahe, H. 1963. Die Struktur der alteuropäischen Hydronomie. Wiesbaden: Akademie der
Wissenschaften und der Literatur.
Kremenetski, C. V. 2003. Steppe and forest-steppe belt of Eurasia: Holocene environmental
history, chapter 2 in Prehistoric Steppe Adaptation and the Horse (McDonald Institute
Monographs), M. Levine, C. Renfrew and K. Boyle (eds), 11–27. Cambridge: University of
Cambridge.
Kremenetski, C. V., Chichagova, O. A. and Shishlina, N. I. 1999. Palaeoecological evidence for
Holocene vegetation, climate and land-use change in the low Don basin and Kalmuk area,
southern Russia, Vegetation History and Archaeobotany, 8 (4), 233–46.
Kristiansen, K. 1998. Europe Before History (New Studies in Archaeology). Cambridge and New
York: Cambridge University Press.
Kristiansen, K. 2005. What language did Neolithic pots speak? Colin Renfrew’s European
farming-language-dispersal model challenged, Antiquity, 79 (305), 694–95.
Kroonen, G. 2012. Non-Indo-European root nouns in Germanic: Evidence in support of the
Agricultural Substrate Hypothesis, in A Linguistic Map of Prehistoric Northern Europe, R.
Grünthal and P. Kallio (eds), 239–60.
Mémoires de la Société Finno-Ougrienne 266. Helsinki: Société Finno-Ougrienne.
Kulcsár, G. and Szeverényi, V. 2013. Transition to the Bronze Age: issues of continuity and
discontinuity in the first half of the third millennium BC in the Carpathian Basin, in Transitions to
the Bronze Age. Interregional Interaction and Socio-Cultural Change in the Third Millennium BC
Carpathian Basin and Neighbouring Regions, V. Heyd, G. Kulcsár and V. Szeverényi (eds), 67–
92. Budapest: Archaeolingua.
Kunst, M. 2001. Invasions? Fashion? Social Ranks? Consideration concerning the Bell Beaker
phenomenon in Copper Age fortifications of the Iberian peninsula, in Nicolis (ed.) 2001, 81–90.
Kunst, M. 2007. Zambujal (Torres Vedras, Lisboa): relatório das escavações de 2001, Revista
Portuguesa de Arqueologia, 10 (1), 95–118.
Kuzmina, E. E. 2007. The Origin of the Indo-Iranians, ed. J. P. Mallory. Leiden and Boston: Brill.
Labaune, M. 2013. Bell Beaker metal and metallurgy in Western Europe, chapter 16 in Prieto
Martinéz and Salanova (eds) 2013, 177–88.
Lacan, M. 2011. Ancient DNA suggests the leading role played by men in the Neolithic
dissemination, PNAS, 108 (45), 18255–59.
Lacey, B. 2006. Cenél Conaill and the Donegal Kingdoms, AD 500–800. Dublin: Four Courts
Press.
Lane, A. 2014. Wroxeter and the end of Roman Britain, Antiquity, 88, 501–15.
Lapidge, M., Blair, J., Keynes, S. and Scragg, D. (eds). 1999. Blackwell Encyclopaedia of Anglo-
Saxon England. Oxford: Blackwell.
Laws of the Earliest English Kings, ed. and trans. F. L. Attenborough. 1922. Cambridge:
Cambridge University Press.
 Cambridge University Press.
Lazaridis, I. et al. 2014. Ancient human genomes suggest three ancestral populations for present-
day Europeans, Nature, 513, 409–13.
Lazarovici, C.-M. 2010. New data regarding the chronology of the Precucuteni, Cucuteni and
Horodisștea-Erbiceni cultures, PANTA RHEI: Studies on the Chronology and Cultural
Development of South-Eastern and Central Europe in Earlier Prehistory Presented to Juraj
Pavúk on the Occasion of his 75th Birthday. Studia Archaeologica et Mediaevalia, 11, 71–94.
Bratislava: Comenius University.
Lebor Gabála Érenn, ‘The Book of the Taking of Ireland’, Irish Texts Society, vols 34 (1938), 35
(1939), 39 (1940), 41 (1941) and 44 (1956).
Lechterbeck, J. et al. 2014. How was Bell Beaker economy related to Corded Ware and Early
Bronze Age lifestyles? Archaeological, botanical and palynological evidence from the Hegau,
Western Lake Constance region, Environmental Archaeology, 19 (2), 95–113.
Lee, E. J. et al. 2012. Emerging genetic patterns of the European neolithic: Perspectives from a
late neolithic bell beaker burial site in Germany, American Journal of Physical Anthropology,
148 (4), 571–79.
Lejars, T. 2003. La nécropole celtique de Roissy, Pour La Science, 306.
Lemercier, O. 2012. Interpreting the Beaker phenomenon in Mediterranean France: an Iron Age
analogy, Antiquity, 86 (331), 131–43.
Leslie, S. et al. 2015. The fine-scale genetic structure of the British population, Nature, 519, 309–
14.
Lhuyd, E. 1707. Archaeologia Britannica: giving some account additional to what has been
hitherto publish’d, of the languages, histories and customs of the original inhabitants of Great
Britain: from collections and observations in travels through Wales, Cornwal, Bas-Bretagne,
Ireland and Scotland, vol. 1: Glossography. Oxford: Printed for the author.
Lintott, A. 1999. The Constitution of the Roman Republic. Oxford and New York: Oxford University
Press.
Lira, J. et al. 2010. Ancient DNA reveals traces of Iberian Neolithic and Bronze Age lineages in
modern Iberian horses, Molecular Ecology, 19 (1), 64–78.
Lives of Saints Declan and Mochuda, ed. and trans. P. Power. 1914. Irish Texts Society 16.
Livy. History of Rome, trans. Rev. Canon Roberts. 1905. London: J. M. Dent & Sons, Ltd. History
of Rome: Books Nine to Twenty-Six, trans D. Spillan and C. Edmonds. 1868. London: Henry G.
Bohn.
Lorrio, A. J. and Zapatero, G. R. 2005. The Celts in Iberia: an overview, e-Keltoi: the Journal of
Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 167–254.
Luján Martínez, E. R. 2006. The language(s) of the Callaeci, e-Keltoi: the Journal of
Interdisciplinary Celtic Studies, 6: The Celts in the Iberian Peninsula, 715–48.
Lynn, C. J. 1989. An interpretation of ‘The Dorsey’, Emania 6, 5–10.
Mac Carron, P. and Kenna, R. 2012. Universal properties of mythological networks, Epl
(Europhysics Letters), 99, 28002.
McEvoy. B. and Bradley, D. G. 2006. Y-chromosomes and the extent of patrilineal ancestry in
Irish surnames, Human Genetics, 119, 212–19.
McEvoy, B. P. and Bradley, D. G. 2010. Irish Genetics and Celts, chapter 5 in Cunliffe and Koch
2010, 107–20.
MacFarlane, J. 1922. The History of Clan MacFarlane. Glasgow: David J. Clark.
McKie, R. 2006. The Face of Britain: How Our Genes Reveal the History of Britain. London:
Simon & Schuster.
MacKillop, J. 2004. A Dictionary of Celtic Mythology. Oxford: Oxford University Press.
McManus, D. 1996. Ogham, section 26 in The World’s Writing Systems, P. T. Daniels and W.
McManus, D. 1996. Ogham, section 26 in The World’s Writing Systems, P. T. Daniels and W.
Bright (eds), 340–45. Oxford and New York: Oxford University Press.
MacNeill, E. 1920. Phases of Irish History. Dublin: M. H. Gill and Son.
Maggi, R. and Pearce, M. 2005. Mid fourth-millennium copper mining in Liguria, north-west Italy:
the earliest known copper mines in Western Europe, Antiquity, 79 (303), 66–77.
Maier, F. 1991. The oppidum of Manching, in Moscati et al. (eds), 1991, 530–31.
Makhortyk, S. V. 2008. On the question of Cimmerian imports and imitations in Central Europe, in
Import and Imitation in Archaeology, P. F. Biehl and Y. Ya. Rassamakin (eds), 167–86.
Langenweißbach: Beier & Beran.
Mallory, J. P. 1989. In Search of the Indo-Europeans: Language, Archaeology and Myth. London:
Thames & Hudson.
Mallory, J. P. 1992. The world of Cú Chulainn: the archaeology of the Táin Bó Cúailnge, in
Aspects of the Táin, J. P. Mallory (ed.), 103–59. Belfast: December Publications.
Mallory, J. P. 1993. The Archaeology of the Irish Dreamtime, in Proceedings of the Harvard Celtic
Colloquium, 13, B. Hillers, P. Hopkins and J. Hunter (eds), 1–24. Department of Celtic
Languages and Literature: Harvard University.
Mallory, J. P. 1998. The Old Irish Chariot, in Mír Curad: Studies in Honour of Calvert Watkins, J.
Jasanoff, H. C. Melchert and L. Olivier (eds), 451–64. Innsbruck: Sonderbruck.
Mallory, J. P. 2013. The Origins of the Irish. London and New York: Thames & Hudson.
Mallory, J. P. and Adams, D. Q. (eds). 1997. Encyclopedia of Indo-European Culture. London:
Fitzroy Dearborn Publishers.
Mallory, J. P. and Adams, D. Q. 2006. Oxford Introduction to Proto-Indo-European and the Proto-
Indo-European World. Oxford: Oxford University Press.
Mallory, J. P. and Mair, V. H. 2000. The Tarim Mummies: The Mystery of the First Europeans in
China. London and New York: Thames & Hudson.
Manco, J. 1998. Saxon Bath: The Legacy of Rome and the Saxon Rebirth, Bath History, 7, 27–54.
Manco, J. 2015. Ancestral Journeys: The Peopling of Europe from the First Venturers to the
Vikings. Revised and updated edition. London and New York: Thames & Hudson.
Manley J. and Rudkin, D. 2005. A pre-AD 43 ditch at Fishbourne Roman palace, Chichester,
Britannia, 36, 55–99.
Manning, W. H. 1995. Ironworking in the Celtic world, chapter 17 in Green (ed.) 1995, 310–20.
Manti, P. and Watkinson, D. 2008. From Homer to hoplite: scientific investigations of Greek
copper alloy helmets, History of Mechanism and Machine Science, 6, 167–79.
Manzura, I. 2005a. Steps to the steppe: or, how the North Pontic region was colonised, Oxford
Journal of Archaeology, 24 (4), 313–38.
Manzura, I. 2005b. The proto-Bronze Age cemetery at Durankulak: a look from the East, in
Prehistoric Archaeology and Theoretical Anthropology and Education (Reports of Prehistoric
Research Projects 6–7), L. Nikolova, J. Fritz and J. Higgins (eds), 51–55. Salt Lake City and
Karlovo.
Marciniak, A. 2011. The Secondary Products Revolution: empirical evidence and its current
zooarchaeological critique, Journal of World Prehistory, 24, 117–30.
Markoe, G. E. 2000. The Phoenicians (Peoples of the Past). London: British Museum Press.
Martirosyan, H. 2013. The place of Armenian in the Indo-European language family: the
relationship with Greek and Indo-Iranian, Journal of Language Relationship, 10, 85–137.
Maryon, H. 1936. Excavation of two Bronze age barrows at Kirkhaugh, Northumberland,
Archaeologia Aeliana (4th series), 13, 207–17.
Mata, D. R. 2002. The ancient Phoenicians of the 8th and 7th centuries BC in the Bay of Cadiz:
state of the research, in The Phoenicians in Spain: An Archaeological Review of the Eighth–
 Sixth Centuries B.C.E. A Collection of Articles Translated from Spanish, M. R. Bierling (ed. and
trans.), 155–98. Winona Lake, IN: Eisenbrauns.
Matasović, R. 2007. Insular Celtic as a language area, in The Celtic Languages in Contact:
Papers from the Workshop within the Framework of the XIII International Congress of Celtic
Studies, Bonn, 26–27 July 2007, H. L. C. Tristram (ed.), 93–112. Potsdam: Universitätsverlag
Potsdam. Matasović, R. 2012. The substratum in Insular Celtic, Journal of Language
Relationship, 8, 153–68.
Mazzieri, P. and Dal Santo, N. 2007. Il sito del Neolitico recente di Botteghino (Parma), Rivista di
Scienze Preistoriche, LVII, 113–38.
Mees, B. 2003. Stratum and shadow: a genealogy of stratigraphy theories from the Indo-
European West, in Language Contacts in Prehistory: Studies in Stratigraphy: Papers from the
Workshop on Linguistic Stratigraphy and Prehistory at the Fifteenth International Conference on
Historical Linguistics, Melbourne, 17 August 2001. (Amsterdam Studies in the Theory and
History of Linguistic Science, Series IV: Current Issues in Linguistic Theory, 239), H. Anderson
(ed.), 11–44. Amsterdam and Philadelphia: John Benjamins Publishing Company.
Megaw, R. and Megaw, V. 1995. The nature and function of Celtic art, chapter 20 in Green (ed.)
1995, 345–75.
Menk, R. 1979. Le phenomène campaniforme: structure biologiques et intégration historique,
Archives Suisses d’anthropologie générale, 43, 259–84.
Meyer, K. (ed. and trans.). 1911. Selections from Ancient Irish Poetry. London: Constable &
Company.
Militarev, A. Y. 2005. Once more about glottochronology and the comparative method: the
Omotic-Afrasian case, Aspects of Comparative Linguistics, 1, 339–40.
Mödlinger, M. et al. 2013. Archaeometallurgical characterization of the earliest European metal
helmets, Materials Characterization, 79, 22–36.
Moffat, A. and Wilson, J. 2011. The Scots: A Genetic Journey. Edinburgh: Birlinn.
Moore, C. and Chiriotti, C. 2010. Reinventing the wheel: new evidence from Edercloon, Co.
Longford, chapter 5 in Creative Minds: Production, Manufacturing and Invention in Ancient
Ireland, Proceedings of a Public Seminar on Archaeological Discoveries on National Road
Schemes, August 2009, M. Stanley, E. Danaher and J. Eogan (eds). (Archaeology and the
National Roads Authority, Monograph Series No. 7.) Dublin: National Roads Authority.
