 Who were the Celts? Oh, so many things to so many people. According to Julius Caesar, the Gauls called themselves Celts. So let's start with them. This is a Roman denarius, minted in 48 BC, showing Gallic warriors. How do we know their Gauls? Because this coin could be an illustration of the words of Deodorus Siculus, writing around that time, who said, The Gauls are very tall, with white skin and blonde hair, not only blonde 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 to 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. Some of them think so little of death that they fight wearing only a loincloth, without armour of any kind. Here's a modern interpretation. One big difference is that they're wearing a lot more clothes. That enabled the artist to depict the tartan still beloved by the Scots. But the charioteer has jettisoned his jacket to show off some swirling tattoos in the Celtic style. The other big difference is that the warrior has been domesticated. In the background are some Celtic roundhouses. The children are loving the show. Looks like a lap of honour for the winner of the trophy for fastest chariot carrying armour. So there we have the stock pitch deconstructed a bit. For Diodorus Siculus the Celts were very tall, but how tall was tall in the past? We see here the ups and downs of average height for men and women combined in three regions of Europe. Bones from 314 sites were measured for this study. One big surprise was that Mediterranean seemed to have been taller before the Roman Empire got a grip on them. There's this dip in the Roman period. It had been surmised that Imperial wealth would ensure better living conditions. Perhaps it did for some, but not for everyone. So around the time that Diodorus Siculus was writing the average Roman was about 164 centimetres tall which is about 5 feet 4 inches. At the same period the average for Central Western Europe was a mammoth 5 feet 6 inches. Well that seems big to me to fit into this and while you may well ask on the face of it the map is telling us that Ireland is part of northeastern Europe. I suspect that what the author is trying to tell us is that there were no samples from Ireland. The important point is that the Celts did not need to be giants to seem very tall to Mediterranean's. Indeed we can see that there was a pretty consistent height advantage for northeuropeans over the centuries. Perhaps it was the result of ample milk in the northern diet. Whatever the reason height difference may explain the name Celts. There have been a number of attempts to find a meaning for it in a Celtic language but what makes most sense to me is one linguist's idea that Celtoi the Greek version had a Greek meaning the tall ones. Certainly it was recorded first by Greek writers. For the Greeks the world was a lot smaller than it is for us with our globes and atlases and satellite views. The Greeks mainly explored and colonised the Mediterranean and Black Sea. In the 5th century BC Herodotus had heard of Celtoi beyond the Pillars of Hercules which we call the Strait of Gibraltar and he knew that the river Ister the Danube to us rose in the land of the Celts. But his geography seems to preserve to modern readers. He thought that the Ister rose at the city of Pyrene somewhere in Iberia. Aristotle later made it clear that Pyrene here was intended to be the Pyrenees. Now 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 and that could lead to a mental map of one region being stitched directly onto that of another without realising 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 horses until the 3rd century BC but Herodotus does list the tributaries of the Danube from east to west. The last one of which he names is the Alps rising north of the land of the Ombrians which must mean the Ombrians of Italy. So here it seems we have one of the Alpine tributaries so we can work out that some Celts were living north of the Alps around the head of the Danube. To a modern map we can see where ancient Greek traders and colonists encountered the people they called Keltoi. Herodotus knew of some outside the Strait of Gibraltar next to the western most people of Europe. Now that fits a description in the 4th century BC. A very prosperous market called Tartesos a famous city with much tin carried by river as well as gold and copper from Celtic lands. One king of Tartesos was called Argantoneus and this name is Celtic derived from the word Santa meaning silver. Then we have a complex source the Oremaritima. Its Roman author describes the coast from Brittany to Marseille. Not in his own day but drawing on Phoenician and Greek authors from the 6th and 5th centuries BC. He says that Ligurians had been displaced by Celts somewhere on route but it seems it's very unclear where and it's not much help. More interesting for us from the Isles is this section. 