Conclusions

Archaeology and Celtic art

To the question 'can we meaningfully talk about Celtic art?', we have answered in the affirmative, but not simply as careless shorthand for La Tene art. It is possible to argue from the documentary record of ancient historians and geographers that there were people known to the classical world as Celts, and even people who regarded themselves as Celts. Furthermore, the evidence of personal and place-names, admittedly in many cases known from Roman period sources, allows us to infer, in the absence of any evidence for ethnic incursions to account for their recent introduction at that late date, that the Iron Age populations of Western and west Central Europe spoke related languages that are classified as Celtic. A case can therefore be made for regarding the art of any of these population groups, whether identified as Celtic on ethnic or linguistic grounds, as Celtic art. Plainly this must include La Tene art: indeed, the magnificence of the La Tene art phenomenon justifies giving it pride of place in any treatment of the subject. The fact that the archaeological distribution of La Tene culture is not coterminous with those areas recognized as Celtic ethnically or linguistically need excite no particular difficulty. Not since the generation of Gordon Childe have archaeologists expected such close correlations for many reasons, not least that the dynamic process of cultural interaction and change would have obscured geographical territories and boundaries over time. But it is also abundantly clear that Atlantic Europe had a different character archaeologically from the Urnfield-Hallstatt-La Tene sequence of Central Europe, and without presuming an Atlantic unity that may be more apparent than real, the Celtic zone evidently included a non-La Tene as well as a La Tene aspect.

Celtic art therefore is emphatically not exclusively La Tene. In fact, the persistence of that equation has greatly hindered an understanding of Celtic art in its broader geographical and chronological context. Conversely, the exclusive equation of La Tene material culture with people of Celtic speech or ethnicity is a more difficult issue. Setting aside obvious examples of prestige goods that were carried by trade, diplomatic gift, plunder or spoils of war well beyond the La Tene cultural world, there is a hinterland between the north-alpine zone and the North European Plain where, from at least the Augustan era into the first century ad, it is unclear how far Celtic and northern, Germanic-speaking groups may have integrated.

The La Tene-Celtic equation has also dominated assessments of later Insular Iron Age art. In addressing the genesis of symbol stone art in eastern Scotland ('Pictish' art), students are drawn to question whether it represents a late resurgence of Celtic art, by which they mean implicitly La Tene art, and if so, what had happened to the tradition in the intervening centuries? The answer must be that it is not a late resurgence of La Tene art, which is not to say that it is not another manifestation of Celtic art. It might include echoes of major themes from earlier La Tene art, but it is also different in important aspects, such as the depiction of the human form in battle or hunting scenes, in which it also echoes Celtiberian art. Likewise the art of Early Christian Ireland may be regarded as Celtic art, but its debt to the La Tene tradition is perhaps less significant than that to Germanic and other traditions. Once we break the exclusive equation between Celtic art and La Tene art, it is possible to view the regional and chronological differences across the Celtic world of later prehistory and early history in a different light. Instead of trying to trace influences of one upon the other, or trying to derive one from the other, on the basis of perceived similarities in style or motifs, sometimes valid and sometimes tenuous, we can ask why certain periods in certain regions produced outstanding art works while others did not. What circumstances or combination of circumstances catalyzed the Golden Ages of Irish art or early La Tene art in Europe?

The traditional approach to Celtic art through a chronological sequence of 'styles' fails to address key issues of social and cognitive archaeology implicit in the evidence. The different styles in any event are not strictly comparable, being manifest in different sets of material representative of limited social groups - aristocratic drinking services, high-status warrior equipment, or high denomination coinage — and at no stage represent more than a limited spectrum of society. The art of the potter seldom reflects the art of the metal-worker, but cannot be simply dismissed simply as vernacular art since there are reasonable grounds for believing, for example, that some ceramics had a particular funerary role, perhaps quite different from their domestic counterparts. Celtic art has many regional sub-sets, and few pan-European traits, other than in the most general terms. Dragon-pairs represent one of the few images that have a widespread geographical distribution, but they are still only representative of a restricted if influential warrior elite. What was the status of the artists, and how did specialist armourers, jewellers, moneyers, professional brooch-makers and gold-workers operate over a millennium or more? Why did Celtic art develop so spectacularly within the La Tene world rather than in the non-La Tene Atlantic north and west? And why did it enjoy a spectacular revival in the early historic period, notably in Early Christian Ireland?

Our answer to these questions will depend in some measure upon our definition of art. The supremacy of the La Tene art style in archaeological convention derives fundamentally from the equation of art with ornamental styles embellishing high-status artifacts that display intrinsic value or high technical accomplishment. Though less prestigious, a carefully crafted brooch or spear-head, completely devoid of superficial ornament, but deposited in a grave assemblage or in a ritual hoard, might nevertheless qualify in an anthropological definition as a significant art object. The Maskfibeln of the early La Tene period in Central Europe or jinete of Celtiberian Spain certainly qualify as art objects, even though they bear no ornamental embellishment, beyond stylized human or animal imagery, that might assign them to any particular 'style'. Equally, however, wooden images that will have perished, unless preserved by unusual environmental conditions, or paintings on the daub walls of timber-and-wattle buildings, could have served the social and ritual needs that art evidently fulfilled in Celtic society.