Moore, L. T. et al. 2006. A Y-chromosome signature of hegemony in Gaelic Ireland, American
Journal of Human Genetics, 78 (2), 334–38.
Morgunova, N. L. and Khokhlova, O. S. 2013. Chronology and periodization of the Pit-Grave
Culture in the region between the Volga and Ural Rivers based on radiocarbon dating and
paleopedological research, in Proceedings of the 21st International Radiocarbon Conference, A.
J. T. Jull and C. Hatté (eds), Radiocarbon, 55 (2–3), 1286–96.
Morley, C. 2008. Chariots and migrants in East Yorkshire: dismantling the argument, in
Movement, Mobility and Migration, Archaeological Review from Cambridge, 23.2, E. Lightfoot
(ed.), 69–91.
Moscati, S. et al. (eds) 1991. The Celts. Milan: Bompiani; London: Thames & Hudson.
Moseley, C. (ed.). 2010. Atlas of the World’s Languages in Danger (3rd ed.). Paris: UNESCO
Publishing. Online version.
Muhr, K. 1996. Place-Names of Northern Ireland, 6: Northwest County Down / Iveagh. Belfast:
Cló Ollscoil na Banríona.
Muller, J.-C. 1986. Early stages of language comparison from Sassetti to Sir William Jones
(1786), Kratylos, 31, 1–31.
Müller, J. and van Willigen, S. 2001. New radiocarbon evidence for European Bell Beakers and
the consequences for the diffusion of the Bell Beaker Phenomenon, in Nicolis (ed.) 2001, 59–
75.
Müller, R. et al. 2007. Zambujal and the beginnings of metallurgy in southern Portugal, in Metals
and Mines: Studies in Archaeometallurgy: Selected papers from the Conference Metallurgy: A
Touchstone for Cross-cultural Interaction held at the British Museum 28–30 April 2005 to
celebrate the career of Paul Craddock during his 40 years at the British Museum, S. La Niece,
D. Hook and P. Craddock (eds), 15–26. London: Archetype Publications in association with the
British Museum.
Murray, A. C. 2003. (ed. and trans.) From Roman to Merovingian Gaul: A Reader. Toronto,
Ontario: Higher Education University of Toronto Press.
Myres, N. et al. 2011. A major Y-chromosome haplogroup R1b Holocene era founder effect in
Central and Western Europe, European Journal of Human Genetics, 19 (1), 95–101.
Nakhleh, L., Ringe D. and Warnow, T. 2005. Perfect phylogenetic networks: a new methodology
for reconstructing the evolutionary history of natural languages, Language, Journal of the
Linguistic Society of America, 81 (2), 382–420.
Nichols, J. 1997. The epicenter of the Indo-European linguistic spread, in Archaeology and
Language I: Theoretical and Methodological Orientations, R. Blench and M. Spriggs (eds), 122–
48. London: Routledge.
Nicolaisen, W. F. H. 1982. Old European names in Britain, Nomina 6, 37–42.
Nicolis, F. (ed.) 2001. Bell Beakers Today: Pottery, People, Culture, Symbols in Prehistoric
Europe. Proceedings of the International Colloquium, Riva del Garda (Trento, Italy), 11–16 May
1998. 2 vols. Trento: Provincia Autonoma di Trento.
Nikitin, A. et al. 2012. Mitochondrial haplogroup C in ancient mitochondrial DNA from Ukraine
extends the presence of East Eurasian genetic lineages in Neolithic Central and Eastern
Europe, Journal of Human Genetics, 57, 610–12.
Nocete, F. 2006. The first specialised copper industry in the Iberian peninsula: Cabezo Juré,
Antiquity, 80 (309), 646–57.
Nocete, F. et al. 2011. Direct chronometry (14C AMS) of the earliest copper metallurgy in the
Guadalquivir Basin (Spain) during the third millennium BC: first regional database, Journal of
Archaeological Science, 38 (12), 3278–95.
Nocete, F. et al. 2014. Gold in the southwest of the Iberian peninsula during the 3rd Millennium
BC, Journal of Archaeological Science, 41, 691–704.
Northover, J. P. N., O’Brien, W. and Stos, S. 2001. Lead isotopes and metal circulation in
Beaker/Early Bronze Age Ireland, Journal of Irish Archaeology, 10, 25–47.
Novembre, J. et al. 2008. Genes mirror geography within Europe, Nature, 456, 98–101.
NPNF2:6 = The Nicene and Post-Nicene Fathers, second series, vol. 6: Jerome: Letters and
Select Works, trans. W. H. Fremantle. 1893.
Nutt, A. T. 1900. Cuchulainn, the Irish Achilles. London: D. Nutt.
O’Brien, M. A. 1962. Corpus Genealogiarum Hiberniae, vol. 1. Dublin: The Dublin Institute for
Advanced Studies.
O’Brien, S. R. et al. 1995. Complexity of Holocene climate reconstructed from a Greenland ice
core, Science, 270 (5244), 1962–64.
O’Brien, W. 2004. Ross Island: Mining, Metal and Society in Early Ireland (Bronze Age Studies 6).
Galway: National University of Ireland, Department of Archaeology.
Ó Corráin, D. 2006. Annals, Irish, in Koch (ed.) 2006, 69–75.
Ó Croidheáin, C. 2006. Language from Below: The Irish Language, Ideology and Power in 20th-
Century Ireland. Oxford, Bern, Berlin, Bruxelles, Frankfurt am Main, New York, Wien: Peter
Lang.
 Lang.
Ó Cróinín, D. 1995. Early Medieval Ireland 400–1200 (Longman History of Ireland). Abingdon and
New York: Routledge.
Ó Cróinín, D. 2005. Ireland, 400–800, chapter 7 in Ó Cróinín (ed.) 2005, 182–234.
Ó Cróinín, D. (ed.) 2005. A New History of Ireland, vol. 1: Prehistoric and early Ireland. Oxford
and New York: Oxford University Press.
ODNB = Oxford Dictionary of National Biography. Oxford University Press 2004–14.
Odriozola C. P. and Hurtado Pérez, V. M. 2007. The manufacturing process of 3rd millennium BC
bone based incrusted pottery decoration from the Middle Guadiana river basin (Badajoz, Spain),
Journal of Archaeological Science, 34, 1794–803.
O’Dushlaine, C. T. et al. 2010. Population structure and genome-wide patterns of variation in
Ireland and Britain, European Journal of Human Genetics, 18, 1248–54.
Ó Floinn, R. 2000. Freestone Hill, Co. Kilkenny: a reassessment, in Seanchas. Studies in Early
and Medieval Irish Archaeology, History and Literature in Honour of Francis J. Byrne, A. P.
Smyth (ed.), 12–29. Dublin: Four Courts Press.
Ó hUiginn, R. 2006. Táin Bó Cuailnge (‘The Cattle Raid of Cooley’), in Koch (ed.) 2006, 1646–47.
Olalde, I. et al. 2014. Derived immune and ancestral pigmentation alleles in a 7,000-year-old
Mesolithic European, Nature, 507, 225–28.
O’Laverty, J. 1878. An Historical Account of the Diocese of Down and Conor, Ancient and
Modern, vol. 1. Dublin: James Duffy and Sons.
Olivier, L. 1999. The Hochdorf ‘princely’ grave and the question of the nature of archaeological
funerary assemblages, chapter 8 in Time and Archaeology, T. Murray (ed.), 109–38. London
and New York: Routledge.
Orme, N. 2000. The Saints of Cornwall. Oxford and New York: Oxford University Press.
Østmo, E. 2012. Late Neolithic expansion to Norway. The beginning of a 4000 year-old
shipbuilding tradition, chapter 6 in Prescott and Glørstad 2012, 63–69.
Outram, A.K. et al. 2009. The earliest horse harnessing and milking, Science, 323 (5919), 1332–
35.
Panegyrici Latini. In Praise of Later Roman Emperors: the Panegyrici Latini, ed. and trans. C. E.
V. Nixon and B. Saylor Rodgers. 1995. Oakland, CA: University of California Press.
Pardee, D. 2009. A new Aramaic inscription from Zincirli, Bulletin of the American Schools of
Oriental Research, 356, 51–71.
Pare, C. 1991. Fürstensitze, Celts and the Mediterranean World: Developments in the West
Hallstatt Culture in the 6th and 5th centuries BC, Proceedings of the Prehistoric Society, 57 (2),
183–202.
Pare, C. F. E. 2000. Bronze and the Bronze Age, in Metals Make the World Go Round: The
Supply and Circulation of Metals in Bronze Age Europe. Proceedings of a Conference Held at
the University of Birmingham in June 1997, C. F. E. Pare (ed.), 1–32. Oxford: Oxbow Books.
Parker Pearson, M. 1995. Southwestern Bronze Age pottery, in ‘Unbaked Urns of Rudely Shape’:
Essays on British and Irish Pottery for Ian Longworth, I. Kinnes and G. Varndell, (eds), 89–100.
Oxford: Oxbow.
Parkinson, W. A. et al. 2010. Elemental analysis of ceramic incrustation indicates long-term
cultural continuity in the prehistoric Carpathian Basin, Archaeology, Ethnology and
Anthropology of Eurasia, 38 (2), 64–70.
Parpola, A. 2008. Proto-Indo-European speakers of the Late Tripolye culture as the inventors of
wheeled vehicles: Linguistic and archaeological considerations, in Proceedings of the 19th
Annual UCLA Indo-European Conference (The Journal of Indo-European Studies Monograph
Series 54), K. Jones-Bley, M. E. Huld, A. Della Volpe and M. Robbins Dexter (eds), 1–59.
Washington, DC: Institute for the Study of Man.
 Washington, DC: Institute for the Study of Man.
Parsons, J. 1767. The Remains of Japhet, being historical enquiries into the affinity and origins of
the European languages. London: Printed for the author.
Patrick. Confessio, trans. P. McCarthy. 2003. Dublin: Royal Irish Academy.
Pausanias. Description of Greece, trans. W. H. S. Jones and H. A. Ormerod. 1918. 5 vols (Loeb
Classical Library). Cambridge, MA, Harvard University Press; London, William Heinemann Ltd.
Pereltsvaig, A. and Lewis, M. 2015. The Indo-European Controversy: Facts and Fallacies in
Historical Linguistics. Cambridge: Cambridge University Press.
Peška, J. 2013. Two new burial sites of Bell Beaker Culture with exceptional finds from Eastern
Moravia/Czech Republic, chapter 6 in Prieto-Martinez and Salanova (eds) 2013, 61–72.
Pezron, [P. Y.]. 1706. The Antiquities of Nations, More particularly of the Celtae or Gauls, Taken
to be Originally the same People as our Ancient Britains, Englished by Mr. Jones. London:
Janeway.
Piggott, S. 1995. Wood and the wheelwright, chapter 18 in Green (ed.) 1995, 321–27.
Piguet M. and Besse, M. 2009. Chronology and Bell Beaker common ware, Radiocarbon, 51 (2),
817–30.
Pliny the Elder. Natural History, trans. J. Bostock and H. T. Riley. 6 vols. 1855–57. London: H. G.
Bohn.
Plutarch. The Life of Caesar, in The Fall of the Roman Republic: Six Lives by Plutarch, trans. R.
Warner. 1958. Harmondsworth: Penguin.
Polyaenus. Stratagems of War, trans. R. Shepherd. 1793. London: George Nicol.
Polybius. The Histories, trans. R. Waterfield with notes by B. McGing. 2010. Oxford: Oxford
University Press.
Pomponius Mela’s Description of the World, ed. F. E. Romer. 1998. Ann Arbor: University of
Michigan Press.
Prescott, C. 2012. Third millennium transformations in Norway: modelling an interpretive platform,
chapter 10 in Prescott and Glørstad (eds) 2012, 115–27.
Prescott, C. and Glørstad, H. (eds). 2012. Becoming European. Oxford and Oakville: Oxbow
Books.
Price, T. D., Knipper, C., Grupe, G. and Smrcka, V. 2004. Strontium isotopes and prehistoric
human migration: the Bell Beaker period in Central Europe, European Journal of Archaeology, 7
(1), 9–40.
Prieto-Martínez, M. P. 2012. Perceiving changes in the third millennium BC in Europe through
pottery, chapter 4 in Prescott and Glørstad (eds) 2012, 30–47.
Prieto-Martínez, M. P. and Salanova, L. (eds) 2013. Current Researches on Bell Beakers:
Proceedings of the 15th International Bell Beaker Conference: From Atlantic to Ural. 5th–9th
May 2011. Poio (Pontevedra, Galicia, Spain). Santiago de Compostela.
Prósper, B. M. 2010/2011. The Hispano-Celtic divinity ILVRBEDA, gold mining in western
Hispania and the syntactic context of Celtiberian arkatobezom ‘silver mine’, Die Sprache, 49 (1),
53–83.
Prósper, B. M. 2014. Sifting the evidence: new interpretations on Celtic and Nonceltic personal
names of western Hispania in the light of phonetics, composition and suffixation, in Continental
Celtic Word Formation. The Onomastic Data, J. L. G. Alonso (ed.), 181–200. Salamanca:
Ediciones Universidad de Salamanca.
Ptolemy, Claudius. The Geography. No reliable complete English translation is in print. That used
is trans. E. L. Stevenson. 1932. New York: New York Public Library.
Querré, G. and Convertini, F. 1998. Apports des études céramologiques en laboratoire à la
connaissance du Campaniforme: résultats, bilan et perspectives, Bulletin de la Société
préhistorique française, 95 (3), 333–42.
 préhistorique française, 95 (3), 333–42.
Raftery, B. 2005. Iron-Age Ireland, chapter 6 in Ó Cróinín (ed.) 2005, 134–81.
Raghavan, M. et al. 2014. Upper Palaeolithic Siberian genome reveals dual ancestry of Native
Americans, Nature, 505, 87–91.
Rance, P. 2001. Attacotti, Déisi and Magnus Maximus: the case for Irish federates in Late Roman
Britain, Britannia, 32, 243–70.
Rankin, D. 1987. Celts and the Classical World. London and New York: Routledge.
Reaney, P. H. and Wilson, R. M. 1997. Oxford Dictionary of English Surnames (3rd ed.). Oxford:
Oxford University Press.
Redmonds, G., King T. and Hey, D. 2011. Surnames, DNA and Family History. Oxford and New
York: Oxford University Press.
Rees, B. R. 1998. Pelagius: Life and Letters. Woodbridge: Boydell Press.
Renfrew, C. 1987. Archaeology and Language: The Puzzle of Indo-European Origins. London:
Jonathan Cape.
ResGestae Divi Augusti, in Compendium of Roman History. Res Gestae Divi Augusti, trans. F. W.
Shipley. 1924. (Loeb Classical Library). Cambridge, MA: Harvard University Press.
Reynolds, S. 1983. Medieval origines gentium and the community of the realm, History, 68, 375–
90.
Ringe, D., Warnow, T. and Taylor, A. 2002. Indo-European and computational cladistics,
Transactions of the Philological Society, 100 (1), 59–129.
Ríos, P. 2013. New dating of the Bell Beaker Horizon in the region of Madrid, chapter 9 in Prieto-
Martínez and Salanova (eds) 2013, 97–109.
Ritchie, J. N. G. and W. F. 1995. The army, weapons and fighting, chapter 4 in Green (ed.) 1995.
Rivet, A. L. F. and Smith, C. 1979. The Place-Names of Roman Britain. London: Batsford.
Robb, J. 2009. People of stone: stelae, personhood and society in prehistoric Europe, Journal of
Archaeological Method and Theory, 16 (3), 162–83.
Roberts, B. 2008. Creating traditions and shaping technologies: understanding the earliest metal
objects and metal production in Western Europe, World Archaeology, 40 (3), 354–72.
Roberts, B.W. 2009. Production networks and consumer choice in the earliest metal of Western
Europe, Journal of World Prehistory, 22 (4), 461–81.
Roberts, S., Sofaer, J. and Kiss, V. 2008. Characterization and textural analysis of Middle Bronze
Age Transdanubian inlaid wares of the Encrusted Pottery Culture, Hungary: a preliminary study,
Journal of Archaeological Science, 35 (2), 322–30.
Rocca, R. et al. 2012. Discovery of Western European R1b1a2 Y chromosome variants in 1000
Genomes project data: an online community approach, PLoS ONE 7 (7), e41634.
Rojo-Guerra, M. Á. et al. 2006. Beer and Bell Beakers: drinking rituals in Copper Age inner Iberia,
Proceedings of the Prehistoric Society, 72, 243–65.
Roller, D. W. 2006. Through the Pillars of Herakles: Greco-Roman Exploration of the Atlantic.
New York and London: Routledge.
Rolley, C. (ed.) 2003. The Tombe Princière de Vix. Paris: Société des Amis du Musée du
Châtillonais.
Ross, A. P. 1981. The Table of Nations in Genesis 10 – Its content, Bibliotheca Sacra, vol. 138.
Ruzickova, P. 2009. Bow-shaped pendants of the Bell Beaker Culture, Acta Archaeologica
Carpathica, 44, 37–72.
Salanova, L. 2004. The frontiers inside the western Bell Beaker block, in Czebreszuk (ed.) 2004,
63–75.
Salanova, L. 2008. Le temps d’une diffusion: la céramique campaniforme en Europe, in
Construire le temps. Histoire et méthodes des chronologies et calendriers des derniers
millénaires avant notre ère en Europe occidentale. Actes du XXXe colloque international de
 millénaires avant notre ère en Europe occidentale. Actes du XXXe colloque international de
Halma-Ipel, UMR 8164 (CNRS, Lille 3, MCC), 7–9 décembre 2006, Lille, A. Lehoërff (ed.), 135–
47. Glux-en-Glenne: Bibracte.
Salanova, L. et al. 2011. Archaeometric analysis and Bell Beaker pottery circulation: a synopsis of
the results in Western Europe. Paper presented at the 15th Bell Beaker International Congress
– From Atlantic to Ural, 5–9 May 2011.
Sarauw, T. 2007. Male symbols or warrior identities? The archery burials of the Danish Bell
Beaker Culture, Journal of Anthropological Archaeology, 26 (1), 65–87.
Sarauw, T. 2008. Danish Bell Beaker pottery and flint daggers – the display of social identities?,
European Journal of Archaeology, 11 (1), 23–47.
Schiffels, S. et al. 2014. Insights into British and European population history from ancient DNA
sequencing of Iron Age and Anglo-Saxon samples from Hinxton, England. Paper presented at
the American Society of Human Genetics Annual Meeting, October 8–22, 2014.
Schloen, D. and Fink, A. S. 2009. New excavations at Zincirli Höyük in Turkey (Ancient Sam’al)
and the discovery of an inscribed mortuary stele, Bulletin of the American Schools of Oriental
Research, no. 356, 1–13.
Schrijver, P. C. H. 2006. Review of G. Meiser, Veni Vidi Vici: Die Vorgeschichte des lateinischen
Perfektsystems, in Kratylos, 51, 46–64.
Schrijver, P. 2007. What Britons spoke around 400 AD, in Britons in Anglo-Saxon England, N. J.
Higham (ed.), 165–71. Woodbridge: Boydell Press.
Scozzari, R. et al. 2012. Molecular dissection of the basal clades in the human Y chromosome
phylogenetic tree, PLoS ONE 7 (11): e49170.
Seguin-Orlando, A. et al. 2014, Genomic structure in Europeans dating back at least 36,200
years, Science, 6 November 2014.
Senna-Martinez, J. C. et al. 2011. First Bronzes of North-West Iberia: The Date from Fraga dos
Corvos Habitat Site, in Povoamento e Exploração de Recursos Mineiros na Europa Atlântica
Ocidental/Settlement and Mining in the Atlantic Western Europe. Proceedings of the First
International Congress, Braga, 10 December of 2010, C. B. Martins et al. (eds), 381–94. Braga:
CITEM, APEQ, FEUP.
Shepherd, I. 2012. Is there a Scottish Chalcolithic, chapter 11 in Allen, Gardiner and Sheridan
2012, 164–71.
Sheridan, A. 2008. Upper Largie and Dutch-Scottish connections during the Beaker period, in
Between Foraging and Farming: An Extended Broad Spectrum of Papers Presented to
Leendert Louwe Kooijmans (Analecta Praehistorica Leidensia, 40), H. Fokkens et al. (eds),
247–60. Leiden: University of Leiden.
Sherratt, A. 1981. Plough and pastoralism: aspects of the secondary products revolution, chapter
10 in Pattern of the Past: Studies in Honour of David Clarke. I. Hodder, G. Isaac and N.
Hammond (eds), 261–305. Cambridge: Cambridge University Press.
Sherratt, A. 1986. Two new finds of wooden wheels from later Neolithic and Early Bronze Age
Europe, Oxford Journal of Archaeology, 5 (2), 243–48.
Sherratt, A. 1987. Cups that cheered, in Bell Beakers of the Western Mediterranean (BAR
International Series 287), W. H. Waldren and R. C. Kennard (eds), Oxford: British
Archaeological Reports, 81–114. Revised and reprinted as: Cups that cheered: the introduction
of alcohol to prehistoric Europe, in A. Sherratt, 1997. Economy and Society in Prehistoric
Europe: Changing Perspectives, 376–402. Princeton: Princeton University Press.
Shisha-Halevy, A. 2003. Celtic syntax, Egyptian-Coptic syntax, in Das Alte Ägypten und seine
Nachbarn: Festschrift Helmut Satzinger, M. R. M. Hasitzka, J. Diethart and G. Dembski (eds),
245–302. Krems: Österreichisches Literaturforum.
Shrimpton, G. S. 1991. Theopompus The Historian. McGill: Queens University Press.
Sikora, M. J., Colonna, V., Xue, Y. and Tyler-Smith, C. 2013. Modeling the contrasting Neolithic
male lineage expansions in Europe and Africa, Investigative Genetics, 4:25.
Silva, L. 2013. Viriathus: and the Lusitanian Resistance to Rome 155–139 BC. Barnsley: Pen and
Sword Military.
Simmons, I. and Tooley, M. (eds). 1981. The Environment in British Prehistory. London:
Duckworth.
Sims-Williams, P. 2006. Ancient Celtic Place-Names in Europe and Asia Minor (Publications of
the Philological Society, 39). Oxford and Boston: Blackwell Publishing.
Sims-Williams, P. 2007. Common Celtic, Gallo-Brittonic and Insular Celtic, in Gaulois et Celtique
Continental, P.-Y. Lambert and G.-J. Pinault (eds), 309–54. Geneva: Droz.
Skene, W. F. (ed. and trans.). 1868. The Four Ancient Books of Wales. 2 vols. Edinburgh:
Edmonston and Douglas.
Smith, W. (ed.). 1854. Dictionary of Greek and Roman Geography. London: Walton and
Maberly/John Murray.
Smyntyna, O. V. 2007. Late Mesolithic of the Ukrainian part of the Lower Danube region: New
perspectives of human adaptation and interpretation of natural environments, Quaternary
International 167–68, 114–20.
Solinus. Collectanea Rerum Memorabilium, ed. T. Mommsen. 1895. Berolini [Bern]: Weidmann.
Stevens, C. J. and Fuller, D. Q. 2012. Did Neolithic farming fail? The case for a Bronze Age
agricultural revolution in the British Isles, Antiquity, 86 (333), 707–22.
Strabo. The Geography, trans. H. L. Jones. 1917–32. 8 vols (Loeb Classical Library). Cambridge,
MA: Harvard University Press.
Struble, E. J. and Herrmann, V. R. 2009. An eternal feast at Sam’al: the new Iron Age mortuary
stele in context, Bulletin of the American Schools of Oriental Research, 356, 15–49.
Stupak, D. 2006. Chipped flint technologies in Swiderian complexes of the Ukrainian Polissya
Region, Archaeologia Baltica, 7, 109–19.
Summerfield, T. 2011. Filling the Gap: Brutus in the Historia Brittonum, The Anglo-Saxon
Chronicle MS F and Geoffrey of Monmouth, The Medieval Chronicle 7, 85–102. Amsterdam and
New York: Rodopi.
Szécsényi-Nagy, A. et al. 2015. Tracing the genetic origin of Europe’s first farmers reveals
insights into their social organization, Proceedings of the Royal Society B, 282 (1805),
20150339.
Tacitus. Agricola, Germania, trans. H. Mattingly, rev. with intro and notes by J. B. Rives. 2009.
London and New York: Penguin Books.
Tacitus. Annals. Available in a number of translations into English. A recent edition is trans. J. C.
Yardley. 2008. Oxford and New York: Oxford University Press.
Táin Bó Cúailnge survives in three recensions, the third of which is found only in fragmentary
form. The first two have been published in several translations, of which the most recent are: C.
O’Rahilly (ed. and trans.), Táin Bó Cualnge from the Book of Leinster [Recension 2], 1970, and
C. O’Rahilly, Táin Bó Cúailnge. Recension I, 1976. Dublin: School of Celtic Studies, Dublin
Institute for Advanced Studies. Accessible translations combining elements of both recensions
are The Táin, trans. T. Kinsella, 1969, Dublin: Dolmen Press and 1970, Oxford: Oxford
University Press; and The Táin, trans. C. Carson, 2007, London: Penguin Classics.
Talbert, R. J. A. 2000. Map-by-Map Directory to Accompany Barrington Atlas of the Greek and
Roman World. Princeton: Princeton University Press.
Taylor, J. 1980. Bronze Age Goldwork of the British Isles. Cambridge and New York: Cambridge
University Press.
Taylor, J. J. 1994. The Oliver Davis Lecture: the first Golden Age of Europe was in Ireland and
Britain (circa 2400–1400 BC), Ulster Journal of Archaeology, Third Series, 57, 37–60.
Telegin, D. Ya. and Mallory, J. P. 1994. The Anthropomorphic Stelae of the Ukraine: The Early
Iconography of the Indo-Europeans. Journal of Indo-European Studies Monograph 11.
Washington: Institute for the Study of Man.
Ternes, E. 1992. The Breton language, chapter 7 in The Celtic Languages, D. MacAulay (ed.),
371–452. Cambridge: Cambridge University Press.
Tertullian. De Anima (c. AD 203), trans. by P. Holmes as A Treatise on the Soul, in Translations of
the Writings of the Fathers Down to AD 325. 1870. Edinburgh: T. & T. Clark.
Thissen, L. 1993. New insights in Balkan-Anatolian connections in the Late Chalcolithic: old
evidence from the Turkish Black Sea littoral, Anatolian Studies, 43, 207–37.
Thomas, C. 1981. Christianity in Roman Britain to AD 500. Berkeley and Los Angeles: University
of California Press.
Thomason, S. G. 2001. Language Contact: An Introduction. Edinburgh: Edinburgh University
Press.
Thornton, D. E. 2003. Kings, Chronologies, and Genealogies: Studies in the Political History of
Early Medieval Ireland and Wales. Oxford: Unit for Prosopographical Research, Linacre
College.
Thorpe, L. (ed. and trans.). 1966. Geoffrey of Monmouth, The History of the Kings of Britain.
Harmondsworth: Penguin.
Tian, C. et al. 2008. Analysis and application of European genetic substructure using 300 K SNP
information, PLoS Genetics, 4 (1), e4.
Tian, C. et al. 2009. European population genetic substructure: further definition of ancestry
informative markers for distinguishing among diverse European ethnic groups, Molecular
Medicine, 15 (11–12), 371–83.
Tinner, W. et al. 2003. Climatic change and contemporaneous land-use phases north and south
of the Alps 2300 BC to 800 AD, Quaternary Science Reviews 22, 1447–60.
Tinsley, H. 1981. The Bronze Age, in Simmons and Tooley (eds) 1981, 210–49.
Tomlin, R. S. O. 1992. Voices from the Sacred Spring, Bath History, 4, 7–24.
Toner, G. 2000. Identifying Ptolemy’s Irish places and tribes, in Ptolemy: Towards a Linguistic
Atlas of the Earliest Celtic Place-Names of Europe, D. N. Parsons and P. Sims-Williams (eds),