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 Hyone and nearby lies the land of the Albionis. Sacred Isle is a misconstruction of the name of island. In ancient Greek Hyeros meant sacred. The Hyone are easily identified as the Irish. The Irish name for themselves can be reconstructed as Iverni from the Irish name for island Iveria the fertile land. Albion corresponds to the old Welsh Elbid meaning world, earth, land, country, district something like that. Further east we have references to Keltoy in what is now the south of France. A Greek colony was founded around 600 BC on the Mediterranean coast at Masalia modern Barçae by treaty with the Ligurians whose land it was. But Masalia was said to be near Keltiki the land of the Celts who had their own port at Narbo presumably present day Narbonne and of course we have Keltoy around the head of the Dutch. Such scattered references might suggest that the early Greeks used the term Celts for any foreigners north of the Mediterranean. But if we zoom in on Europe as known to Herodotus we find a whole lot of different foreigners starting at the west we have Sinesians we have the Tartesians Iberias the Ligis or Ligurians the people of Italy Illyria and Thrace then further north the Getai and various other peoples encountered probably by traders along the Danube. So Keltoy meant something more specific than random foreigner. On the other hand the Keltoy were far too widespread to be a single tribe. Indeed when Roman geographers went to work centuries later they recorded dozens of tribes in Gaul alone. Just to confuse matters they weren't all Gauls. Caesar famously said that Gaul was divided into three parts. The people in each part were distinguished by language customs and laws from the others. Aquitanni lived in southwest Gaul and spoke a language that we can now identify as a precursor of Basque. The Belgi lived in northeast Gaul and were relatively recent arrivals from across the Rhine. The Gali or Gauls to us lived in the rest. So the terms Gali, Belgi and Aquitanni were one step up from tribal. They denoted an ethnicity. Now that's not the same as a nation far from it. Tribes could fight each other shouting insults at each other in the very same language. There was no long term central political controlling Gaul until Caesar brought it all into the Roman Empire. The takeaway message is that language was important to the ancients in defining identity. Naturally they had no such thing as a linguist in modern terms. Any trading nation needs to communicate and the Romans had translators. So they were not merely aware that there were people in Europe who did not speak Latin. They knew that Greek was different from Celtic and so on. Not that they were invariably spot on in modern terms. Caesar thought that the Belgi spoke a language different from the Gauls. Modern linguists perceived a little difference going by place and personal names. So maybe it was more of a dialect difference. When modern linguists were let loose on the data, they recognized that Gaulish belonged in a family of languages. Welsh scholar Edward Lloyd, keeper of the Ashmolean Museum in Oxford, was deeply interested in the living languages related to Gaulish. In the absence of published dictionaries for them all, he traveled through Cornwall, Scotland, Isles and Brittany, taking notes. Sorry, Ireland and Brittany. It was quite an adventure without modern transport. In 1707 his archaeology of Britannica appeared. It provided the first detailed account of the living Celtic languages, exploring their affinity with each other and with Gaulish. The concept of the language family was thus set on a solid footing. The label Celtic came later, but it was an obvious one to look to use. This was the name that transcended tribal and geographical boundaries. Gradually linguists have added other languages to the family, such as Celtiberian, once spoken in Iberia. Some linguists argue that the language of a group of inscriptions in southern Portugal is Celtic. Without the case, it would be the earliest Celtic language recorded in writing for the inscription star in the 8th century BC. But until and unless linguists can agree on this, the earliest written Celtic language remains La Pontic, spoken in northern Italy with inscriptions from about 600 BC. Putting together all the linguistic evidence, inscriptions, place names and living languages, this is what we get. Celtic, ancient and modern. Today Celtic languages cling to precarious life on the northwest fringes of Europe. Delving to the pre-Roman past, and we find Celtic spoken across the continent and even into Anatolia. Now this is a sort of map you'll find in atlases of the Celtic world. This one is actually based on a map in the superb historical encyclopedia of Celtic culture edited by Celticist Professor Dong Cook. For this and his other working Celtic studies, he has adopted a working definition of a Celtic as someone who spoke a Celtic language. I've done the same for my book, but not everyone