The New Image of Celticity
I am looking at the editorial of British Archaeology (Sept-October 2015), greeting a lavish prestige exhibition on The Celts at the British Museum in London: “It is with Celts, however, that academic and public versions of history seem to diverge more than any other time. […] many archaeologists say Celts never existed.” The editor hedges his bets: “There may never have been Celts, but there were Celtic fashions and crafts”[.] Classical writers never apply the word “Celt” to Britain. Calling the ancient British or Irish Celts is a misnomer. Those people actually existed even if we simply call them Western Indo-Europeans. Has a historical mythology of celticity been constructed? And what elements of actuality do we have left if we blow and cause the fragile and fumy mythology to dissolve?
Fortunately, a great deal of very astute and fundamental and compelling work has already been done on this problem. A range of writers have demolished the notion, developed in the late 19th and early 20th centuries, of the sociological link between Celtic languages and the Hallstatt and La Tene cultures, and between those societies and the British societies which, many centuries later, spoke languages known in a classificatory way as Celtic. The genetic link between barbarian aristocrats in Central Europe and similar in Ireland and Britain has been severed.
According to Almagro-Gorbea, the new image started around 1980 and is mainly the product of British archaeologists affected by theory, i.e. by scepticism about the stories validated and written by mid-century archaeologists. The disbelief in frequent and large-scale migrations is on important part of this. It would be wrong to leave out Raimund Karl - this is not merely a British phenomenon (and there has been no committed resistance by Continental archaeologists). The volume “The Celts in Theory”, edited by Karl and by David Spitzer, is a handy anthology of the critics of the historical myth of Celtdom.
The older theory involved four rings which it was necessary to have slide over each other and coincide. These were the historical narrative of Classical writers, the fragmentary records of Celtic language on the Continent, and the deposit pattern of artefact styles called Hallstatt and La Tene. The fourth ring was the antecedents of the British, since the theory held that there had been a cultural river flowing westwards from the Hallstatt centre in Austria, and that this had (somehow) pushed further West to engulf Britain - where languages of the family labelled “Celtic” were found being spoken, quite abundantly, by about 600 AD. The idea that these four things could fit over each other was a breakthrough at the time. It is associated particularly with Henri Hubert and TGE Powell, although others may have held this idea rather earlier. Hubert’s book was published in 1914 and Powell’s in 1958. The new wave has not yet succeeded in finding the first statement which links the “four rings”. It was a breakthrough, in a way. A recent and unmoderated statement of the theory is in Atlas of the Celts, by John Haywood (2001).
One aspect of this four ring theory is as prehistoric English imperialism. In order to supply the ancestors of the people of Britain, there had to be a coherent ethnographical narrative. As Celtic languages were so well evidenced in the British Isles, where they were still widely spoken in 1900, then the Celts had to come into existence to supply the back story for those ancestors. The coherence of the legendary about the Celts resembles, therefore, the firmness of the belief of Victorian scholars in their ancestry. A variant on this was the projective nationalism of French and Irish scholars, equally willing to find ancestors for their own ethnic groups - heroic, noble, and successful, if possible. Napoleon founded an Academie Celtique whose mandate was something like promoting the military prowess of the Celts as ancestors of the French, in order to project Napoleon’s Empire into prehistory.
The first ring: the Keltoi of classical historians
It may be helpful if we go back to an older book, H. J. Massingham’s Pre-Roman Britain (1927). It is a sixpenny book from Benn’s sixpenny library, a sort of predecessor of Penguins, and HJ’s qualifications to write it were no more than a fluency with words and an amateur interest in the deep past. He was probably recommended by one of his associates in the “Children of the Sun” school, for example W.J. Perry, an unlucky link as it may seem to us now that the school has been consigned to the outer realms of Deep Time crackpottery (or psychoceramics). He says of the arriving Celts: “these men were swift-moving warriors and they broke up the megalithic civilisation of Britain, as their kinsmen did that of Spain and Minoan Crete.” (p.72)... “The Celts were deficient in constructive works […] for the very simple reasons that their energies were diverted into a different channel. Their mission was primarily a destructive one”[.] “the Celtic crematory urn is much coarser and of more careless workmanship than the “Beaker” is, […] Their system of cultivation was mean and sparsely distributed in comparison with the patient labour of the terraces.” Apart from metalwork “in every other respect their civilisation betrays a relapse into barbarism [.]” He sees the society produced by the Celtic overrunning of the megalithic civilisation as “an intermingling of idea, custom and faith between a broken civilisation and its barbarian overlords. The Celts lost certain elements of megalithic culture and retained others in a garbled, traditional, and moribund form”. Massingham was affected by the Great War in some way I can’t reconstruct. He saw the Celts as the inventors of warfare: “So for as we can gather, organized warfare, as an absorbing occupation for men, did not seriously develop in Western Europe before the last millennium BC, when the waves of Celtic invasion and intertribal conflict crumbled the stability of the megalithic period.“ At some deep emotional level he identifies them with the “Central Powers” of the time - something boastful and malevolent that came out of central Europe.