First, we should try to draw together the various strands of information that have been studied in the foregoing chapters, to see whether it is possible to distil a provisional re-assessment of what constitutes Celtic art, based upon the material inventory across those regions that we have identified as probably or potentially Celtic on ethnic or linguistic grounds. Among the pre-Roman Celtic groups there are certain recurrent themes or component elements, and certainly recurrent artefact types, in the inventory. Broadly following and developing Jacobsthal's analysis, we may identify four elements: (1) anthropomorphic imagery; (2) animal imagery; (3) abstract curvilinear ornament; and (4) geometric ornament as components of early Celtic art. These elements are obviously not unique to Celtic art, and might equally be a description of other ancient or more recent ethnographic art styles. Nevertheless, they are distinctive and characteristic, and we may therefore begin with a brief review of these four aspects of early Celtic art:

1 Anthropomorphic imagery. Jacobsthal underlined the absence of a figural or narrative dimension in La Tene art, in spite of the model of figural scenes on Greek vases or the proximity of situla art around the head of the Adriatic with its ceremonial and festive scenes, so that the rejection of figural or narrative themes was clearly by deliberate choice or inclination. The scene depicting riders and foot soldiers on the La Tene A scabbard from grave 994 at Hallstatt, commonly cited as an illustration of Celtic warriors in battle, is so obviously influenced by the frieze narratives of situla art as to be scarcely representative of Celtic art. Equally the contesting warriors on later Iron Age painted vases from Numantia are part of a regional tradition that is already much influenced by southern, Iberian styles. Schematic figural scenes are represented from the eastern Hallstatt zone, on the pottery from Sopron, and simpler 'matchstick' images are occasionally found elsewhere across Celtic Europe. In the north-alpine Hallstatt world, life-sized stone sculptural figures, notably from Hirschlanden, from the early La Tene at Glauberg, or in the Castro culture of northern Portugal, may be representations of deities or ancestor-heroes. In fact, much of the human or humanoid representation in Celtic art is sufficiently stylized to be regarded as formulaic or representative of otherworld beings, rather than figural in the normally-accepted sense. Green (1989, 206ff) has argued that images of deities may be exaggerated or schematized in order to dehumanize the portrayal, in effect deliberately to distance the supernatural from the natural. This may manifest itself in an enlargement of the head, in bi-cephalic or tri-cephalic heads, or the truncation or schematic depiction of other limbs. Even in later symbol stone art or Early Christian manuscript art the human form is depicted in schematized form or in postures that are improbably contorted, so that it need not be concluded that all such images are intended to be supernatural or divine.

2 Zoomorphic imagery is certainly not unique to Celtic art, though it is a significant element from the Urnfield period through to the symbol stone art of the later Iron Age in Scotland. At one level, it constitutes a pan-European theme in Celtic art, and is one theme in which La Tene art was certainly not lacking antecedents. It could take the form of animal images modelled in the round or representations in two dimensions on pottery or metal-work, and originally doubtless was also deployed on perishable organic materials or as painted images that would not survive. Animal and bird images may assume several guises. Some are of recognizable breeds, like boars or birds of prey, either complete at various scales from miniatures to life-size, like the Hounslow or Neuvy boars, or as animal-headed protomes like the Bra cauldron bovines. Alternatively, they may mutate into exotic or fantastic beasts, like those on the rim of the Durrnberg flagon or the handle of the Borsch flagon. Fabulous creatures like centaurs and winged horses on coins are adopted into the Celtic pantheon from classical sources. Finally, zoomorphic imagery may be introduced as implicit elements in more complex designs, as in the Cheshire style or in lyre-motifs that become dragon-pairs in La Tene art.

It is worth remarking that most of the creatures represented — boar, bull, horse, stag, snake or fish — are conventionally regarded as 'noble' beasts, and it is hard not to believe that animal and bird imagery had a special significance for the Celts. Doubtless hunting was a highly regarded aristocratic pursuit, illustrated by the model of the Celtiberian boar hunt or by hunting scenes of 'Pictish' symbol stones. Animal or bird imagery was also evidently closely associated with the arts of war, as the ^iumejti helmet or Witham shield testify. Even so, imagery like the serpent held by the squatting figure on the Gundestrup cauldron or the snakes of symbol stone art are suggestive of a symbolic or ritual dimension in animal imagery. In the Romano-Celtic pantheon Epona, Artio and Arduinna are identified with horse, bear and boar, and even in an Early Christian context lion and eagle are symbolic of the Evangelists.

