There is no evidence to suggest that anyone in prehistoric or early historic times thought of himself or herself as a Celt. On the contrary, names such as Celt, Gaul or Gael were given by outsiders, whether Greek geographers, Roman historians or Anglo-Norman conquerors, according to their own perceptions of themselves and other peoples. Such large-scale Identities were never shared by the so-called Gelts themselves, despite a certain similarity of material culture, especially in such areas as art and prestige goods, throughout much of Europe, nor was there ever any political unity at such a scale.
The largest social or political groupings that may have been meaningful were much smaller. The classical authors refer to many groups such as the Arverni, the Helvetii or the Iceni; Caesar’s regular word in Latin for such groups is civitas, usually translated into English as ‘tribe’, and though they may have been very different in internal organization from the classical city-state, his use of the term shows that he recognized them as political entities. At their largest, they could number several hundred thousand people; when Caesar conquered the Helvetii, he gave their population as 263,000 and, although there may be some doubt about the exact figure, the general order of magnitude must be correct {De Bello Gallico 1.2-30).
Some of these tribes are known to have comprised a number of smaller groups, for which Caesar’s term is a pagus; it is not clear, however, whether these were equal subdivisions, perhaps on a territorial basis, or whether they were groups subordinate to the dominant one which gave its name to the larger entity. Whether they had an ideology of identity, at the level of either the civitas or thepagm, based on concepts of ethnicity or descent rather than just on political allegiance to a common elite and a single individual, is obscure. Archaeology has had little success in identifying common traits in sites or material to match these political groups. In the Late Iron Age, the production and distribution of coinage has often been attributed to a tribal origin (e. g. Nash 1978a), but that may merely reflect the political purpose of the coins; otherwise, the very rarity of such apparently tribal traits as the gold neck-rings of the Iceni in eastern England or the distinctive inhumation burial tradition of the Parisl in east Yorkshire simply emphasizes how little such identities were signalled in ways that archaeology has yet been able to recover, if indeed they were signalled at all.
In Early Christian Ireland, the fundamental unit of social organization was the tuath (Kelly 1988: 3-6), of which there were at least 150, each comprising up to several thousand people; it was larger than a kin-group, and it was a political rather than an ethnic or cultural entity. Membership of a tuath was an important part of an individual’s identity, not least because of the political obligations that it entailed. Individual tuatha varied considerably in size and prestige, and their power fluctuated over time; at times one tuath could exercise dominance over a few others. By the eighth century AD some dynasties were beginning to be able to exert more stable authority over even larger territories. As with the prehistoric groups, archaeology has had little or no success in identifying the extent of the tuatha.
There have been various attempts to define political groups on archaeological grounds alone, mainly by defining territories around hypothetical political centres. One of the most successful concerns the late Hallstatt fortified sites with rich burials of the sixth century BC in southern Germany (Harke 1979); if these really were the centres of independent polities, then each one dominated a territory about 100 kilometres across, much bigger than the Irish tuath. On the other hand, the rich burials of the fifth-century Early La Tene period in the Rhineland have a much closer spacing, suggesting a very different sort of social organization.
It would be dangerous to extrapolate uncritically from the few areas and periods for which we have good documentation, whether archaeological or historical, because the patterns they reveal are so different. There is no reason to think that political groups of the size of the larger late iron age tribes described by Caesar or of the Irish kingdoms of the late first millennium AD were typical of other periods. These may well have been the product of unusual historical circumstances, and we should expect the more common pattern to have been one of much smaller political units, perhaps of the scale of the tuath. Such political groupings may therefore have offered what was generally only a very small stage for the political activities which took place within and between them.
Perhaps the most important form of identity for an individual was as a member of a kin-group. According to the early Irish literature, the most informative source for this, many social rights and obligations were exercised by the derbfine, or four-generation descent group sharing a common patrilineal great-grandfather (Kelly 1988: 12-16). The group held land in common; it was responsible for offences committed by any of its members, and likewise sought retribution in common for grievances. By the eighth century AD a three-generation group with a common grandfather was beginning to be a more important social unit, but kinship was still an important basis for social relationships. It is highly likely that kinship was the dominant factor in social relationships in earlier periods also, but the classical authors tell us little of social organization at this level, and archaeology has not been good at exploring social organization in such fine detail.
We know even less about the fundamental forms of social identity within the kin-group, in particular concepts of male and female or adult and child. It is clear that categories of male and female were well established; prehistoric burial traditions regularly show distinctive sets of personal ornaments and possessions which are correlated with sexual identification of the skeletons to show gender divisions in society. The Irish laws also treat men and women differently, in terms that seem familiar to modern readers. It would be tempting to think that these categories were unproblematic, but we do not really know the conceptual structure that underlies them.
The difference between the categories of adult and child is more varied. The best archaeological evidence is from cemeteries; most regions show a significant underrepresentation of young people, which may suggest that the very young were thought of as a separate category not to be treated in the same way as older people. The treatment of older children also differs from place to place; in some cases they are buried with the same rites and accompanying goods as adults, but in others they have distinct rites and sets of ornaments, suggesting a variability in the way that differences between adults and children were constructed.
The Irish laws show the practice of one particular region. Until the age of 7, children were reared in their parents’ household, but then left to join the family of foster-parents; this was a common practice, but we do not know whether it was universal or limited only to particular levels of Irish society. After the end of fosterage at the age of 14, females were ready to join new households through marriage, but young males could not establish themselves as full members of the adult community with their own household until they had inherited land. There were thus large numbers of young landless males of intermediate status who joined together in bands or fianna, which engaged in hunting, fighting and on occasion acts of brigandage. Such age-grade institutions can be found in other Indo-European societies, and may originally have been more widespread (McCone 1990: 203-17). They are not documented for the prehistoric Iron Age, but the regular occurrence of Celts as mercenaries in the armies of the Mediterranean states may well be connected with such a tradition of young men’s wandering and fighting.
World History