73–82. Aberystwyth: CMCS.
Turek, J. 2006. Beaker barrows and the houses of dead, in Archaeology of Burial Mounds, L.
Šmejda (ed.), 170–79. Plzeň, Czech Republic: University of West Bohemia.
Turek, J. 2012. Origin of the Bell Beaker phenomenon: the Moroccan connection, chapter 8 in H.
Fokkens and F. Nicolis (eds), Background to Beakers: Inquiries into Regional Cultural
Backgrounds of the Bell Beaker Complex. Leiden: Sidestone Press.
Turner, J. 1981. The Iron Age, in Simmons and Tooley (eds) 1981, 250–81.
Uhlich, J. 2007. More on the linguistic classification of Lepontic, in Gaulois et Celtique
Continental, P.-Y. Lambert and G.-J. Pinault (eds), 373–412. Geneva: Droz.
Valério, M. 2014. The interpretive limits of the Southwestern script, Journal of Indo-European
Studies, 42 (3 & 4), 439–67.
Vander Linden, M. 2007. What linked the Bell Beakers in third millennium BC Europe?, Antiquity,
81 (312), 343–52.
Vasil’ev, S. 1999. The Siberian mosaic: Upper Paleolithic adaptations and change before the Last
Glacial Maximum, chapter 14 in Hunters of the Golden Age: The Mid Upper Palaeolithic of
Eurasia 30,000-20,000 BP (Analecta Praehistorica Leidensia, 31), W. Roebroeks, M. Mussi, J.
Svoboda and K. Fennema (eds), 173–95. Leiden: University of Leiden.
Velušček, A. 2004. Past and present lake dwelling studies in Slovenia: Ljubljansko barje (the
Velušček, A. 2004. Past and present lake dwelling studies in Slovenia: Ljubljansko barje (the
Ljubljana Marsh), chapter 5 in Living on the Lake in Prehistoric Europe: 150 Years of Lake-
Dwelling Research, F. Menotti (ed.), 69–82. Abingdon and New York: Routledge.
Verger, S. 2006. La grande tombe de Hochdorf, mise en scène funéraire d’un cursus honorum
tribal hors pair, Siris, 7, 5–44.
Vergnaud, L. 2013. The Bell Beaker funeral group from Sierentz ‘Les Villas d’Aurèle’ (Haut-Rhin,
France), chapter 5 in Prieto-Martinez and Salanova (eds) 2013, 51–59.
Villar, F. 2000. Indoeuropeos y no Indoeuropeos en la Hispania prerromana. Salamanca:
Ediciones Universidad de Salamanca.
Villar, F. 2004. The Celtic Language of the Iberian Peninsula, in Studies in Baltic and Indo-
European Linguistics in Honor of William R. Schmalstieg, P. Baldi and P. U. Dini (eds), 243–74.
Amsterdam: John Benjamins.
Vitali, D. 1991. The Celts in Italy, in Moscati et al. (eds) 1991, 220–35.
Vouga, P. 1923. La Tène. Monographie de la station publiée au nom de la Commission des
Fouilles de La Tène. Leipzig: Karl W. Hiersemann.
Všianský, D., Kolář, J. and Petříka, J. 2014. Continuity and changes of manufacturing traditions of
Bell Beaker and Bronze Age encrusted pottery in the Morava river catchment (Czech Republic),
Journal of Archaeological Science, 49, 414–22.
Walsh, A. 1991. Excavation at the Black Pig’s Dyke, Clogher Record, 14 (1), 9–26.
Warner, R. 1990. The ‘Prehistoric’ Irish Annals?, Archaeology Ireland, 4 (1), 30–33.
Warner. R., Moles, N. and Chapman, R. 2010. Evidence for early Bronze Age tin and gold
extraction in the Mourne Mountains, County Down, Journal of the Mining Heritage Trust of
Ireland, 10, 29–36.
Watkins, C. 1966. Italo-Celtic revisited, in Ancient Indo-European Dialects H. Birnbaum and J.
Puhvel (eds), 29–50. Berkeley: University of California Press.
Watkins, C. 2001. An Indo-European linguistic area and its characteristics: Ancient Anatolia. Areal
diffusion as a challenge to the comparative method?, in Areal Diffusion and Genetic Inheritance:
Problems in Comparative Linguistics, A. Y. Aikhenvald and R. M. W. Dixon (eds), 44–63.
Oxford: Oxford University Press.
Wei, W. et al. 2013. A calibrated human Y-chromosomal phylogeny based on resequencing,
Genome Research, 23(2), 388–95.
Whitehouse, N. et al. 2014. Neolithic agriculture on the European western frontier: the boom and
bust of early farming in Ireland, Journal of Archaeological Science, 51, 181–205.
Wilde, S. 2014. Direct evidence for positive selection of skin, hair, and eye pigmentation in
Europeans during the last 5,000 y, PNAS, 111 (13), 4832–37.
William of Newburgh, The History of English Affairs, Book 1, ed. with trans. and commentary by P.
G. Walsh and M. J. Kennedy. 1988. Warminster: Aris.
Winney, B. et al. 2012. People of the British Isles: preliminary analysis of genotypes and
surnames in a UK-control population, European Journal of Human Genetics, 20, 203–10.
Włodarczak, P. 2009. Radiocarbon and dendrochronological dates of the Corded Ware culture,
Radiocarbon, 51 (2), 737–49.
Woodward, A. and Hunter, J. 2011. An Examination of Prehistoric Stone Bracers from Britain.
Oxford: Oxbow Books.
Woolf, A. 2007. From Pictland to Alba 789–1070 (The New Edinburgh History of Scotland 2).
Edinburgh: Edinburgh University Press.
Wright, D. M. 2009. A set of distinctive marker values defines a Y-STR signature for Gaelic
Dalcassian families, Journal of Genetic Genealogy, 5 (1), 1–7.
Zavaroni, A. 2009. Footprints as a symbol of a divine guide of souls: focus on the Scandinavian
rock art, Papers presented at the XXIII Valcamonica Symposium 2009, 420–29.
rock art, Papers presented at the XXIII Valcamonica Symposium 2009, 420–29.
Sources of Illustrations
Images are listed by figure numbers.
Title page: Musée de l’Oise, Beauvais; 1 Drazen Tomic, after Koch 2006, p. 373; 2 British
Museum, London; 3 from Francis Grose, The Antiquities of England and Wales, new edition, Vol.
IV, c. 1784; 4 Joseph Christian-Leyendecker for The Century Illustrated Magazine, January 1907;
5 National Monuments Service, Ireland. Department of Arts, Heritage and the Gaeltacht; 6
National Museum of Wales, Cardiff; 7 British Library, London; 8 Drazen Tomic, adapted from
Soares 2010; 9 Drazen Tomic, after Richard Rocca; 10 Private Collection; 11 Drazen Tomic,
based on Koch 2007; 12 Musée Cantonal d’Archéologie, Neuchâtel; 13 from P. Vouga, La Tène,
1923; 14 British Library, London; 15 Musée d’Archéologie Nationale, Paris; 16 Nationalmuseet,
Copenhagen; 17 Francois Guillot/AFP/Getty Images; 18 Heritage Image Partnership Ltd/Alamy;
19 Drazen Tomic; 20 José-Manuel Benito Álvarez; 21 Wessex Archaeology. Drawing by Elizabeth
James; 23 Drazen Tomic, adapted from Myres 2011; 24 Drazen Tomic; 25 Drazen Tomic, based
on Sims-Williams 2006; 26 adapted from Nahkleh, Ringe and Warnow; 27 Drazen Tomic; 28
David Bezzina; 29 Drazen Tomic, after Inizan 2012; 30 Drazen Tomic, adapted from Lazaridis
2014; 31 Drazen Tomic, adapted from David Anthony 2007; 32 Dnipropetrovsk National Historical
Museum named after D. I. Yavornytsky, Ukraine; 33 Drazen Tomic, with information from Telegin
and Mallory figs 2, 19, and Dimitriadis, G. 2008. Looking for metals: megalithic monuments
between reality and mythology, in Geoarchaeology and Archaeomineralogy. Proceedings of the
International Conference, 29-30 October 2008 Sofia, R. I. Kostov, B. Gaydarska, M. Gurova (eds),
205–10. St. Ivan Rilski: Sofia, fig. 3; 34 Histria Museum, Constanta; 35 Musées Cantonaux du
Valais, Sion. Photo Hervé Paitier; 36 Musée Fenaille, Rodez; 37 Drazen Tomic; 38 DRCC-Museu
da Guarda; 39 British Library, London; 40 David Bezzina, after Kunst 2001; 41 Muzeul Național
de Istorie a României, Bucharest; 42 Drazen Tomic; 43 Museo de Valladolid; 44 Landesmuseum
Württemberg, Stuttgart; 45, 46 Drazen Tomic, adapted from Cunliffe 2008; 47 Erich Lessing/akg-
images; 48 Landesmuseum Württemberg, Stuttgart; 49 Photo Keltenmuseum Hochdorf/Enz,
Eberdingen; 50 above Musée du Pays Châtillonnais, Châtillon-sur-Seine; 50 below Musée du
Pays Châtillonnais, Châtillon-sur-Seine/Gianni Dagli Orti/The Art Archive; 51, 52 akg-images; 53
Drazen Tomic; 55 Drazen Tomic, after Koch 2007; 56 DEA Picture Library/Getty Images;
57Čestmír Štuka; 58, 59 The Trustees of the British Museum, London; 60 Drazen Tomic, after
Cunliffe 1997, fig. 55; 61 Staatliche Museen zu Berlin; 62 Posavski Muzej, Brežice; 63 Musei
Capitolini, Rome; 64 Drazen Tomic, after Lorria and Zapatero 2005, from Albertos 1990; 65
Private Collection; 66 Drazen Tomic, after Richard Rocca; 67 ML Design; 68 Werner
Forman/Corbis; 69, 70, 71 Drazen Tomic; 72 Musée Calvet d’Avignon; 73 Musée Crozatier, Le
Puy-en-Velay; 74 Drazen Tomic; 75 Museum of London/The Art Archive; 76 Drazen Tomic; 77
Meigle Sculptured Stone Museum, Perthshire; 78 Drazen Tomic; 79 LVR -LandesMuseum, Bonn.
Photo Jürgen Vogel; 80 The Trustees of the British Museum, London; 81 Miss G. D. Jones; 82
Carmarthenshire County Museum, Abergwili. Photo Dara Jasumani; 83 National Museum of
Ireland, Dublin; 85 Drazen Tomic, after Atlas of Irish History, 2nd ed. (2000), p. 19; 86 Patrick
Frilet/Marka/SuperStock; 87 ML Design, based on Campbell, John and Wormald 1982, fig. 50; 88
Royal Commission on the Ancient and Historical Monuments of Wales (RCAHMW); 89, 90
Drazen Tomic; 91 Trinity College Library, Dublin; 92 Drazen Tomic, after Koch 2007, §24, §390;
93, 94 Drazen Tomic; 95 National Portrait Gallery, London; 96 from Edward Lhuyd, Archaeologia
Britannica, Vol. 1, 1702; 97 Victoria and Albert Museum, London; 98 The Illustrated London
News, 1862; 99 Fred Tanneau/AFP/Getty Images; 100 Educational Images/UIG/Getty Images;
101 Hemis/Alamy.
Acknowledgments
Colin Ridler of Thames & Hudson had the idea for this book, Jim Mallory acted as midwife to its
birth, while Miranda Aldhouse-Green and David Miles supplied expert comments, which is not to
say that any of them shares all the views expressed herein. James Honeychuck was kind enough
to proofread the first two chapters. Thanks are due to Alan Reilly for generously sharing his
research on pressure blade-making and pointing out other useful papers. On the genetic side, I
am grateful to Sir Conor O’Brien, Baron Inchiquin, for permission to include his Y-DNA
haplogroup, and to Dennis Wright for consultation on it. Kenneth Nordtvedt was as helpful as
always on the estimated dating of subclades of Y-DNA haplogroup I. DNA distribution maps for
R1b-U152 and R1b-L21 were initially created by Richard Rocca.
Index
All page references refer to the 2015 print edition Numbers in italic refer to illustrations; numbers
in bold refer to maps Aedui 149–50; 71
Aeneas 24
Afontova Gora man 69, 70, 71, 73
Agricola (governor of Britain) 156–58; 76
agriculture, spread of 66–67
Alaric the Goth 163
Alba, kingdom of 184
Alban, St 162–63
Albanian language 104
alcohol 49–50
Aldhelm, abbot of Malmesbury 180
Alesia, siege of 151–52
Alexander the Great, and Celts 129–30
Alps 32, 33, 82–83, 86, 114, 115, 128
Altai Mountains 76
Alteuropäisch see Old European (IE) amber 53, 104–05, 107, 110
Ambiani 68
Ambrona Valley (Spain) 50
America 70
Amesbury Archer 53, 56; 21
Ammianus Marcellinus 123
Anatolia 75–76, 136
Galatian Celts 131–33
PIE homeland theory 66–67
ancestral north Eurasians (ANE) 70, 73; 30
ancient DNA, European
earliest dates for 44–45
Indo-European origins 202
Insular Celts and 202
and spread of R1b-L21 27
Yamnaya culture 76
Andronovo culture 78
Angles 13, 174; 78
Anglesey (Ynys Môn) 22, 177, 178
Anglo-Saxons
and Britons 174–77
genetics 175
 genetics 175
see also Saxons
animals, draught 74
Annales Cambriae (Welsh Chronicles) 23
annals 19, 21
Annals of the Four Masters 19–20
Annals of Ulster 168
Anthony, David 68
Antoninus Pius, emperor 156
Antonine Wall 156
Aosta, necropolis 89–90, 97, 115; 53
Aquae Sulis (Bath) 163
Aquitani 31; 71
Aragon 135
Arbroath, Declaration of (1320) 182
archaeology 7
and literature 18–20
archery equipment 52, 53
Argantonios 43
Ariovistus 149–50
Aristotle 31, 102
Arkaim 78
Arles, Christian Council 163
Armenian language 77; 31
Armorica 180–81, 187; 89
armour 102–03
see also weapons Arras 140
Arras culture, Yorkshire 38; 55
arrows/arrowheads 53, 112; 21
arsenic-rich copper 51, 74, 89
art, Celtic 118–19; 57, 58, 83
Art Nouveau, Celtic art and 196
Artchorp, Irish genealogy 168
Arthur, British war leader 22, 23
Arverni 149, 151–52; 71
Asia Minor, Galatoi (Galatians) 131–33
Assyrians 67
Asterix 151
Athenaeus of Naucratis 131
Atlantic Bronze Age 100
Atrebates 140, 154–55; 74, 76
Attacotti 159, 161; 78
Attalus I, of Pergamon 132; 63
Aubrey, John 3
Augustine of Hippo, St 163
Augustorium (Limoges) 40
Aulus Plautius 154–55; 74, 76
Aulus Plautius 154–55; 74, 76
Austria 128
Auxilius (missionary to Ireland) 167–68
Avienus, Rufus Festus, Ora Maritima 46
Aylesford-Swarling culture 140
axes 52, 75, 91
Aztecs 70