would agree. As Cook says, the concept of the Celts we have today entwines three categories of evidence. One, people called Celtoi by the Greeks. Two, the ancient Celtic languages. And three, the Hausstadt and Latin archaeological cultures. In the 1990s some archaeologists began to argue that only number one on the list really counted. Since there was no evidence that Greek or Roman authors had ever referred to people in Britain or Ireland as Celts, or indeed that early British or Irish authors dubbed their people Celts, they therefore could not be Celts. Now this missed the point by a mile. Celtic is our modern word for a widespread people linked by language. We borrowed it from the Greeks. Just as today we refer to the inhabitants of Polynesia and their family of languages as Polynesian. Now that word is from the Greek from many islands. It wasn't named by ancient Greeks of course. They didn't get that far. It's the creation of later European explorers. But we don't agonise over whether the Polynesians ever used that name for themselves and state that if they didn't they can't be Polynesian. But moving to number three on the list, how did these cultures come to be linked to the Celts? That takes us back to number one. People called Celtoi by the Greeks. In the mid 19th century the scholarly focus fell on those around the head of the Danube. As archaeology began to take shape as a discipline amateur pioneers in the field identified two successive cultures at the right place and time to correspond to these historic Celts. Johann Jörg Ramsauer was in charge of the salt mines at Hausstadt in the Austrian Alps in 1846 when he discovered a cemetery nearby. Instead of simply plundering it as antiquity hunters of his day were going 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. Scraps of cloth showed the plaid type of weave that we associate today with the Scots. So this early foray into archaeological methodology remains one of the most remarkable sites ever linked to the Celts. He got the full television treatment from the BBC on Monday night in the first programme of the series The Celts, Blood, Iron and Sacrifice. We now know that the site belongs mainly to the seventh and sixth centuries BC. When similar artefacts were found elsewhere they were naturally identified as in the style of Hausstadt. So the site gave its name to a culture eventually found to be widespread over Central Europe. If you watched the programme on Monday night Neil Oliver made the classic error of saying that the culture spread from the small village of Hausstadt. This is the problem of calling a culture after the first place it happens to be discovered. The site at Hausstadt was just one of many in a wide intercommunicating zone growing wealthy on trade. Salt was just one of the items traded. The site at La Taine on Lake Neuschatter in Switzerland should be forever associated with the next phase of the Central European Iron Age. In 1857 local collector Hansley Cop 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 artefacts. The flowing forms of the decorative metalwork attracted most attention. Here we see stylised animals enlivening a scabbard in a confidently fluid and playful style. But like the salt of Hausstadt the oxygen-free lake mud also preserved wooden objects including a complete chariot wheel. The chariot was to prove a key distinction between the earlier Hausstadt culture and La Taine. Lightly objects in the same style were dug up in France and northern Italy. The Italian finds were crucial. The Roman historian Livy describes the Gauls crossing the Alps in waves to settle in the Po Valley, driving out those Etruscans who lived there. So the cemetery of an intrusive people in the ruins of an Etruscan town could be firmly identified as Celtic. The Latin culture thus could be seen as the material manifestation of the Celts. So neat and pleasing was this identification that it stiffened into certainty that no other culture could be Celtic. Making it a huge puzzle the Celtic speech elsewhere did not always seem connected to it. Remember our language map? Celtic languages down the Danube and across into Anatolia can be explained by recorded events in the last few centuries BC as you just saw. La Taine artifacts went along with the language. But what about Iberia? That was barely touched by La Taine influence and yet we have evidence of Celtic speakers there. Indeed we have a king called Argentonius who died before the development of La Taine. Indeed we have actual inscriptions in a Celtic language in northern Italy before La Taine. Not that La Pontic is a problem. It's generally accepted that the La Taine culture developed out of the Halschlok culture which is assumed to be Celtic speaking. It's Iberia an island that are the real problem. In the Aura Maritima we have Celtic names for island and its people that seem to date from two or three centuries before La Taine material appeared