3 Abstract curvilinear ornament is commonly regarded as the essence of Celtic art, simply because of the mistaken equation between Celtic art and La Tene art and because of its prominence in the latter style. In fact, not all La Tene styles, as we have seen, are curvilinear and among those that are, repetitive designs often underlie ornament that at first sight is characterized by freestyle complexity. The Waldalgesheim or Vegetal Style is often regarded as the high point of La Tene curvilinear ornament because it is in this style that derivative classical motifs are deconstructed and re-emerge with the 'assured irrationality' (Megaw, 1970a, 89) of a vibrant and independent Celtic creation. In fact, the culmination of the La Tene tradition might better be attributed to the later Plastic Style, in which the art is not simply applied as ornament but is the artefact itself fully integrated with its embellishment.

4 Geometric ornament. Geometric ornament, rectilinear, curvilinear and emblematic, occurs widely in Celtic art, separately or in combination. Emblematic here refers to the use of symbols like the 'sun-disc' or perhaps the swastika that may be supposed to have had a 'meaning' beyond simple ornament. Geometric ornament is especially characteristic of Urnfield and Hallstatt Iron Age pottery and metal-work, but continues throughout the La Tene period, as in the arc-and-circle style in the eastern La Tene zone or on the Braubach style of pottery, though it is otherwise seldom accorded a diagnostic role in schemes of classification. In the Celtiberian world, geometric, including emblematic, ornament is the norm in pottery and metal-work. Geometric ornament is widespread prior to the appearance of the La Tene curvilinear styles, but should not be regarded as necessarily earlier in any formal sequence from 'geometric' to 'curvilinear'. In fact, the two should not be seen as opposites, since geometric and even compass-aided designs, as we have seen, are sometimes implicit in curvilinear composition.

On the principle that the object is as significant as or more significant than the ornament, we should perhaps first consider the predominant types that constitute the body of Celtic art, earlier and later, since these should be indicative of the social role and function of art objects in the Celtic world. We need not assume, of course, that priorities will remain constant over space or time. In fact, since beliefs and values may have changed with developing social and political structures, we might anticipate significant changes over the period of nearly two thousand years between the Urnfield late Bronze Age and the later Iron Age. For the present we may begin with an inventory of the principal classes or groups of artefacts that have been discussed in the context of Celtic art.

• weaponry, especially sword, spear and shield

• equestrian gear and vehicles

• personal ornaments, both everyday and high-status

• drinking and feasting equipment

• public art: sculpture, stele and related monuments

• manuscript illumination

• pottery and domestic equipment.

Ethnographic studies suggest that works of art conferred prestige upon those who 'owned' them or had custody of them. In the case of early Celtic art, and in the context of a chiefdom-based society rather than an urbanized state society, we might infer that prestige accrued from controlling the resources for production in the case of locally produced artefacts or the negotiated means of access in the case of imported goods. It has commonly be assumed that southern imports, initially Greek and subsequently Etruscan, in the Hallstatt C-D phases and into La Tene A and B, sustained a prestige goods economy among the north-alpine aristocratic dynasties (Frankenstein and Rowlands, 1978), though the evidence for redistribution has been challenged (Dietler, 1990). Indeed, it may have been the failure to sustain the mutual obligations implied by the redistribution model that strained the social fabric of the late West Hallstatt chiefdoms to the point of collapse. The status conferred by Mediterranean contacts would equally have been visible in the mud-brick walls and bastions at the Heuneburg or in the novel treatment of Greek and orientalizing motifs in early La Tene art. In the hierarchical structure implied by the late Hallstatt Fürstengräber it may be supposed that prestige was acquired through the capacity to dispose of the wealth represented by 'art objects' in burials or hoards. Since disposal in hoards or high-status burials represents a primary means whereby distinctive 'art objects' enter the archaeological record, it would be easy to assume that this represented some kind of norm, whereas self-evidently the rich late Hallstatt tombs of south-west Germany or the early La Tene chariot-burials of the Champagne are very much in the minority, even in the well-documented archaeological record of these regions at those times. In all probability the majority of 'art objects' were curated and inherited over many generations, resulting in some of the composite artefacts that we have here examined.

From the eight principal categories listed above it is clear that not every category is equally represented in all periods or in all regions covered by this survey. The late west