Baetica 134
Baia de Cris (Romania) 82
Baikal, Lake 69, 71; 29
Balkan languages 77; 31
Balkans 60
collapse of farming 75–76
La Tène culture 128–31
metallurgy 84
Yamnaya culture 72
Baltic 104–05
Barrington Atlas of the Greek and Roman World (2000) 59
Basque language 31
Bathanattus 131
Bede, the Venerable
Dál Riata 182–83
Picts 182–83, 194
Ecclesiastical History of the English People 28,174,194
beehive quern stones 137
beer, Bell Beaker and 49–50
Belgae 31, 60, 123–24, 133; 71
against Rome 150–51
coins 140; 68
in Britain 139–40; 68
language 133
see also Gauls
Belgium 30
Bell Beaker 48–56, 73; 20; 19
eastern and western 97–98
language 57–61; 23
mobility 56, 115
and stelae 89–91; 39
Bell Beaker routes, and development of Celtic cultures 94–99, 115; 42
Bell Beaker ware 48–51, 73, 89; 20; 19
All Over Corded (AOC) 73, 76, 91–92
Epi-Bell-Beaker in Portugal 52
geographic locations 51, 91
Maritime 91–93, 96, 98; 40
origins of 91–94
Protruding Foot Beaker (PFB) 92
 Protruding Foot Beaker (PFB) 92
style of ornamentation 91–92
belts 82, 91; 34
Benraw (Ireland) 98
Berber language 95, 96
Beringia land bridge 70
Bernicia, Angle kingdom 13; 87
Bettelbühl necropolis 106–07
birch bark hat 107; 48
birds, and Celts 122; 58
Bithynia 131
Black Pig’s Dyke (Armagh) 18; 67
Black Sea 25, 60, 70, 75, 77, 93, 103; 31, 33, 45
boar’s tusks 97
Bohemia 117, 118, 120; 55
Boii tribe 120, 123, 128, 143, 144, 148; 57
Bolgios 130
Botteghino (Italy) 83
Boudica 36, 110
bow 89, 91
Bragança district (Portugal), bronze workings 52
Brennos, attack on Delphi 130–31
Breton language 21, 31, 116, 187–88; 92
Bretons 209–10
Brian Boru 207–09
bridles 104
Brigantes 155–56; 67, 74, 76
Britain
ancient genome sequences 175
British and Anglo-Saxon kingdoms 87
Celtic name for 45–46
chariot burials 36, 38, 124
gold 53
ogham 164, 165; 82
Saxon provinces 175; 87
wine trade 125–26; 59
Yr Hen Ogledd (The Old North) 175, 187
Britannia (Roman province) 12, 21, 59, 152–61, 170
attacked by barbarians 158, 159–61; 78
Celtic placenames 59, 123, 137–38, 168; 25
Irish raiders 163–64
leaves Roman empire 161
Pritani 169–70
Roman conquest of 153–61; 74, 76
Saxon provinces 175
British Celts, DNA profile 26–27
British Isles
 Christian Celts 162–85
development of Celtic 116
names for Brittonic people 169–70, 201
source of tin 52, 98
tribes 152–59; 74, 76
Britons
abroad 180–82
and Anglo-Saxons 174–77
genetics 175
inter-marriage with Saxons 176
slaves in Ireland 163
Brittany
British migration to 180–81, 187; 89
economic emigration from 191
Inter-Celtic Festival (Lorient) 197–98; 99
Maritime Bell Beaker 96
stelae 98
and Y-DNA R1b-L21 26–27
Brittonic 12–13
poetry 14–15
see also Celtic languages
Brixlegg (Austria) 83, 84
bronze 74, 102, 106, 137
Bronze Age 35
armour and weapons 102–03; 44
Bell Beaker folk and 51–52
mobility of Celtic speakers 99–100
Brutus (legendary founder of Britain) 22, 24, 25, 193–94; 7
Buchanan, George 193–94; 95
Bulgaria 72, 82
burials
Bell Beaker 52–56, 89–90
chariot 34–38, 117, 123–24; 14
Copper age 84
cushion stones 53
flat graves 117
grave goods 52–6, 117; 21, 22
Heuneburg 106–07
Hochdorf 107–09; 49
Iberia 89
kurgan burials 74–75, 76, 80–81
Lady of Vix 109–11; 50
practices 38
Single Grave culture 76
stelae 80–82, 84; 35
wagon burials 104–11, 117; 49
Yamnaya 74–75, 80–82
 Yamnaya 74–75, 80–82
see also cremation
Byzantium 131