An early appearance of the Celts (Keltoi) is in accounts of Alexander meeting, in around 330 BC, an army of them, across a river, north of Macedonia. The Greeks were likely to find a name for such a striking phenomenon as an army from central Europe appearing on their borders. But this kinetic, aggressive, and inherently unstable military entity did not straightforwardly match onto a society -something inherently stable, productive, and settled. The size of that army does not prove that its members came from one single tribe. The record of a Greek name for the mercenaries or invaders who poured down through the Balkans does not show that there was one people, of huge numbers, who were also the Celts. Still less, that there was one people which inhabited all Europe from Bohemia to Ireland.
Massingham was a spinner of robust fantasies rather than a thinker. It is touching nonetheless to see two ideas which he was attracted to. First, the idea that folk dances of modern times were faithful records of prehistoric events. Secondly, the association of the arrival of metals with the decline of cultural standards: something which we find among the hippy writers of the late 1960s, and which they had probably picked up directly from Massingham’s later books, still in print. Another idea is the correspondence of high cultures and peripheral ones, with the latter producing wretched and yet faithful imitations of the fabulous productions of the great cultural centres.
The relevance of Massingham (who had no background in research) is that he casts a side-light on the people who identified the Celts with the archaeological cultures of Hallstatt and La Tene: the evidence of that time allowed only wild conjectures. But also, that Massingham is right about the warrior aspect of what Classical historians record about the Celts: quite possibly the Celts were not a people but a phase, a state of excitement which some people enter into, forming armies and invading other countries, and the word “Celt” is not a people at all. The militarism is so prominent in all that the Classical writers record. This does not fit very well with an entire society. The people of Central Europe were an agricultural society - which does not exclude them being a militarised agricultural society. However, if there was a military identity which allowed mobile armies to form, including armies composed of individualist from any different ethnic groups, that identity might be what the word “Celt” (Keltoi, celtae) is a label for.
There is a classic analysis in Malcolm Chapman’s book about Celticity in which he describes the use of the word Keltoi, throughout Anna Comnena’s 12th century Byzantine history, to refer to the Franks, who were a pest from the west afflicting Byzantium in her time. You can take this as simply a mistake (the Franks could not be Celts) or as a loyal continuation of Greek semantics and social categorisation which had persisted, in a rigidly conservative and learned society, and exactly reflected what the word Keltoi had meant in 300 BC: a military entity, of Western origins, kinetically swarming, which overflowed proper territorial boundaries and was out to cause untold damage to Greek communities. This military activity was institutionally close to the profession of being a mercenary: as mercenaries, warriors, essentially not functioning in ethnic units, acquired the knowledge of geography and tactics which were the basis for being marauders and invaders. The predatory armies were a product of the surplus wealth of southern neighbours; the temptation of loot was able to unite mobile individuals from many tribes in one unit, a shared endeavour. Massingham’s strictures might, thus, not be so far short of the mark.
If the Celts were just a phase of the social reality of much larger and more permanent social identities, they could not be the ancestors of Insular societies. Moreover, most people speaking the Celtic languages were not susceptible of being warriors, being female, children, or of a social status incompatible with acquiring the panoply, the equipment of costly metal and so on, which defined the warrior. In this model, the amazing geographical extensiveness of the Celtic realms resolves itself as simply a limited and incurious view of Central and North-west Europe: because armies came from that general direction, called in a classificatory fashion Celts, it seemed as if everyone living in vaguely that direction was Celtic. In fact the armies or armed bands were linked by mutual interest rather than by ethnic affiliation.
Another interpretation of the Celts is as an impact zone of mediation between the Mediterranean and the barbaric hinterland. That is, their behaviour would depend on geography more than ethnicity. There is a certain relationship not just between mercenaries and invasion, but between mercenaries and the import of slaves from the periphery.