Hallstatt Furstengraber are especially distinguished by their wealth in terms of the drinking and dining service, but are not noted for warrior equipment. Weapons are certainly represented among the combination of grave-goods (Kossack, 1959) but not especially in the princely tombs. Ethnographic evidence (Keeley, 1996, 144) suggests that the warrior class, vital in war, was not necessarily accorded the highest rank in peace, when the skills of negotiation, wealth acquisition and knowledge of ritual may have been accorded priority. Where weapons are included in the wealthiest grave inventories, as at Hochdorf, the dagger and its sheath are particularly splendid, underlining the fact that this is as much a symbol of power as a weapon of combat. Such a scenario would not be incompatible with the evidence from the late west Hallstatt world, where the 'paramount chieftains' (Frankenstein and Rowlands, 1978) would have vied for control of the lucrative Mediterranean markets in order to sustain control over their 'vassal chiefs', and yet would doubtless have required periodically the force of arms to bolster their status and resolve disputes. Trade and exchange, as Keeley observed (1996, 125) are no more a guarantee of peace than inter-marriage, and social tensions created by contacts with the Mediterranean world are likely to have been as significant as their material benefits. That the hierarchical structure of late west Hallstatt society collapsed abruptly at the end of the sixth century is widely acknowledged by prehistorians, though whether this was occasioned by a widespread uprising against a hated feudal tyranny (Pauli, 1985) or as a result of a dynastic power-struggle between ruling kin-groups, as envisaged by Arnold (1995) for the Heuneburg, remains arguable. What is clear from the archaeological evidence is that there was major disruption of the political and social order at the end of the sixth century and the beginning of the fifth century, and that a shift in emphasis in terms of southern contacts, from the Massilian Rhone axis to the trans-alpine axis from Etruria to the middle Rhine, was either a contributory cause or a consequence of this upheaval. A further concomitant was a progressive decline in the deposition of lavish grave-goods in the La Tene A and La Tene B phases, though hardly to the point of egalitarianism (Pauli, 1985, 35), and an increase during middle La Tene in the number of graves that display the standard equipment of a 'warrior burial'. Such must surely be the simplest interpretation of tombs like Somme-Bionne and La Gorge Meillet in the Champagne, despite attempts in some quarters to 'pacify the past' (Keeley, 1996, 19; Collis, 2003, 213). In effect, there was a widespread change in the social infrastructure of north-alpine Europe that led to the kind of warrior chiefdoms and proto-states whose activities from the fourth to the first centuries bc are described extensively in classical sources.

This change in social infrastructure in west Central Europe coincides with the appearance in the mid-fifth century of early La Tene art. In the establishment of the new order in the fifth century, it doubtless provided a means for princely or warrior patrons to express their identity in contrast to the displaced order, and especially asserting their independence in the adopting and adapting of southern models into a new and vibrant art style to adorn their table services, their weapons and their personal ornaments. When Jacobsthal made his famous observation that La Tene art was an art with no genesis, therefore, he was in effect remarking the fact that the La Tene hierarchy was explicitly expressing its independence from the older Hallstatt order. La Tene art was an art with no genesis because it was intended to symbolize the new order in contrast to the old, not because its social, economic or technological roots were not embedded in the past.

Despite some recent reservations regarding the interpretation of 'warrior burials', warriors, or more strictly equites or mounted warriors constituted one of the two classes in Gaulish society that Caesar reckoned to be of any significance (de Bello Gallico, VI, 13; note that Caesar did not say that the equites and druids were the only classes apart from theplebes, merely that they were the two that were of any account from his perspective), and there is no archaeological basis for believing that this was not in essence true. Contemporary opinion in Spain (Lorrio, 1997; Almagro-Gorbea, 1998) certainly argues for the importance of the warrior elite in leading Celtiberian resistance to Rome. From the later Bronze Age onward the sword, predominantly long, together with the spear, were the prime weapons of Celtic society, while defensive equipment regularly included a shield, with helmet and body-armour perhaps reserved for warriors of greater status or distinction. The 'triple panoply' of sword, spear and shield, reflected widely in north-alpine Europe in the La Tene Iron Age, is the same essential combination that also characterizes Celtiberian cemeteries in Spain. These weapons doubtless were treated as highly personal items of equipment. The well-known inscription Korisios stamped in Greek letters on a late La Tene sword from Port, Bern (Wyss, 1955) is generally supposed to enshrine the name of the owner or the maker of the sword, but equally it could have been the name of the sword itself, personalized like the weapons of epic literature. The Port sword is also distinguished by a stamped impression adjacent to the name depicting a pair of goats flanking the tree of life, an 'orientalizing' image doubtless borrowed from Mediterranean sources. A series of swords distributed from Switzerland (Drack, 1955) through the middle Danube to Hungary (Szabo and Petres, 1992, 61-3) and dating from the middle La Tene into the first century bc displays stamped motifs, anthropomorphic or zoomorphic, that have generally been taken to be either the makers' marks or emblems of the swords' owners. Located just below the hilt they would scarcely have been visible to an adversary in combat, but nevertheless may have served a talismanic purpose. Recurrent among the zoomorphic images is that of a boar. Anthropomorphic images takes a variety of forms, but one variant, of a human head in profile, certainly resembles the style of contemporary coinage.

