Ethnicity and Material Culture in North Cameroon

1 downloads 0 Views 61KB Size Report
non-nucleated but focussed on a mountain, was the largest political unit (Martin 1970). Each was organized by segments of exogamous patrilineal clans and two ...
Canadian Journal of Archaeology, Vol. 15, 1991

171

Ethnicity and Material Culture in North Cameroon Nicholas David, Kodzo Gavua, A. Scott MacEachern and Judy Sterner Department of Archaeology University of Calgary

INTRODUCTION In The Uses of Style in Archaeology (Conkey and Hastorf 1990), Whitney Davis (1990:19) defines style for the art historian’s purposes as a description of a polythetic set of attributes present by virtue of common descent from an identifiable artifactproduction system. Likewise ethnicity appears to the outsider as a polythetic set of cultural behaviours explicable in terms of shared membership in an identifiable multipurpose socio-cultural group. Multiple referents, some but not all necessarily nested one within another, are obviously possible in both areas; and neither style nor ethnicity need have any substantial historical depth. Cubism was a brief phenomenon while ethnicity is always emergent and may be transitory. This paper draws on the work of several members of the Mandara Archaeological Project team first to describe multiple ethnic referents found among the small-scale, largely acephalous, montagnard societies of the northern (near Mora) and central (near Mokolo) Mandara highlands of north Cameroon (Figure 1). Material culture plays a constitutive role in the forging of their ethnicities. Second, both Davis and Conkey (1990:15) insist on the importance of studying “design and the organisation of production as integral to any study of style”. Another aim is then to characterize the production systems behind the variation observed. In this region the stylistic variation in material cultures can best be described in terms of shared access to a “symbolic reservoir” on which different groups draw in different ways (Sterner 1990). Lastly we comment on the relevance of our work for archaeology and in particular “taken-for-granted notions about the inherent boundedness of groups” (Conkey 1990:12).

SOCIETY, SYSTEMS OF PRODUCTION AND MATERIAL CULTURE The “tribes” or “ethnic-linguistic groups” found on maps of the peoples of Africa are commonly figments of the administrative or pedagogic imagination. So it is on Figure 1. We shall take the Mafa and the Bulahay as examples of a southwestern cluster of peoples, and the Muraha, Plata and Uldeme as representatives of a northeastern group. The Mafa, numbering about 240,000, only began to become conscious of their existence as an entity after independence, when would-be politicians attempted to mobilize a shared language in their favour. Before that the individual community, non-nucleated but focussed on a mountain, was the largest political unit (Martin 1970). Each was organized by segments of exogamous patrilineal clans and two endogamous castes-cultivators and “transformers”, that is to say specialist iron-workers, funeral directors, diviners, potters and midwives. What little temporal authority was exercised above the level of the localized clan segment was the affair of a council of clan elders, a priest-chief primarily responsible for the ceremonial cycle and perhaps another for

172

Journal canadien d’archéologie, Vol. 15, 1991

Figure 1 Parts of northern Nigeria and Cameroon, showing “ethnic-linguistic groups”. Northeastern and southwestern clusters are indicated by dashed lines.

rain and the crops. In both northeastern and southwestern clusters, a few pope-like figures extended a ritual influence over several groups. In most communities most clans had traditions of their founder’s arrival from other Mafa settlements; a minority were of other linguistic groups (Miiller-Kosack 1987). In at least one case, at Mendeje, a disparate group of immigrants who had moved into new territory formally constituted themselves as a clan. Mafa material culture nonetheless contrasts with that of their non-Mafa neighbours in terms of, for example, the layout of their homesteads, the form of granaries, pottery typology, iron smelting furnaces and forges, and mortuary practices. Mafa communitiesalso use material culture to differentiate themselves from Mafa neighbours, for example by reversing left and right in the layout of compound and rooms, by the form and decoration of the same emic pot type, or by the way waterpots are carried on head or shoulder (Gavua 1999). Furnaces varied, we believe, by both village and clan, but generally clans were not distinguished by their material culture. The “Bulahay” are a classic case of the non-equivalence of language and ethnicity. In its common usage the term, rejected by those it purports to designate,