Massingham had looked at the Classical sources and at the warrior-display nature of the artefacts most
cited by archaeologists describing Hallstatt and La Tene. The artefacts were surely useful to a particular section of society, to a function (male, warrior, mobile, “excited”) rather than to all the elements which we know those societies must have consisted of. They were not artefacts relevant to sheep and cattle, to wheat and barley, to cooking and keeping house. It followed that their distribution was not proof of ethnic relationship, instead of aspiration to a heroic ideal which could be transmitted by warfare. It could reflect the acceptance of rivalry between different families, and the mutual imitation of enemies. The splendid artefacts deposited in the graveyard of Hallstatt, and in the lake used as offering site of La Tene, were specialised and their extension to a whole society with all its inhabitants was a leap which was brilliant at establishing connections between different provinces but could also cast real social and ethnic relations into obscurity. As soon as you establish that it characterised the nobility rather than the peasantry, you have exposed the possibility that the signs of nobility drew on emulation and display which could unite families from ethnic groups which were not related, and also that the idea of the long distance was perhaps basic to aristocratic success, in establishing alliances and in acquiring artefacts - probably also in physical movement. This physical movement is encoded in the preoccupation with horse gear and chariots. The symbols of aristocracy were related to a function, that indeed of moving over long distances. The long distances explain how similar artefacts could be found over such a wide area, from Bohemia to (eventually) Ireland. This is surely a different phenomenon from ethnicity.
Massingham sees celticity as something disembodied from ethnic reality. If the phenomenon of militarism has such a long history in Europe, a history of success, actually, then we would not want to say that all militarised groups were ethnically related. This is a basis for arguing that the warrior nature of specific groups in Ireland and in Celtic Scotland did not necessarily reflect linear transmission from the militarism of Iron Age Central Europe: they were different kinds of military ideology, even if both shared a preoccupation with courage and with metallurgy and arms technology.
Central Europe was inhabited by Central Europeans rather than by a discrete group called Celts. The art style is not an index of ethnicity but rather of competition and aspiration to prestige. Speaking of “the La Tene area” does not imply a central political authority or similar customs and laws, still less linguistic unity.
The Classical writers do not describe any Celtic state, still less a “Celtic empire”. There was no ethnic unity spanning wide areas of Iron Age Europe, this unity is just a blur caused by inaccurate observation by outsiders. The conditions of the period did not allow for large-scale political units.
The third ring: Celtic languages
The linguistic traces of a Celtic language family do not correlate with the La Tene or Hallstatt style areas and in fact there is no linguistic evidence of relevant date for most of those areas. The geography of Celtic languages before writing is wholly unclear. The evidence for Celtic languages in the Hallstatt area is remarkably thin. The basis for attributing Celticity to the Hallstatt geographical area was weak throughout the career of that theory. There was a total lack of evidence of appropriate date, and the evidence from a thousand years later did not show what people wished it to show. The exhibition catalogue (for Hallstatt) is noncommittal about the Hallstatt= Celts equation. The situation is different for La Tene, where we have quite good evidence for the linguistic Celticity of Northern Gaul, and northern Gaul is a key part of the geographical extent of the La Tene style. This is a momentous withdrawal. If the artefact style of Hallstatt gave rise to La Tene, and Hallstatt was not the artisanal work of speakers of a Celtic language, the basis for linking La Tene with Celticity is seriously weakened.
A major advance in 20th century archaeology was the hypothesis that an area with a coherent set of deposited artefacts also represented an ethnic group - as in 20th century Europe. This hypothesis is no longer accepted by the majority of archaeologists, certainly in Western Europe and North America. So beside the Hubert thesis that the La Tene artefact area has the same borders as those of an ethnic group, the Celts, we can place another thesis, that La Tene represents a horizon of imitation, where elites over an extraordinarily wide area see the prestige latent in certain very showy, parade, artefacts, preciously worked, and acquire imitations of themselves - by importing artisans, importing patterns, or importing them as finished goods. The idea of a sword can spread from one tribe to another and in fact the source of the transfer may be battles caused by ethnic rivalry. So the use of a swor does not prove who your grandparents were. The archaeology of those cultures was unfortunately focused on the showy artefacts, possibly available to very few people, rather than on subsistence patterns. This “horizon of imitation” idea itself needs testing, naturally.
The makers of La Tene artefacts may have spoken half a dozen different languages. Some of them certainly spoke Gaulish (P-Celtic). But that just shows that the ethnicity of those artefact styles is problematic. More probably they are exactly artefact styles and not the authentic and unfakable utterance of an ethnic ideology and self-image, which itself would have to be confined to speakers of one language, or perhaps three or four closely related ones. In fact, they were a variant on Etruscan artefacts, which themselves had strong influences from imported Greek gear.