Cabezo Juré (Spain) 87; 37

Cabrières (France) 86
Cadfan, king of Gwynedd 22, 178; 6
Cadiz (Spain) 43
Caesar see Julius Caesar
Caledonia and the Caledonii 59–60, 156, 158
Caledonians 9, 59
Calleva 156
Camden, William 194
Campbell, Ewan 182
Canada, Scottish Gaelic 191
Cantiaci 140; 74, 76
Caratacos (Caratacus) 155
carnyx/carnyces (war horns) 39–42; 39, 17, 18
replica created 40
carpat (chariots, Irish) 36
Carpathian Basin 82, 84, 93, 96, 97, 98, 102, 104, 128; 31
Begleitkeramik 97, 98
Bell Beaker folk 96–97
Carpathian Mountains 74
Cartagena (Iberia) 143, 146
Carthage 143, 144
Carthaginians 46, 143
wars with Rome 143–46
see also Phoenicians
Cartimandua 155
carts 74
Caspian Sea 103
Cassander of Macedon 130
Cassius Dio 158
Cassivellaunos 153–54
Castro culture 100
Cathbad 15–16
Catuvellauni 140, 153–55; 74, 76
Caucasus 68, 103
cauldron, bronze 108; 49
cavalry 30
Çayönü Tepesi, obsidian blades 70
Ceamurlia de Jos (Romania) 82
Celestine, Pope 163, 164
Celtiberians 98, 133–35
mercenaries 145
source of slaves for Rome 146
 source of slaves for Rome 146
Celtic art 118–19, 196; 57, 58, 83, 97
Celtic chiefdoms, and trade routes 46
Celtic and Italic languages 6, 56–61, 57, 200; 1, 23, 24
Celtic languages
conjectural tree 54
cultural/romantic interest in 195–98; 97, 98, 99
development in British Isles 95, 116
linguistics 202
loss of and revival of 187–205
lost in Iberia and Gaul 162
P-and Q-Celtic 116
population movement and 189–92
see also Brittonic
Celtic names, in Iberia 43–45
Celtic placenames 25
Asia Minor 133
British Isles 59, 123, 137–38, 168; 25
Europe 60
Gaul 31, 58–59, 123; 25
Germany 60
Iberia 123, 133–35; 64
Ireland 137–38, 168
Italy 60, 127; 24
see also placenames
Celtic tribes
British Isles 152–59; 74
Gauls 36–37, 40, 46, 116, 149–52; 71
Ireland 67
see also individual tribes by name Celtici 134
Celtoscepticism 200–02
Celts
21st century 204
and Alexander the Great 129–30
ancestral heroes 80
aristocracy 107
in Balkans 130–31
in the Balkans and Asia Minor 131–33
Bell Beaker and 47
bronze spearman 61
carnyces (war horns) 39–42
Christian Celts 162–85; 80
and Cimmerians 104
collective names for 201
Continental 9, 34, 36
deities 41
expansion eastwards 128–33
 expansion eastwards 128–33
first inscriptions 113, 128
Golasecca culture and 113–15
Insular 9, 26, 78, 158
and Macedonia 129–31
as mercenaries 126, 129, 131
migration 42–47
religion 40–41
settlements 119–20
shrines and sanctaries 40, 80
source of slaves for Rome 146
theories of origins 7–10
timelines 8
tribal culture 100, 201
tribal names 36–37, 116, 138; 67
see also Hallstatt culture (Hallstatt C and D); La Tène culture Cenomani 127, 143, 144
Centre for Advanced Welsh and Celtic Studies 7
Cerdic 176–77
Cernavoda culture 75
Channel Isles, stelae 89
chariots 30, 34–38; 4, 10
burials 34–38, 117, 123–24; 14
Celts and 117–18
Gauls and 30–31; 10
in the Iliad 36
La Tène 33–34, 117–18; 13
reconstruction 118
Sintashta 78
Chastel-Arnaud (France) 86
Childe, Vere Gordon 48, 50
Christianity
in the British Isles 162–64; 80
in Ireland 16–17, 162, 167–69, 173; 86
saints and druids 16–17, 180–81
chromosomes 26
churches, in Brittany 180–81
Ciempozuelos (Iberia), Bell Beaker ware 92–93, 98; 20
Cimmerians 24, 25, 103–04; 45
Claudius, emperor 154, 155; 76
climate
effect on population growth 137
effect on population movement 72, 74, 75–76, 135
Clontarf, battle of 207
Clusium (Chiusi) 124, 127
Coeurdoux, Gaston 64–65
Coggalbeg (Ireland) 22
coinage/coins 10, 68
coinage/coins 10, 68
Atrebates and Catuvellauni 154; 75; 74
Belgae, Britain 139–40
Celtiberian 134
Parisii 118; 56
Coligny calendar 115–16
Columba, St 17, 184–85
Conchobar mac Nessa 16
Connachta 139
Constantine (king of Dumnonia) 180
Constantine (Roman emperor) 163
Constantius Chlorus 158
copper 51, 72, 82–83
mines 51, 84–86
trade 43
working 51, 82–89
Copper Age 9
Bell Beaker folk and 49–50
and PIE 68
Yamnaya culture 74–75
Coptic 95, 96
Corded Ware 73, 76, 91–92
Corlea trackway (Ireland) 20; 5
Cornwall
Celtic identity 177
Cornish language 187, 192; 100
tin 52, 109, 155
Uí Liatháin settlement 167
Corsica 136
couch, bronze 108; 49
cremation 100
grave goods 104; 44, 59
Urnfield culture 100
Croatia 130
Cruithin/Cruithni 169–74
genetics 171
Csepel Island (Danube), Bell Beaker 96–97; 19
Cú Chulainn 11, 17, 36; 4
Cucuteni-Tripolye culture 71, 72, 74, 76, 78; 41
culture 7
Cumbric language 187
Cunedda 179–80
Cunliffe, Barry 94–96
Cunobelinos 154; 75
curse tablets 12
Cymmrodorion, Honourable Society of 196
Cymry 177, 178
Dacians 39
daggers 53, 75, 84, 86, 89, 91, 104, 108; 35, 48
dairy farming 71, 74
Dál nAraidi 170
Dál Riata 182–83, 184; 90
‘Dame de St Sernin, La’ 36
Danube 31–32, 72, 75, 76, 82, 106,128
Proto-Celtic 94
Urnfield 104
Dark Ages 21, 22, 27
Darvill, Tim 53
Deira, Angle kingdom 13; 87
Déisi Muman 166–67; 85
genetics 168
Delphi (Greece) 123, 130–31; 60
Demetae 165–66; 74, 76
Demetia (Dyfed) 177
Irish settlement of 165–67
dendrochronology see treering dating Dermot MacMurrough, king of Leinster 190
Deskford (Scotland), carnyx 40, 42
Desborough, mirror 2
dialect, and language 65–66
Diocletian 163
Diodorus Siculus 109
on druids 37–38
on Gauls 30, 125
Dionysius of
Halicarnassus 124
DNA
basics 26–27
gender and 26
and language 9–10, 57; 23
see also Y-DNA haplogroups Dnieper River 78
Dnieper-Donets I and II 72, 73
Dniester River 76
Dobunni 154, 155; 74, 76
Dorsey earthwork 18
druids 15–17, 38, 123, 133; 3
and ogham 164–65
Roissy burial 37; 15
Dumbarton, lord of 12, 13
Dumnonia 180; 87
Dumnonii 156, 180; 74, 76
Dunaverney (Ireland) 58
Durocortorum 124
Durotriges 155; 74, 76
Dyfed (Wales) 177; 87
Dyfed (Wales) 177; 87
Irish settlements 165–67
Dying Gaul, the 118–19; 63

early European farmers (EEF) 73; 30

earthworks 18, 177
Ecclesiastical History of the English People 28, 174
see also Bede
Edercloon (Ireland) 34–35
Edict of Toleration (AD 311), of Christianity 163
Edinburgh Rock 13
Elizabeth I, settlement in Ireland 190–91
Emer, wife of Cú Chulainn 11
England, lingustic effect of Norman conquest on 190
English language, dominance of 187, 192–93
English settlers, in Celtic lands 190–91
Eochaid Airem, king of Tara 18–19
causeway 20; 5
Eochaid son of Artchorp 166–67, 168
Eóganacht 28, 173; 85
Ephorus 43
Érainn 138
Erzgebirge Mountains, tin 52
Estremoz (Portugal) 98
Etruria 104, 105
Etruscans 58, 113, 127, 143
driven out by Celts 48, 126
Gauls and 34, 126, 127–28; 60
and Golasecca culture 112
Europe
federalist 199, 200
European Coal and Steel
Community 199
European Economic
Community (EEC) 199
Ezerovo 78

farming
Balkan farmers 76
beginning of in Europe 67
dairy-farming 71, 74
Dnieper-Donets II cattle farmers 72, 73
Feast of Bricriu, The 11
feasting 12, 13, 50; 58
Fenech 28
Féni 28, 173, 178
Fénius Farsaid, Irish language and 28
Ferry Fryston, chariot burial 38, 124
Finnish 68
Fir Domnann 28, 156
Fishbourne (Roman palace) 154, 156
footprints 82; 32, 34
foraging, transition to farming 72
fortified settlements 78
Fraga dos Corvos (Portugal) 52
France 51
suppression of regional languages 188
Franco, General 200
French language, dominance of 187
Furger-Gunti, Andres 118
Gaedil 25
Gaelic language 12, 31, 116
Irish and Scottish 88, 182, 198
see also Irish language
Gaels 182
Galatians (Galatoi) 131–33; 63
Gaulish language 133
St Paul and 133
Galicia
language 133–34
and Y-DNA R1b-L21 27
see also Gallaecia (Spain) Gallaeci (Callaeci) 100, 147
Gallaecia (Spain) 134, 181–82; 64, 89
see also Galicia
Gallay, Alain 89
Gallia Narbonensis 148
Gallic Chronicle, on Saxons in Britain 174, 175
Gaudo (Italy) culture 84
Gaul 30–31
Celtic placenames 31, 58–59; 25
roads and bridges 148
Transalpine and Cisalpine 30
Vercingetorix rebellion 151–52; 73
Gauls
carnyces (war horns) 39–42; 17
chariots 30–31; 10
Cisalpine 30, 143–44, 148
connection with British Isles 98–99
deities 41
description of 30, 31
Dying Gaul, the 118–19; 63
east of the Rhine 60
eastward expansion 128–33
in Iberia 135
 in Iberia 135
Julius Caesar and 30–31, 35, 40–41, 123, 147–52; 71
languages 31, 115
move into Italy 124–28; 60
political systems 147
as seen by Romans 9–10, 28, 30–31
territorial expansion 123–40, 136, 137; 60, 66
Transalpine 30
tribes 36–37, 40, 46, 116, 149–52; 71
and wine 125–26
Y-DNA R1b-U152 136; 136
see also Belgae
genealogies
Irish and Welsh 166–67
spurious 24–25
Genesis 24, 25, 194, 195
genetic code 26, 44
genetics
Bell Beaker folk 57; 23
British and Irish 26–27, 139, 168, 175, 208–11
Déisi Muman 168
Europeans’ sources 73
Mal’ta boy and Afontova Gora man 70, 71, 73
People of the British Isles study 175
Picts 159
R1b in British Isles 99, 136, 139, 208–11
Royal Stewart 210
Scandinavian branches of R1b-DF27 99
spread of Gauls 136; 66
traces of La Tène in Ireland 139
Geoffrey of Monmouth 21–23, 28; 75
History of the Kings of Britain (1138) 22–23
Geraint, king of Dumnonia 180
Gerald of Wales 190
Gergovia 151
Germani 9; 71
against Rome 150–51
expansion 120, 123, 148
mercenaries in Gaul 149–50
Germania 123
Germanic 76, 94
Germanus, St 163
Germany (North), Bell Beaker 97
Gildas 21
on Anglo-Saxon incomers 174, 175
on British kings 162, 163, 165–66
and king Constantine 180
 and king Constantine 180
on Maelgwyn 178
Ruin of Britain 21
Glauberg (Germany) 111–13, 116; 51; 46
Göbekli Tepe (Anatolia) 50
Gododdin 12–14, 36, 178, 186; 87
Goidel 25
Golasecca culture (Italy) 113–16, 126; 53
gold
Bell Beaker folk and 53–56, 98
in Britain 155
in Ireland 137
ornaments and jewelry 53, 55, 75, 107, 108–10, 117; 22, 50, 52
trade, Tartessos 43
Gomer 25
Gorge Meillet, La 34; 14
Gotthard Pass 114; 53
Gray, Thomas, The Bard 195
Greek 77
and Sanskrit 63–64
Greeks, knowledge of Celts 31–32
Gundestrup Cauldron 39, 41–42; 16
Gwent 177; 87
Gwynedd 22, 177–78; 87
Cunedda dynasty 178–80
Gwyneddigion Society 196