there. Awareness of these flaws in the concept of Celtic equals La Taine has generated new interest in an alternative model in which the Belbica culture was the vector for the earliest forms of Celtic. Belbica leapt across Europe not settling everywhere but picking certain patches often ones with ores to exploit. It's 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 to Belbica for the origins of the Celts. Support for this model is starting to arrive from ancient DNA. So far we have wide DNA from Belbica sites only in Germany but every one of them has turned out to be haplogroup R1B. R1B dominates Western Europe today and there's significant overlap with the region's formerly Celtic speaking. Just a few years ago it was firmly believed that this modern pattern derived from the days of hunter-gatherers. Ancient DNA has put paid to this idea. The earliest R1B in Europe appears on the European steppe, long thought to be the homeland of the whole family of Indo-European languages to which Celtic belongs. It arrives from the Asian steppe just in time for one branch to take up farming and move with farmers in a minor way but the major line of R1B spread over Europe in the Copper Age. The subclad of R1B that dominates Western Europe today is the one we call P312 for short. That's been found in one Belbica male as you see at the bottom of the table. One sign of a population explosion is a starburst of subclads appearing from one node on a tree. Increasing population means more suns arriving, more daughters too, but I'm focusing on YDNA because it's most informative about the Copper Age developments. P312 had a burst of activity that produced three major subclads. Such population growth is compatible with migration spreading people into new territories. The L21 subclad predominates today in Britain and Ireland, particularly in the regions where Celtic languages survive the longest. We can easily guess that before the arrival of the Anglo-Saxons, L21 was dense all over the Isles. Confirmation comes from a site in eastern England. Two pre-Roman males, buried at what is now Hinxton in Cambridgeshire, proved to carry YDNA L21. The people who carried YDNA R1B and its cousin R1A also brought a genetic signature new to Europe in their full genomes. It's been dubbed Ancient Northern Eurasian or ANE for short. After some large recent studies of ancient DNA, we now know that the European gene pool is composed of three main elements. The Western hunter-gatherers who were in Europe first, the early European farmers who spread in from the Near East, and this last group who spread in in the Copper Age, whom we can logically link to the spread of Indo-European languages. As Irish historian Owen McDeele sagely said in 1920, there's 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. He felt that all the present nations of Europe are a mixture of the same ancestral components in varied proportions. A perceptive man, he's been proved right. So ancient DNA is giving us a picture of massive population change arising from upheavals in the Copper Age, but we need not picture vast armies charging across Europe. The Bell beaker people seemed to have arrived initially in small groups scouting for metal. They had the greatest scope for settlement and eventual dominance in regions where farming had started to struggle, which included Ireland. They were the first metal workers to enter the Isles, homing in on the copper belts of Ireland and Wales. Around 2400 BC, they left their characteristic occurs at a copper mine on Ross Island in County 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. Analysis of chemical composition shows that copper from Ireland was traded into Britain. The gold of the Morne Mountains was another attraction. Gold was prized by the Bell beaker elite. It's too soft for practical uses, but it's eternal glitter is ideal for ornament. The earliest gold object found in Ireland arrived with Bell beaker people. It was the ear pendant on the left there. It was not of Irish gold, so it was imported ready-made. Similar objects have been found in Portugal, so it seems likely that some Bell beaker people arrived in Ireland by the Atlantic route. In 2009, a gold lunala and a pair of gold discs hit the headlines. It's an extraordinary story. They'd been discovered in Pete in 1945 at Kollerbeg, but they were lost to scholars until they were found among the contents of a safe stolen from a pharmacy in 2009. They're now in the National Museum of Ireland. Both Lunalae and discs had been found before in Ireland, but not together. In fact, the only other such set discovered is in Portugal. However, Lunalae and discs are found in other parts of the Bell beaker world and fit into a pattern of solar imagery in rock art and other objects in copper and bronze age Europe. Ireland seems to have been the main centre of production of Lunalae. It can boast over 80 out of the hundred or so recorded. In archaeology