If we give up the Celtic identity of Hallstatt, because there is no evidence for it, then we have to give up the binding link between La Tene and Celts, because La Tene clearly is the successor to Hallstatt. In this case, the westward impulse which carries the centre of La Tene to the west of the centre of Hallstatt is no longer able to sustain a “westward movement theory” whereby population groups would continue westwards and bring languages of Central European origin to the British Isles. This vector just stops working. The cultural impulses which came from central Europe to Western Europe in Iron Age times, mighty as they were, did not carry the Celtic languages. The evidence for Celtic being spoken in Central Europe is very thin. Reference to Celts in Greek sources describe a warrior life-style but have no component of linguistic evidence. The place-name evidence, late as it is, shows a concentration of celticity in Northern Gaul. Extensions into the east with Singidunum (Belgrade) and Carrodunum (Cracow) bearing the -dunum formative which everyone agrees to be Celtic and linked to well-attested nouns and verbs in Irish, are a very thin scatter and easily explicable as migrations from west to east after 300 BC. The artefact styles which moved from east to west are not equatable with linguistic Celticity.
La Tene and Hallstatt artefacts are rare in Spain. However, Celtiberian is definitely of the Celtic language family. This was made clear by Tovar’s work of 1946. The Celtiberians do not link to Central Europe. This is another reason why the language evidence does not convincingly link Britain to central Europe. The discovery of the Botorritos bronzes, that is significant inscriptions in Celtiberian, in the 1970s, has provided further ammunition for developing theories about Celtic origins, leaving the Central European theory high and dry.
If these languages did not flow, in some way, from central Europe to Gaul and then to Britain, their historical geography must follow some other pattern. One recent suggestion is “Celtic from the West”, starting in south-west Spain and spreading by a seaborne route along the Atlantic seaboard to the British Isles.
Someone wrote an essay on a word in a Celtiberian inscription piricantam. This is in a partly syllabic script so it is credible that the pronunciation was really brigantem. This word also appears in Gaulish inscriptions and in two Welsh words, brenhin and braint. The intervocalic -g- is lost so we have brenhin, which means king (-nt- further becoming -nh-) and braint which means privilege or right, something appertaining to a king. Indeed, copyright is hawlfraint.
This word also appears in the name of a tribe in Northern England, the Brigantes. The older translation of the tribal name as “the high ones” probably needs to be replaced by something like the ones endowed with authority or the lords. There was a variant brigantinos which is the source of brenhin and of an Old Breton word, breenhin (?).
The Iron Age was big on hill-forts. Some of these evolved into towns. In this context, the Celtiberian word for hill, briga, also came to mean “town“. So this word briga, which originally certainly meant something like “high, top” is related to the brigantes word - but has developed along a different route. I have to say, as well, that we don’t know what “piricantam” really means in Celtiberian - it looks like the words in Welsh etc. but its meaning as a social term expressing relations or distinctions in a Spanish society may have been distinctly different. The proliferation of the briga root gives us insight into the powers of this language, or cluster of languages, but it is also a reproof to us when we want to assign a meaning to a word in a language we hardly know. We recognise the root but without dense context we cannot say if that root meant in this vestment a hill or a legal privilege.
The fourth ring
This link Brigantem - brenhin – braint is a win. Quite clearly there are weak connections between Celtiberian, Gaulish, and Brittonic, and probably much stronger ones between Brittonic and Welsh. The area of “Celtic studies” does not immediately have to disband itself because the idea of Celticity has dissolved. There is an object of linguistic study. Forcing a separation between Continental Celtic and Insular Celtic is hardly tenable. However, the utterly fragmentary nature of the surviving ruins of Continental Celtic languages means that correlations with Insular Celtic can never be any more than fragments of those fragments. This is not the sort of structural material on which we can found large-scale historical theses. It is fascinating but is not for that reason useful. The advances in recovery and decipherment of Spanish inscriptions has brought a set of fundamental new possibilities to cleric linguistics, and it would be irrational to suggest that this new evidence has been assimilated yet. The Celtic nature of this new material is conceded by most scholars.
The arrival in the Atlantic islands of Celtic languages, ancestral to Irish and British, is undated but does not seem to have involved any large-scale invasion. There is no “interruption horizon” which shows the arrival in Britain of a new group we could label “the Celts”. In deference to this, the British Iron Age is now populated by “Iron Age Britons” and not by “Celts”. This does not preclude the arrival of new technology, in an adaptation or imitation process related to what has happened in Europe since 1300 AD. The wish for artefacts to speak a language, and to utter in that language signs of ethnic identity, is a projection of 19th C nationalism. We want those dead but exciting humans to be our ancestors, that is extensions of us. For some reason we regard humans not related to us as not being proper objects to project imaginary emotions onto. This is not rational, since all humans are related. The imposition of boundaries, of property, is itself an imaginary act which has no sensible basis.