Hadrian, emperor 156

Hadrian’s Wall 13, 156, 158
hair binders 53, 75, 80
Hallstatt (site) 33
Hallstatt culture 33, 94; 11, 46
aristocrats 104–06
in Britain 124
sword 106–06; 44
trading routes 105; 46
wagon burials 104, 117
weapons 103; 44
see also Celts; La Tène culture Hamangia-Baia (Romania), stela 82, 91; 34
Hamilcar Barca 143
Hannibal 143–45
Hanno 146
Harrison, Richard 89
Hattic 67
Hecataeus of Miletus 31
Helvetii 148; 71
Henry II, king of England, invasion of Ireland 190
Henry II, king of England, invasion of Ireland 190
Herodian 158
Herodotus 31, 32, 43
Heuneburg (Germany) 106–07, 116; 47; 46
Heyd, Volker 89
high crosses, Irish 36
hillforts, Hallstatt 106; 47
Himilco 46
Hinxton, DNA 26, 175
Hirschlanden, Warrior statue 107, 109; 48
Hispania (Roman province) 147
Historia Brittonum 21, 23–25, 28, 167; 7
History of the Kings of Britain (1138) 22–23
see also Geoffrey of Monmouth Hittite/Hittites 67, 75
Hochdorf (Germany) 107–08, 110; 49; 46
Hochmichele (Germany) 109
horses/horseriding 74, 75, 87, 96, 104; 18
Hulín-Pravčice (Czech Republic) 53
Hungary 76, 96, 97
hunter-gatherers, western (WHG) 73; 30
Hwicce 176

I Celti exhibition 199–200

Iberia
Bell Beaker burials 52
Castro culture 100
Celtic languages and 94–96, 98
Celtic placenames 123, 133–54; 64
Copper Age culture 87–89; 37
Gauls 135
gold 53
horses 87
Indo-European placenames 62
La Tène culture 42
Neolithic burial chambers 89
Rome versus Carthage 144–47; 70
stelae 88–89, 98, 88
tin 52
trade with Greek world 43
see also Hispania; Portugal Illiad 24, 36, 75, 102
Illyrians
conquered by Rome 148
and Gauls 123, 128–29, 130, 131
India, Andronovo culture and 78
Indo-European languages 62, 63–68, 72; 27
ancient DNA and 202
relationship tree 64
word order and Celtic 95–96
 word order and Celtic 95–96
Indo-Europeans
dispersal of 75–78, 98
homeland 28
lifestyle 72–75
origins 66–75
Ine, of Wessex 180
Insubres 143, 144
inter-marriage 192
Saxons with Britons 176
International Congress of Prehistoric Anthropology and Archaeology (Bologna 1871) 34
Iona 184; 90
Iranian language, contact with Proto-Celtic 94, 104
Ireland 12
Anglo-Norman incursions 190–91
Bell Beaker culture 52, 98
Celtic names 45–46, 138; 67
Celtic placenames 137–38, 168
chariot burials 34–35
Cruithin incomers from Britain 169–74
division of 28, 138
dynastic kindreds 170–74; 85
early Christians 164, 167–69
Easter Rising (1916) 198
emigration of Irish speakers 191–92
Gaelic language festivals 198
Gaelic language health 88
Gaelic League 198
gold 53
La Tène culture 42, 135–38; 67
language shift 173
missions of Palladius and Patrick 164, 167–69
nationalist politics 198, 199
ogham 164–65; 81
Plantation of Ulster 191
raids on Britannia 163–64
Romano-British presence 156; 67
see also Northern Ireland
Irish
DNA profile 26–27
heroic poetry 14
Iwerni 46
migration to Scotland 182–85; 90
myths and legends 11–12
Irish language 28
Feiseanna and Oireachtas festivals 198
ogham 164–65; 81
 ogham 164–65; 81
Old Irish 14, 165, 173, 182
Primitive Irish 164, 165, 173
Q-Celtic 116
in the Republic of Ireland 88
see also Gaelic language
Irish surnames 106, 168, 170, 171, 206, 207–09
Iron Age 18, 33, 100; 5
wheel 35
iron-working 98, 104, 106, 120, 124
Hallstatt culture 105–06
La Tène culture and 117, 124
Irvine surname 208
Isidore of Seville 24, 25, 36, 158
Isle of Man 165, 178, 187, 197, 198; 97
isotopes 38, 53, 56, 97
Israelites 28
Italic languages 201–02
Italo-Celtic language 60–61; 42
Italy
Celtic placenames 60, 127; 25
La Tène culture and 34, 127–28
Y-DNA R1b-U152 136; 66
Iwerni (in Ireland) 138; 67

James I, of England, and VI of Scotland, Plantation of Ulster 191

Japheth 24–25
Jerome, St 163
on Galatian language 133
Jones, Sir William 63–66, 76
Josephus 24–25, 28
Julius Caesar
in Britain 153–54
chariots in Britain 36
conquest of Gaul 30, 147–52; 71
druids and verse 15
elected consul 147
on Gaul and Gauls 30–31, 35, 40–41, 42, 123, 193
Lepontii 113
oppida 119–20
Jutes 174; 78
Jutland, Bell Beaker 97

Kanesh (Neša) 67
Kargaly 75
Kells, Book of 185; 91
Keltoi 31, 201
Kernosovka (Ukraine), stela 32
Kernosovka (Ukraine), stela 32
Kilnasaggart Stone 86
Kirkhaugh (England) 53
knives, copper 53
Knox, Archibald 97
Koch, John
origins of Celtic 94–96
Tartessian language theory 43–45
Kopp, Hansli 33
Kortlandt, Frederik 61
Krahe, Hans 61–62
krater, bronze 110
Kromsdorf (Germany) 57
Kültepe, excavations 67
kurgan burials 74–75, 76, 80–81
La Tène (site) 33; 12, 13
La Tène culture 33–34, 116–21; 12, 13
in the Balkans 128–31
in Britain 38, 42, 116, 124–26, 137; 65
in British Isles 135–38
Celtic art 118–19; 57
and Celtic language 94, 116
chariots 33–34, 117–18; 13
and Hallstatt 116
insignificant in Iberia 42
in Ireland 42, 135–38; 67
in Italy 34, 127–28
warriors 116–18; 61
see also Celts; Hallstatt culture Laigin 178; 85
Lake Trasimene, battle 143
language
Bell Beaker 57–61; 23
Cimmerians and Celts 104
Cisalpine Gaulish 128
and dialect 65–66
DNA signatures and 9–10, 57; 23
Indo-European 62, 63–66, 72; 26; 27
Lepontic 113–15, 128
Proto-Indo-European (PIE) 66–72
see also individual languages by name Languedoc, copper cultures 86; 36
Latin
dominance in Roman period 187
in Ireland 173
and Proto-Indo-European 115–16
and Sanskrit 63–64
in western Europe 162
Lebor Gabála Érenn (The Book of the Taking of Ireland) 25, 28–29, 156
Lebor Gabála Érenn (The Book of the Taking of Ireland) 25, 28–29, 156
Leceia (Portugal) 88, 94; 37
Lepontic language 113–15, 128
Lepontii 113–15
Levant 68
Lhuyd, Edward, Archaeologia Britannica 195; 96
Life of Patrick 16
Ligurians 31, 61, 95, 100, 136, 144
Limovici, Gaulish tribe 40
linguistics 65, 202, 204
literacy 14, 67
literature (Irish) 14, 17–20, 156, 174
and archaeology 18–20
monastic annals 19–20
Annals of the Four Masters 19–20
Annals of Ulster 168
Expulsion of the Déisi 166
Táin Bó Cúailnge 17–18, 35, 36
Livy 34, 124, 131
Ljubljana marshes 84–86
Llangadwaladr (Anglesey) 22, 178; 6
Llŷn Peninsula, and Irish settlers 178
lockenringe see hair binders Longroiva (Portugal) 89
Los Millares (Spain) 88; 37
lunula 55–56; 22
Lusitanians 61, 100, 133, 147
Lutetia Parisiorum (Paris) 37
Luwian language 75
Luxembourg 30

Macalister, R. A. Stewart 28–29

MacAlpin, Kenneth 9, 184
McCone, Kim 116
Macedonia, Celts and 129–31
MacFarlane, Clan 211–12
MacNeill, Eoin 9, 195, 198
Maeatians 158
Magnus Maximus 167
Magog 25
Maikop culture 68, 74–75
Mallory, J. P. 20, 66, 68
Mal’ta boy 69, 70, 71, 73; 28
mammoth 69; 28
Man see Isle of Man Manching (Bavaria) 120
Mandubracios 153
Maniquet, Christophe 42
Manx language 165, 187
Manx language 165, 187
Marmara, Sea of 71
Marne department (France), chariot burials 34, 37, 123; 14
Mary, Queen of Scots 210, 211–12
Marzobotto (Italy) 34
Massalia (Marseilles) 31, 43
Maxen Wledic 167
mead 50
Medb, queen of Connacht 17
Medes 24
Mediolanum (Milan) 127
megalithic tombs, Bell
Beaker burials 52
memorial stones
in Gwynedd 178
Kilnasaggart Stone 86
Latin and ogham 166; 82
Llangadwaladr 22, 178; 6
Penbryn 88
menhirs, statue-82
mercenaries 126, 129, 131, 145, 149–50, 174
Merlin 15
metallurgy 48
and Bell Beaker culture 48, 51, 53, 84
France 86
Hallstatt/Celtic culture 105–06
Iberia 87–89; 37
Ireland 137
Italy and Tyrol 84
Maikop culture 74–75
PIE spread 68
Urnfield culture 104
migration 10
Bell Beaker routes 94–99; 42
Celts and 42–47
Pliny and 124–25
pressure blade-making 69–71; 29
Stelae People 82–84; 33
Mikhailovka culture 80–81
Milan (Italy) 127, 144
milk 66, 74
mitochondrial DNA (mtDNA) 44–45
Mochta 169
Mocissos (Portugal) 89; 37
Mochuda, St 17
Moira, battle of 174
Mongolia 70
Mongolia 70
Mont Lassois (France) 109–11, 116; 50
Monte Loreto (Italy) 84
Moravia 128
Morocco, Bell Beaker 51, 93
Morris, Richard and Lewis 196
Moselle, River 117
Mourne Mountains 52, 53, 137
Muirchú, Life of Saint Patrick 169
Munster 138
myths and legends
Brittonic 12–13
Irish 11–12