One other aspect of the Port sword is worth remarking; it was bent to a right-angle in a manner generally assumed to imply a ritual act (Megaw and Megaw, 2001, 158). This practice is known from both burials and hoards in Iron Age Europe, and is widely documented from the earliest Iron Age in and beyond the Hallstatt culture zone and throughout the La Tene period north and south of the Alps. From the Filottrano cemetery, the sword and scabbard with Vegetal Style ornament was similarly treated. Beyond the Pyrenees in the cemeteries of the Celtiberians a number of burials show deliberate bending of both swords and spears (Lorrio, 1997). Perhaps the most remarkable examples are those from the sanctuaries of Picardy like Gournay-sur-Aronde, where a range of armaments was systematically destroyed. Here, however, we may be seeing the destruction and dedication of the spoils of war, as opposed to the deposition of the proper and traditional perquisites of the dead. Though the practice of decommissioning weapons may be regarded at one level as indicative of a pan-European Celtic ritual, therefore, marked regional differences between funerary practice and the evidence from sanctuaries suggests a degree of complexity that militates against simplistic explanation (Rapin, 1993).

Archaeological classification may be based upon different criteria in successive periods without apparently exciting comment or explanation. Late Bronze Age and

Hallstatt C—D sword typology is based entirely upon sword morphology. For the La Tene period swords themselves are hardly diagnostic, and it is the scabbards that form the basis of classification. Scabbards may be assumed from an early date, since an unsheathed sword is both hazardous and threatening, but examples must have been made of organic materials. For the late Bronze Age and Hallstatt C periods chapes of bronze imply scabbards of which no other archaeological trace survives. Shorter daggers of the ensuing Hallstatt D phase are the first in north-alpine Europe to have metal sheaths. By the early La Tene Iron Age metal scabbards regularly provided a surface for ornament, as doubtless had their leather antecedents, if indeed the Swiss technique of chagrinage is a legacy of that medium. Embellishment of scabbards doubtless also conveyed symbolic meaning and insignia of status, with particular designs like dragon-pairs or the triskele motifs of some Swiss swords probably also indicating achievement or rank among the warrior class.

One widely distributed variant of sword from the early Iron Age is the anthropoid-hilted type, a more elaborate version of antenna-hilted forms. The 'legs' and 'arms' were functional, of course, the one as hand protection, the other to prevent the weapon slipping out of grasp. But the addition of a knobbed pommel, in later Romanizing versions depicted as a recognizable human head, makes it probable that an anthropomorphic image is implied in the design from the outset. Hawkes (Clarke and Hawkes, 1955) had drawn a clear distinction between these later (La Tene 2 and 3) variants and the 'pseudo-anthropoid' precursors in which no human features are represented, but in the context of Celtic iconography it seems wholly unnecessary to demand explicit representational features. Any late Bronze Age ancestry, in the spirally-coiled antenna sword series, and any relationship to the antenna swords of south-western France and the Iberian peninsula (Quesada Sanz, 1997) is perhaps more tenuous. But it is hard to resist the belief that the explicit personification of the anthropoid and pseudoanthropoid series was not implicit in a wider range of weapons in a heroic tradition.

Closely related to warrior equipment is equestrian gear, not necessarily associated with a wheeled vehicle. Two-wheeled chariots were evidently an important accessory to the infantry warrior in the earlier La Tene Iron Age of Continental Europe, evidently surviving as an actual vehicle for engaging in battle in Britain as late as the first century bc. The mounted warrior, conventionally regarded as an introduction of the Hallstatt C Iron Age (Cowen, 1967), is a different style of combat, and again one in which the warrior doubtless took a personal pride in his horse's prowess and ornamental trappings. The commonest indicator of horse-riding is bridle-bits, comprising snaffle-bars and cheek-pieces, sometimes accompanied in grave inventories by other items of harness attachment such as rein-rings and phalerae. All may be ornamented, with phalerae affording the best of limited surfaces for embellishment. Other metal trappings of a not specifically functional kind may have been used, as was inferred from the Hallstatt burials at Court-Saint-Etienne in Belgium (Marien, 1958, Fig. 46). Much of the evidence for Hallstatt C and D comes from vehicle burials, where the implication is that the horse-gear was for paired draught animals, decked out for display, rather than to enhance the martial image of war horses. For ceremonial or parade purposes, we might have expected more evidence like the Torrs pony-cap, though the example of the frozen Iron Age tombs of Siberia serves as a reminder of how much of this equipment, from saddle blankets to ornamental trimmings, might have been of textiles and other perishable materials.

Vehicle embellishments, though not always identifiable with certainty in the absence of contextual data, occur throughout the later Bronze Age and Iron Age. In Hallstatt C, wooden yokes attracted ornament in a number of Central European graves, as in Grave 46 at Hradenin in Bohemia. In the early La Tene a notable example are the plaques emblazoned like a carriage crest at the centre of the Waldalgesheim yoke. Linch-pins attracted elaboration and ornamentation, like the Bohemian jangling pendants of Hallstatt C, or the strikingly whimsical Plastic Style images like those from Manching or Mezek.