The entities with their different boundaries would have to be substantially the same, that is that people in the area of La Tene artefacts would have to be principally Celtic-speaking, and conversely that the principal area where Celtic languages were spoken would have to be within the boundaries where La Tene artefacts were found in generous quantities. To fulfil these conditions, we would have to know where Celtic was spoken in the relevant period, i.e. from BC 1000 to BC 0 for the Hallstatt and La Tene cultures. The statements asserted further that the spread of the Celtic languages was mapped by the spread of Hallstatt and La Tene, so that the theory also gives us an area of origin for the languages and a geography of their diffusion through Western Europe. All of these points were subject to testing in the light of later evidence, and all of them have failed to be substantiated.
There is a language group which survives, naturally, the withdrawal of the label “Celtic”. Gaelic and Welsh are real entities and the rather shadowy entities of Gaulish and Celtiberian are unmistakably related to them, as a family of Western Indo-European languages, whatever we choose to call them.
The speakers of the Insular Celtic languages were the inhabitants of the British Isles and later British prehistory is the history of those people. Labelling them, also, “Celts” may be a source of errors, given how little we know about who the Celts were. Linguistically, the Celts may just have been the western Europeans. This really is not telling us very much about sociology. This leaves us with a puzzle about how Western IE languages (referred to as “Celtic”) came to the islands. Since they are found there by 100 AD at latest, they must have come there at some date. They must have come from the east because they are related to languages spoken in Greece and the Ukraine. We may have “cumulative indo-europeanicity” to complement Hawkes’ compelling idea of “cumulative Celticity”.
BIG people armies MOVING VERY FAST wearing GOLD winning MATCHES ancestors of ME people LIKE ME REBEL HERO ARISTOCRATS living at ease in CONQUERED ESTATES and leaving BURIED TREASURE and yet thousands of years ago being ME and belonging to WHAT'S MINE.
It is easy It's easy to see that narratives based on eye-grabbers like these have a popular appeal which more methodical evaluation of source material does not. Moreover, the primary attraction of archaeologists to the field of study, the infantile infantile component so to speak, may be of this greedy and starry-eyed nature, to be replaced in maturity by intellectual curiosity. Sadly, scholars are capable of nationalistic and other motives.
The relevance of this to poetry is that it displays the egocentric nature of identification, the psychological power of the in-group. When there is discussion about the reception of poetry by people from ethnic minorities, it is very plausible that not being in the in-group is a significant factor in the course of events, and that an idea of pure interest in aesthetic form is not the most powerful factor. If you watch people spending so much energy projecting their self or their family back to 1000 BC, it is hard to avoid the feeling that when they identify with poetry they are affected by narcissism and group solidarity.Addition. Belatedly I have found out that numismatists use the transcription birikantin for the word mentioned above. This let me find out that it appears as the only word on several coins, and that these were found in the Narbonne area of France. My recollection is that other inscriptions from this area, however few, were Iberian and not Celtiberian. So possibly this coin legend (in the Iberian syllabic script) is not Celtic at all. But I suspect that it is Celtic and corresponds to well-known place-names of the form brigantium (possibly Brigantion), ancestral to Bregenz, Briancon, and several others. I doubt it was just brigantin but coin legends habitually curtail words.
One comment – some Internet pages record the idea that “brenhin” (from brigantinos) originally meant “the Brigantian” and slid semantically to mean “lord” because the Brigantii were the top people. I find this problematic and I don’t think there is evidence that the Brigantii as a political formation had survived the Roman occupation. Since “briga” certainly means ”what is high”, and “high” is a common metaphor for people who wield social authority, it is simpler to guess that the word for “high person” meant “lord” and then “king”, and skip the Brigantii altogether. Their tribal name presumably meant “the lords”, a self-aggrandising epithet.
Since the root means “tip (of a growing plant)” in Welsh and Gaelic, it is likely that “tip, top” is the oldest meaning and that even the meaning “hill” is a secondary development. I have just been looking at an essay on folklore from Islay which gives a recipe for making tea from a local plant, saying that you boil it to extract the “briogh”, meaning “essence” (or flavour) in this case. Once you get really involved in the word-group around the root brig-, you will wish you’d never started. The point still stands that you can connect Celtiberian with Insular Celtic word-forms and celtoscepticism does not doubt these connections. The grouping of variants (so that you have briga, brigant, and then also brigantin-) found at both ends makes the link especially convincing.
Brig- is quite close to the Germanic version of the same root, i.e. berg, but we can keep them apart. Other languages may have had similar versions of the same root. An instance of that third language would be an exotic find.