Ireland earlier this year, Mary Cahill argued for the lunala as a symbolic skyboat carrying the sun in its journey. Whatever these objects symbolised, it's clear that their meaning was widely understood. This was a period in which shared concepts and values were spread along long distance networks. Now at this point those keen on the Celtic from the West theory will be feeling joyful and triumphant. Their idea is that Celtic entered Ireland up the Atlantic route, but here we have strange objects telling a different story. The pair on the left are from Ireland. They've been beaten flat, but when found they would have been rolled up like those on the right. They appear in pairs so some archaeologists think that they're earrings. Others believe that they fixed the end of braids in place. I'm in the latter camp, but the importance of the pair on the right is that they were buried with the Amesbury archer and we know where he came from. Isotopes in his bones and teeth show that he came from somewhere near the Alps. So Belbica flowed into the isles by two routes and the eastern group of Belbica seems to have gained in prominence as time went by. Significantly the distribution of Belbica is not exclusive to later Celtic regions. It includes areas that became italic speaking. Linguists agree that italic languages such as Latin had similarities with the Celtic family. The two families could even have a common ancestor as I showed on the Celtic language tree. As I mentioned both families belong to the yet bigger family known as Indo-European. Languages don't spring from nothing. They have parents and grandparents so it's useful to see them in tree form even though I have to add these fat arrows all over the place to explain how languages can continue to interact and influence each other even when they've gone their separate ways. The process of language spread can be anything but straightforward. Many parts of Europe seem to be like a linguistic layer cake. One wave of Indo-European speakers was succeeded by another. And here's my latest guesswork. Place names that can be identified as Indo-European but neither Celtic nor italic are found in Iberia particularly in the south. There are many place names starting with the letter P which was lost in early Celtic. Place name scholars generally cast these as old European by which they mean the earliest Indo-European. We even have some of these in Britain and Ireland. Now don't ask me to tell you what they are of the job of my head. Since the Indo-European seem to be metalworkers and the first metalworkers in Ireland of Belbica that leads me to suppose that the earliest Belbica people in Ireland spoke old European. Meanwhile back in the Carpathian Basin or somewhere thereabouts we can picture the language gradually changing into something not quite italic. That has a three-way split as different groups go their different ways. Some head north to that famous area around the head of the Danube, some head into Italy where we can picture italic developing and some were coastal wanderers. I picture Celtic developing where traditionally thought but earlier than Halstead. Somewhere in late Belbica times. That location is a good fit to the linguistic evidence that Celtic developed in contact with precursors to Germanic. Now I'm not even going to give you a generic battle scene. Television does these things so much better. All I'll say is that this is what Europe looked like when Rome had finished with the Celts. I expect some of you have read the asterisks and obelisk comic books all about a little village in Ormorica, now Brittany, where the Gauls held out supposedly against the Romans with the aid of a druid and magic potion. It's enchantingly funny but the real holdouts were the Irish and the northern British. Celtic languages survived on the northwest frontier of the Roman world. While Romanization gradually straight-jacketed Celtic art over much of Europe, the shape-shifting Latin style continued to develop freely in Ireland into the Christian era, absorbing other influences as it went. I make no apology for showing such familiar images, they are justly celebrated. Now here's Google trying to get to grips with the different characters of modern Celtic nations. Two identical cultural capsules placed in two different locations will immediately start to accumulate different experiences. The Irish, for example, are an island people lapped around by the sea. Google has here illustrated the famous tale of the children of Lear, Lear being a personification of the sea. The modern Bretons, Cornish, Irish, Manx, Scots and Welsh cherish their own particular identities and I've been attacked by an Irishman for daring to push the Irish under the Celtic umbrella. To all such doubters I have only this to say, explain that without resort to words like bards or musical heritage. The word Celtic is trembling on your lips, right? So there we are. Seek within for more.