Personal ornaments include a wide range of types, from prestigious objects like torcs and arm-rings in gold or silver, for which a high-status or even cult significance might be inferred, not least because their weight and embellishment may have made them cumbersome or downright uncomfortable to wear. The occult significance of torcs is inferred from their association with funerary stele, statuary of squatting deities or representations like the horned figure holding serpent and torc on the Gundestrup cauldron. But they are also represented on sculpture from the Dying Gaul to the life-sized warrior at the Castro di Lezenho. Even Boudicca was reported as wearing one into battle, so that a special significance in the context of warfare may reasonably be supposed. Torcs in the strict sense of twisted bars or plaited strands are seldom found in graves, though lavish neck-rings are not uncommon. The sumptuous Snettisham hoards defy adequate explanation. The relative paucity of metal-working evidence does not indicate a major production centre, yet there is no structural or other evidence for its being a ritual focus of outstanding importance.

More everyday items like pins, brooches, belt-buckles and bracelets evidently served primarily as functional dress-fasteners, but these too may have indicated identity or status of the wearer. The pin, universal dress-fastening of the Urnfield late Bronze Age, gave way in the Hallstatt Iron Age to the brooch, of which the 'safety-pin' model in successive guises become the hallmark of the La Tene Iron Age. Pins and brooches are generally regarded as fastenings for a cloak or mantle, worn around the shoulders over a tunic, knee-length for men and ankle-length for women. Trews, originating in the east, first appear in north-alpine Europe on the early La Tene scabbard from grave 994 at Hallstatt. Depicted with horizontal stripes, these are the equivalent of Gaulish bracae, described by Diodorus (Bib. Hist. V, 30) and Strabo (IV, iv, 3), based on Posidonius' observations at the beginning of the first century bc. Diodorus adds that the Gauls wore tunics dyed in various colours, and striped cloaks with chequered patterns. All would have furnished opportunity for displaying local 'tartans' to proclaim identity and social kindred, and the brooches that fastened the cloak may likewise have reflected local fashions and traditions. It is not simply the design or decoration of the artefact itself that could have conveyed the identity or status of the wearer, but the position in which it was worn, and the combination of pieces that made up the ensemble. Hodson (1990) attempted to identify status in both male and female graves at Hallstatt on the basis of grave-good combinations, and similar patterns might be detected in grave-groups from La Tene cemeteries in Bohemia (Waldhauser, 1987). The importance of dress and being correctly equipped may be inferred from the subtle complexities of the anthropomorphic belt fittings of the later Plastic Style in Bohemia (Kruta, 1975) and elsewhere in La Tene Europe in the context of both male and female attire.

If weaponry and jewellery can be regarded in any sense as 'personal' artefacts, sculpture almost by definition connotes a public statement. Sculpture is not uniformly represented in the Celtic world. Where it does occur regularly, it is commonly either in a funerary context, as in the grave markers of the Hallstatt or early La Tene Iron Age, like the Hirschlanden and Glauberg figures, or in cult contexts like the human heads from the Celto-Ligurian sanctuaries of southern Gaul. Stone stele may have been actual grave markers, but the setting up of individual stones could simply have been commemorative without associated burial, or as indicators of territorial identity. The early Iron Age stele of Brittany could have been commemorative, or perhaps focal points with which local communities identified in ceremonial gatherings. The Irish decorative stones like Turoe probably belonged to this category. The guerricos of the Portugese castros probably served a combination of roles, not evidently funerary, but possibly commemorative and cultic in invoking the powers of ancestors, heroes or divinities to ward off threats to the citadels that they guarded. Some sculpture embodied fertility symbolism, like the explicitly phallic Pfalzfeld pillar, a meaning that may have been implicit in other standing stones. Sculpted heads are known in a variety of forms, and their wide geographical and chronological distribution invites glib equation with a pan-European Celtic 'cult of the head'. It does not seem unreasonable to regard double-heads like the Roquepertuse example or the tricephalic head from Corleck, Co. Cavan, as symbols of omniscient deities, and the recurrent 'leaf-crown' symbol certainly implies a special identity, divine, ancestral or royal, for the figures depicted. The Class 1 symbol stones of Scotland may have served a commemorative role, or may have been statements of communal or dynastic identity. Perhaps too they may have provided a focus for assembly and religious observance, a forerunner of the sculpted crosses of the early Christian period.

One medium that was expressly designed for religious purposes was the illuminated manuscripts of the early Christian period. The religious orders of the earlier Iron Age, collectively described by classical writers as druids, did not practise literacy, enshrining their spiritual truths in an oral tradition that was exclusive and probably elitist. In the case of the Gospel Books, the Latin text would still have been accessible only to a limited number in holy orders, and the images and symbolism of the illuminated pages likewise would have been fully comprehensible only to those who were schooled in scriptural exegesis.