Narbo (Narbonne) 31
Native Americans 70
Navan fort (Ireland) 18
Nennius 21
Nervii 176
Nešli language 67
Netherlands, the 30, 123
Bell Beaker ware 91–92
Newbridge (Scotland), chariot burial 38, 124
Niall of the Nine Hostages 139
Nicomedes, Bithynia 131–32
Noah 24–25
nomadic lifestyle 103–04
Normans, in Britain 190
Norn language 190
Northern Ireland
Northern Ireland Assembly 203
Provisional IRA and Protestant para-militaries 199
see also Ireland; Ulster
Nutt, Alfred 36

obsidian 70
Offa’s Dyke 177
ogham 164–65, 165; 81, 82
Old European IE (Alteuropäisch) 61–62, 77, 98; 42
oppida (Celtic settlements) 119–20
oral history 20, 21
Ordovices 178; 74
Orkney 189–90
Oscans 201–02
Ostimioi, Celtic tribe 46
overpopulation 124
oxen 74
P-Celtic 62, 95, 115–16, 137
Palladius, bishop 164, 167–68
Palmela points 96; 43
Paris 37
Treaty of (1951) 199
Parisi (Yorkshire) 38, 124; 74, 76
Parisii (Continental) 37–38, 123–24
coin 118; 56
Parsons, James 195
Patagonia, Welsh-speaking colony 191
Patrick, St 9, 162, 163, 167–69
Confession 162, 167, 169
druids and 16
shrine 83
Paul, St, and Galatians 133
Pelagius 163
Péret (France) 86
Pergamon (Turkey) 132; 63
Petilius Cerialis, governor of Britain 155; 76
Petit-Chasseur, necropolis complex 89–91; 35, 39; see also Sion Petroc, Cornish/Breton saint
180
Petuaria (Brough-on-Humber) 37
Pezron, Paul-Yves 194, 195
Philip II of Macedon 129
Phoenicians 43–44
in Iberia 95
see also Carthaginians
Phokaeans 43
Picts 13, 17, 158–61, 194; 79; 78
Christianity and 184–85; 91
genetics 159
and Scots 182–85
symbols 159; 77
Pillars of Hercules (Strait of Gibraltar) 31
Pinkie Cleugh, battle of 212
placenames 31
Geoffrey of Monmouth and 23
Gwynedd and Armorica 178
Indo-European in Iberia 62
see also Celtic placenames Plachidol (Bulgaria) 77, 78
Plaid Cymru 199
Pliny the Elder 113, 134
Gaulish motives for invading Italy 124–25
ploughs 74, 75
Plutarch 152; 73
Po Valley 34, 48, 60, 126, 127
poetry, Irish 14
poetry, Irish 14
Poland, Bell Beaker 97
Polybius 39, 126–27
Gauls in Po Valley 126–27
the Roman empire 142–43
Pomerania 105, 106
Pompeius Trogus 122, 124
Portugal 44, 51, 55, 80
Bell Beaker 93
metallurgical workings 52, 88–89
see also Iberia
pottery
Begleitkeramik 97
Bell Beaker ware 48–51, 73, 89; 20; 19
in Britain 139
earliest 71
making of 51, 72, 92
Romania 41
Yamnaya Cord-impressed 92
Powys 177
Pratchett, Terry 15
Pre-Germanic dialect 76
pressure blade-making 69–71; 29
Pritani 169–70
prophecy 16–17
Proto-Anatolian 75
Proto-Balto-Slavic 78
Proto-Celtic 61, 94–95, 98, 115, 116
Proto-Germanic 76
Proto-Indo-European (PIE) 66–72; 26; 42
family trees (conjectural) 26, 54
vocabulary 67, 72
Proto-Italo-Celtic 61, 76; 31
Proto-Semitic language 68
Proto-Uralic language 68
pseudo-history 21–25, 193
Ptolemy (Claudius Ptolemy), Geography 36–37, 40, 59, 116, 124, 137–38, 155–56; 67
Lepontii 113
Purser, John 40
Pytheas 46

Q-Celtic 115–16
Quedlinburg (Germany), Y-DNA R1b-P312 57; 23

radiocarbon dating 18, 84, 92, 94

Ramsauer, Johann Georg 33
Red Branch Cycle (Irish literature) 17
Reims (France) 117, 124
Remedello (Italy)
culture 84, 89; 35
daggers 86, 91; 35
Remi 123–24, 150
Renaissance, Celtic and Classical sources 193–94
Renfrew, Colin 66–67
Repin culture 74
Rheged 14–15; 87
Yr Hen Ogledd (The Old North) 175, 187
Rhine, River 60, 123
chariot burials 117
Proto-Celtic 94
Rhône River/Valley 86, 97, 106, 109, 148
Rinaldone (Italy) culture 84
river-names, Alteuropäisch (Old European) hydronymy 61–62, 98
road signs, bilingual
signage 203; 100
Roissy, chariot burials 37, 38, 123; 15
Roman empire 142–61; 69
Christianity in 162–64; 80
fall of 13
Romance languages 58, 162, 187
Romania 72
Cucuteni pottery 93
stelae 82; 34
Romans
in Britain 153–61; 74, 76
in Gaul 147–52; 71
and Gauls in Italy 127
governing method 156–58
roads and bridges 148
self identity 201–02
view of Gaul and Gauls 30–31
Rome
sacked by Gauls 122, 127
slave-taking policy 146
tribes in northern Italy subjugated 143–44
wars with Carthage 143–46
Rome, Treaty of (1957) 199
Ross Island (Ireland) 51, 56

Saami 68
Saguntum 144–45
Saint-Romain-de-Jalionas (France) 109
Salanova, Laure 92–93
salt mines 33
salt mines 33
Samara 71, 73
Samson, Cornish/Breton saint 180–81
Sangmeister, Edward 93
Sanskrit 63–64
Sardinia 72, 84, 86, 97, 143
Sava, River 84
Saxons 159, 161, 162; 78
see also Anglo-Saxons
scabbard 40; 12, 62
Scipio Africanus 143, 145–46
Scipio, Gnaeus 144–45
Scipio, Publius
Cornelius 144–45
Scordisci 128, 131
Scoti (Scots) 159, 162, 182–84; 78, 90
Christianity and 184–85; 91
Scotland 152
Celtic placenames 59–60
Dál Riata kingdom 182–84; 90
emigration of Gaelic speakers 191–92
modern Gaelic 88
nationalist politics 199
Scots language 187, 190; 94
in Ulster 190
Scottish National Party 199
Scottish Parliament 203
Scythians 24, 25, 28, 78, 118; 45
and Celts 28–29, 78
Secondary Products Revolution 74
Secundinus 167–68
Seine River 109, 116
Semitic 95
Senones 143
Septimius Severus, emperor 158
Sequani 149; 71
Serbia 72
Sherratt, Andrew 74
Shetland 189–90
Siberia 68, 70, 71; 29
Sicily 48, 143
Silk Road 76
Silures 177; 74
Silurians 9
Simplon Pass 114; 53
Sims-Williams, Patrick 59
Singidunum (Belgrade) 128
Singidunum (Belgrade) 128
Single Nucleotide Polymorphism (SNP) 26
Sintashta 78
Sion (Switzerland) 89–91, 97, 115; 35, 39; 53
skull shape 56, 97
slaves
Britons in Ireland 163
Roman empire and 146
sledges 74
Slovakia 97, 128
Slovenia 130
Spain 28, 51, 52, 200; see also Iberia spearheads 96, 112
stelae 77, 80–91, 98; 32, 34, 35, 38, 39; 33
and Bell Beaker 89–91; 39
Glauberg prince and 113
sacrifices 81
Stelae People 82–89, 96; 33
steppe, European 25–28, 103–04; 45
Stewart, Royal House of 209–10
Stewarts of Appin 204
Stonehenge 15; 3
as a healing shrine 53
Stonyford (Ireland) 155
Strabo 132–33, 134
surnames 27, 206–12
Irish 106, 168, 170, 171, 206, 207–09
Scottish 208, 209–12
with subclades of R1b-L21 27, 168
Welsh 168, 207
Switzerland 30, 51, 86
Helvetii 148
Sion necropolis complex 89–91, 97, 115; 35, 39
swords 40, 102–03, 104, 106–06, 112; 14, 44
Syracuse (Sicily) 126
Tacitus
on Caesar and Britain 154, 193
on Germani 9
precious metals in Britain 155
Roman colonial government 156–58
Táin Bó Cúailnge (Cattle Raid of Cooley) 17–18, 35, 36
Taliesin, The Book of Taliesin 14–15
Tapada Guarda, stela 38
Tara (Ireland) 18, 36, 166; 85
Tartessos 43–44, 52
Tasciovanus 18
tattoos 158, 169
Taurisci 130; 62
Taurisci 130; 62
Tectosagi 132
Theodosius 161
Thrace 77, 130, 131
Thracian language 31
Thracians, silverwork 16
tin 52, 98, 109
in Britain 155
bronze and 52
Indo-European migrants 98
Tincomaros 154
Tintignac (France) 40–42; 17
Tišice (Bohemia) 52
Tocharian language 76
Togidubnos 156
Togodumnos 155
tools
obsidian 70
pressure blade-making 69–71; 29
torcs 13, 117; 61, 63
Glauberg prince 112; 52
Hochdorf chieftain 108–09
Lady of Vix 110; 50
towns see oppida
trade routes 46
Adriatic to Danube 128
Golasecca culture and 113–15
Iberia and Greeks 43
migration and 125–26
Tartessos 43–44
treering dating 18–19, 20
Treveri 133
Triffyn (Dyfed royal line) 166–67
Trinovantes 153, 154; 74
Troy 22, 24, 25
Turek, Jan 93

Uí Echach Cobha 170, 171

Uí Liatháin 167;85
Uí Néill 28, 139, 173, 174; 85
Ulaid 15, 28, 138, 174, 183
Ulster 18, 138, 199
Plantation of Ulster 190, 191
see also Ireland; Northern Ireland Ulster Cycle 11–12, 15–16, 17, 18, 36, 138
chariot carvings 36
Umbrians 32, 201–02
UNESCO, survey of Celtic languages 187
United Kingdom
United Kingdom
pressure for devolution 199, 203
relations with EEC and EU 199
Ural Mountains 68
Uralic language family 68
Urien, king of Rheged 14–15
Urnfield culture 100, 102–04, 114, 136
Usatovo culture 76; 31
Ushant 46

Valentinian I, emperor 161

Vannes (France) 178; 92
Veneti 151, 178; 71
Venutios 155
Vercingetorix 40, 151–52; 73
Vespasian (Roman general) 154, 155
Victoria, Queen 196
Viking settlement, in Orkney and Shetland 189–90
Vila Nova de São Pedro (Portugal) 88; 37
Villas Viejas del Tamuja (Spain) 134
Virgil, Aeneid 24
Viriatus (Viriato) 147
Virica 154, 155
Vix, Lady of 109–11; 50
Vltava, River 117, 120
Volga River 71, 74
Vortipor 166, 167; 82
Votadini 13

wagons 34, 66, 68, 72, 75, 108, 110

burials 104–11, 117; 49
Wales
Anglo-Norman incursions 190–91
emigration of Welsh speakers 191–92
kingdoms 177–80
National Eisteddfod 197–98; 98
nationalist politics 199
Welsh Assembly 203
warfare 13–14
Warner, Richard 20
Water Newton Christian treasure 80
Watkins, Calvert 61
weapons
Hallstatt C iron sword 103, 105–06; 44
La Tène iron scabbard 62
Maikop culture 75
Remedello dagger 86; 35
 see also armour; arrows/arrowheads; bow; spearheads; swords Welsh language 31, 116, 187,
189; 93
eisteddfodau 196–97
health 189; 93
literary societies 196
Welsh surnames 168, 207
Wessex 177, 180
wheel 33, 34, 66, 72, 74, 75, 78, 84, 86, 118; 13, 14
Wilson, James 159
wine 105, 117, 124, 125–26; 59
woad 158
wrist-guards, archers’ 52, 53, 98; 21

Y Gododdin 12–14, 36
Y-DNA haplogroups
arrival in Europe 46
descendants of Brian Boru 209
in Ireland 170, 171, 209
MacFarlane R1b-L21 211
O’Brien R1b1a2a1a2c1f2a (L226/S168) 109
P, R & Q 70
R 26
R1a 78
R1b1a2a1a2 26
R1b1a2a1a2c (L21) 26
R1b in British Isles 99, 136, 139, 168, 208, 209, 210, 211
R1b-L21 27, 159, 168, 204, 208, 209, 210, 211; 9
R1b-L144 168
R1b-M222 139
R1b-P312 26, 57; 23
R1b-U152 136; 66
Stewart R1b1a2a1a2c1i1i (L744/S388) 210
see also DNA
Yamnaya (Pit Grave)
culture 68, 72–77, 80, 81, 82, 84, 89, 91, 92; 31

Zambujal (Portugal) 88–89, 93, 94, 96; 40; 37

Zavist (Czech Republic) 120, 123
Zincirli (Turkey), Kuttamuwa stela 82
 First published in the United Kingdom in 2015 as
Blood of the Celts

ISBN 978-0-500-05183-2
by Thames & Hudson Ltd, 181a High Holborn, London WC1V 7QX
and in the United States of America by
Thames & Hudson Inc., 500 Fifth Avenue, New York, New York 10110

Blood of the Celts © 2015 Thames & Hudson Ltd, London

This electronic version first published in 2015 by

Thames & Hudson Ltd, 181a High Holborn, London WC1V 7QX

This electronic version first published in 2015 in the United States of America by
Thames & Hudson Inc., 500 Fifth Avenue, New York, New York 10110.

To find out about all our publications, please visit

www.thamesandhudson.com
www.thamesandhudsonusa.com

All Rights Reserved. No part of this publication may be reproduced

or transmitted in any form or by any means, electronic or mechanical, including
photocopy, recording or any other information storage and retrieval system,
without prior permission in writing from the publisher.

ISBN 978-0-500-77295-9
ISBN for USA only 978-0-500-77296-6

Frontispiece: This statuette of a bearded and mustachioed Gallic warrior,

wearing a torc around his neck (c. 0 BC/AD), was found at St Maur-en-Chaussée,
Oise, France.