'Art objects' doubtless conveyed a 'meaning', though not necessarily a universal meaning across the Celtic world, or even to all sections of society within one community. That meaning may have been symbolic without necessarily incorporating a 'code' that the modern researcher might endeavour to crack. Regional and local ceramic styles can be identified variously across Iron Age Europe, but it is unclear when recurrent formulae in ornamentation transcend local fashion or tradition to proclaim group identity or to invoke protective or other supernatural forces. It seems likely that conventions in the geometric ornament of ceramics, notably elaborate in the Hallstatt C phase in southern Germany, for instance, would have reflected local identities and traditions in much the same way that ceramic or textile decoration did in more recent ethnographic and vernacular contexts. Higher-status metal artefacts may have had more specific symbolic or ritual associations. It is not difficult to imagine that weaponry and defensive armour would have displayed images and symbolism that not only proclaimed the identity of the bearer and his kin group, but also offered him talismanic powers of protection. Bird and boar imagery on helmets doubtless were endowed with such occult powers, as well as affording a distraction to the opponent and enhancing the physical stature of the wearer. In the context of pre-state societies, when conflict was arguably more 'ritualized' than 'militarized', even if no less lethal in its consequences (Keeley, 1996), even 'parade' shields of the Witham, Wandsworth and Battersea class may have been employed in the ritualized preamble to battle. Carnyxes, together with related classes of horns from the later Bronze Age onwards, doubtless fulfilled a special role in this context, as attested by both archaeological and documentary sources. Scabbard ornament especially we might expect to have had a symbolic or talismanic dimension, either personal to the user or forbidding to an adversary. The dragon-pair imagery of middle La Tene scabbards, with its very extensive distribution from Eastern Europe to the Pyrenees, comes closest to a pan-European Celtic device, perhaps heralding the rank of the bearer or the supernatural forces whose protection that he was invoking.

One medium in which motifs and designs must have been intended to convey a 'meaning' is coinage. Notwithstanding the intrinsic value of gold and silver coins, coinage within a market economy is essentially endorsed by the issuing authority. In its earliest usage in the Celtic world, coinage may not have been used as currency in this way, but for major social transactions such as diplomatic gift, dowry or payment for mercenary services. But by the late La Tene period smaller denominations, base metal coins and even forgeries would be consistent with a developing market economy. Whether distributions of coin types can be equated with tribal territories may be debated, but it is clear that some regions had distinctive design formulae that doubtless expressed the individual identity of the local political and perhaps religious authority. Particularly bizarre is the Armorican representation of miniature human heads on the ends of braids emanating from a larger human head: to offer an interpretation might be to walk the path that angels eschew, but a special significance or veneration of the head seems to be implied. Other images have a wider currency, especially animals such as boar, bulls and birds: horses are commonly represented too, but are present as a consequence of the coinage's Macedonian models. These were doubtless symbols of strength and aggressive vigour, and hardly exclusively so to Celtic society. The depiction of individuals bearing swords or spears, or wearing torcs, together with cauldrons, wheels or sun-disc, and possibly even altars, underlines the aristocratic, heroic or ritual milieu that the coins invoke. As with later Iron Age symbol stone art in Scotland, it may be that it was the combination of symbols rather than any intrinsic meaning of the images themselves that conveyed their significance.

Finally, from the highlighted categories, pottery and domestic artefacts when recovered from settlements are most likely to reflect everyday activities and social norms, which is not to suggest that a symbolic or ritual dimension need be lacking. In the Urnfield late Bronze Age, cylinder neck urns have a widespread though not uniform distribution for much of the period, extending across the Pyrenees into the Ebro valley. Widely-splayed dishes are frequently deployed as covers for the urn itself or as accessory vessels. Recurrent decorative fashions include fluting or rilling, horizontal, vertical or diagonal, as well as incised or excised (Kerbschnitt) styles. How far these ceramic fashions were designed for funerary as opposed to domestic use is still unclear. But such a widespread currency suggests a compelling convention in the shape of cinerary receptacles, and the possibility of special funerary production should not be excluded. In Hallstatt C funerary contexts, pottery seemingly played an important role in the funerary feast, if the numbers of vessels recorded in burials like those of Hradenin, or recent evidence from Hallstatt itself, are indicative. Hallstatt C wares in the west Hallstatt zone are of high quality with elaborate geometric decoration, their recurrent style suggestive of a recognized convention if not a specific 'meaning' in the funerary ritual. Archaeological classification, too often based upon cemetery assemblages rather than settlement remains, perhaps has accorded too high a priority to grave-groups as cultural indices, and the ceramic assemblages from domestic settlements seldom reflect the range or quality of those from high-status burials. The ornamental designs of painted pottery from early La Tene cemeteries in the Champagne accords closest to that of the metal-worker, perhaps suggesting that these vessels were the product of specialist craftsmen rather than of domestic or village industries. Wheel-thrown pottery is known in north-alpine Europe from the early La Tene, and the continued production of hand-made wares in many regions doubtless reflects only the absence of an economic infrastructure within which the new technology might have been exploited. Nevertheless there is abundant evidence for regional fashions and perhaps an emerging professionalism in pottery production, in distinctive styles of ornament like the La Tene Braubach pottery of Central Europe, the stamped wares of Brittany, or even some of the regional styles of Southern Britain.

A major change in pottery production and its ornamentation takes place in the later La Tene in temperate Europe, with the rise of the oppida economy supporting production and distribution on a scale not hitherto witnessed in Iron Age Europe. Renewed contacts with Italy and the Mediterranean world undoubtedly had a major impact upon production in north-alpine Europe, as well as introducing Italic pottery and metal-work types. The use of the wheel resulted in a greater standardization of shapes, and encouraged production on a commercial scale. Ornamentation, including painting and graphite-coating, may be technically more proficient, but lacks the individuality of earlier ceramic styles. These changes mirror trends already noted in the production of brooches and other small ornaments such as glass beads, and are integral to the emergence of incipient state social structures and a market economy. In Gaul and in the Hispanic peninsula, Roman colonization progressively introduced new styles to regions that lay on the expanding frontiers, and made its impact on native production. Even in regions like Britain and Celtiberia, chauvinistically resistant to political domination, high-status imports were readily acquired by the local aristocracies, and new technology like the potter's wheel and lathe-turning of shale vessels was adopted by local craftsmen.

Over the two millennia that have been the subject of this study we would not expect that the role of Celtic art and art objects would remain unchanged. It seems unlikely that the one-size-fits-all principle will adequately explain the role of Celtic art in all quarters of Celtic Europe from the later Bronze Age to the early historic period. In Urnfield Central Europe and the west Hallstatt chiefdoms, it is probable that the repertory of high-status art objects was strictly controlled by an aristocratic or martial elite, while other forms of art objects were more widely understood, even though they may have been locally produced within a village or household. With the emergence of proto-state societies in parts of Europe in the immediately pre-Roman period we might anticipate from ethnographic analogy (Layton, 1991) that art objects imbued with ritual significance would have been in the custody of a restricted cabal that may also have exercised wider social authority. The Yoruba Ogboni evidently was a secret and highly selective organization that also mediated in disputes and imposed legal sanctions, a role that is also attributed to the Gaulish druids (Caesar, de Bello Gallico 6,

13). On the other hand, the beginnings of a market economy in Iron Age Europe heralded important changes in the nature of production and exchange that must have had the opposite effect of widening the currency of a number of everyday 'art' objects such as brooches or glass bracelets.

Later Celtic art need not be expected to conform consistently with the conventions of earlier Celtic art, since over a thousand years even the most conservative of traditions may be expected to change. In the case of later Insular Celtic art, there are striking innovations, like the almost total predominance of stone sculpture in 'Pictish' symbol art or the novel medium of manuscript art in the Early Christian era. But there are also significant underlying similarities in the various influences that combined to create this renaissance in Insular Celtic art, and in the social and cognitive forces that generated its finest material expressions. There are striking echoes in the genesis of later Celtic art of the factors that were undoubtedly contributory to the creation of the early La Tene phenomenon in north-alpine Europe. First, there is the catalyst of external, even long-distance contacts, resulting in the fusion of stylistic innovations with local traditions. In the early La Tene it was classical and orientalizing influences, deconstructed and re-assembled, that transformed and transcended the prosaic styles of Urnfield and Hallstatt ornament. In the manuscript art of the Early Christian era, a fusion of exotic styles is equally evident, Germanic and Mediterranean as well as 'Ultimate La Tene'. In both instances, there needed also to be a compelling authority to generate and sustain these external links, and furthermore to establish an economic environment that could sustain a high level of specialist skills, one that could control the sources of supply and provide an effective milieu of production. In the earlier Iron Age, this was undoubtedly the paramount prince or princess within a hierarchical social order, an order in which the religious authority may have been as potent as the secular. In the early historic period, that authority, no less potent in its very different way, was the Christian Church, and in particular the monastic system. There is, of course, no guarantee that the concomitance of these factors will necessarily result in an efflorescence of artistic achievement, any more than one could re-create the European Enlightenment of the eighteenth and early nineteenth centuries simply by bringing together individuals of appropriate talent in circumstances that might superficially simulate the original. Not surprisingly, therefore, much of the area and period that we have designated Celtic is not characterized by these outstanding examples of artistic achievement, which does not mean that their more mundane products were any less Celtic.

Raymond Firth (1951a) argued that art was composed in a social setting, and had a context in a specific body of beliefs and values (Layton, 1991, 43). Celtic art was described by Ruth and Vincent Megaw (2001, 22) as 'shape-changing', referring to the ambiguity of interpretation that so often challenges the observer. This in itself was surely a reflection of the Celtic mindset, as revealed especially in later Irish and Welsh literature, in which there is no formal boundary between myth and reality, between this world and the Otherworld. Imbued with spiritual significance, it was to this extent a religious art. But it was also equally essentially a social and political art, and a medium through which identity could be asserted. In sum, Celtic art was fundamentally embedded in Celtic society, custom and belief.

0 0

Post a comment