Canadian Journal of Archaeology, Vol. 15, 1991 refers to the inhabitants, totalling 7,700 in all, of the communities of Mefele, Sirak, Mohour and an outlier, Chougoule, that has a heavy, recent Mafa admixture. They speak dialects of the the same language, baptised mefele by Barreteau (1987) but do not consider themselves to be or act in any way as members of a group. Sirak comprises six clans, not one of which claims autochthony. They are of varied origins. There are no diagnostic features of “Bulahay” material culture, although, with Chougoule excluded, there are commonalities that make it possible to speak of (but probably not identify in the archaeological record) a style in Davis’s polythetic sense. At the level of the community, pottery suites are distinctive, as are elements of costume and other material domains, not to mention ceremonial cycles. Once again we can detect no clan-based differentiation. To the northeast, in the hills and on the plains near Mora, such communities did not exist in precolonial times. There a major disjunction between material culture and language occurred in the cases of the Wandala and the Muraha. The former developed a para-Islamic and later Islamic state in the classic West African mode, adopting an appropriate material culture especially as regards architecture, costume and warfare. We shall consider them no further here. The Muraha speak the same language and have always interacted closely with the Wandala, but have retained much of their traditional religion and material culture. The Muraha and their montagnard neighbours are acephalous collections of small, exogamous patrilineal lineages or clans, existing in a web of traditional relations of alliance and opposition with similar groups, and with historical ties to a particular home territory (MacEachern 1990). The patrilineage or its co-resident segment is here the primary referent of ethnicity, marrying, trading and fighting often with the same neighbours. Neither do these groups have a transformer caste. Some crafts, such as iron-working, are carried on by, but are not explicitly restricted to, particular lineages, while others are generally practised, for example potting by most women. Burial is not the prerogative of specialists. Our knowledge of variation across domains of material culture is less than to the southeast. “Ethnic-linguistic” groups are less distinguishable by their pottery than among the Mafa and their neighbours (MacEachern 1990). Rather there are broader scale traditions that characterize the Plata and groups to the north of them versus the Uldeme and others to the south and east. Spheres of exchange of women in marriage, themselves based in part upon degrees of linguistic similarity and in part on other historical factors, are implicated in such patterning, which is now also influenced by markets. Thus the northern ceramic tradition is spreading southwards, while paradoxically the Plata are becoming more willing to accept identification with their southern neighbours, the Uldeme! Furnaces, on the other hand, seem to have shown considerably more variation by “ethno-linguistic” group and perhaps covaried with the lineages that produced them. At a larger scale there are regional contrasts between the northeastern and southwestern groups. This is well seen in the ceramics, although the boundary between the regions is not a simple one. Thus the Gemjek pottery suite is, in terms of its forms and decoration, allied to the northeast, while it is southwestern in its functional range and diversity. At an even larger scale the style of material and non-material culture of the montagnards of the Mandara highlands is integrated through its derivation from a single “symbolic reservoir” (McIntosh 1989). Sterner (1990 and see David et al. 1988) has argued this largely on the basis of the nature and elaboration of ritual pottery, but it can be seen in many other aspects of culture: ceremonial cycles, for example, or downdraft

174

Journal canadien d’archéologie, Vol. 15, 1991

furnaces for the smelting of iron (David et a/. 1989). Archaeological data (David and MacEachern 1988; Wahome 1989) indicate that this symbolic reservoir began to fill at the start of the regional iron Age, some 2,008 years ago.

INTERPRETATION Ethnicity is contextual and explanation of the relationships between material culture and ethnicity at various levels of contrast requires the integration of historical and anthropological approaches. First we must recognize that the exceedingly high population densities of the present, in the 125 to over 200 per square kilometre range (Boulet 1984:110), are a development of the last few hundred years, during which the montagnards have been subject to slaving and exploitation by the state level societies (Wandala, Sukur, Borno, Fulbe) that dominated the surrounding plains. Before the sixteenth century AD, highland populations were very much smaller and more isolated one from another (MacEachern 1988). Nonetheless even before that time, local famines, droughts and intra-community conflicts must have caused erratic but unceasing movement of individuals, families and occasionally larger groups from one community to another. Subsequently, pressure from the plains led to marked population growth, stimulated on the one hand by the need for warriors and on the other by the demands of a necessarily increasingly labour-intensive agriculture. As usable land -i.e. arable under prevailing political conditions- became a scarce resource, competition between communities would have stimulated stylistic differentiation in material culture (cf. Hodder 1982). On the other hand, the new state-induced stresses speeded up Brownian movement of population into and within the massif and would have generated countervailing pressures towards a growing together of local styles. Such grand scale phenomena do not work their stylistic effects directly but are mediated through production systems. In the casted societies in the south (around Mokolo), transformer monopoly of iron working and particularly ceramics, combined with their role as diviners and ritual specialists, result in their exercising a disproportionate influence upon material culture. Because they constitute less than 5% of the population, transformers are more frequently constrained to marry outside their community than cultivators. Similarly, in a society in which travel was ill-advised and dangerous, iron workers were a scarce resource worth protecting. They could and would go considerable distances to obtain ore, to trade, or even to assist friends and relatives at a smelt. Transformers thus were in a sense a supra-community element, and even, since marriages in communities near linguistic borders took little account of such boundaries, a supra-“ethnic-linguistic”-group element within the population at large. Transformers thus contributed to the spread of ideas within and to the overall coherence of the symbolic reservoir, but simultaneously to the maintenance of local community styles. We might expect contrasts between material culture produced by transformers, that associated with iron working, ceramics and mortuary practices, and other domains, for example architecture and basketry, which the cultivators produced for themselves. What we find are distributions that, in the case of architecture, tend to vary as much with political alliances as with ethno-linguistic groups. Thus the Hide and Mabass are members of the Wandala linguistic group, but their architectural forms-though not the layout of their compounds- are strongly allied with that of the Mafa (of the Mafa group) and of the Wula (of the Kapsiki [more properly psikye] group) respectively. Such political alliances are expectably expressed also in bilingualism and in exchange of marriage partners.

Canadian Journal of Archaeology, Vol. 15, 799 1

175

Bi- and multi-lingualism are extremely common in the northeastern part of the area, where communities are so small that a man -women to a rather lesser extent must often operate in three or more languages and cultures in a single day (MacEachern 1996). Women very frequently marry into a lineage of a different fathertongue, and are constrained to learn it if they have not already done so from multilingual relatives. Members of these societies could and frequently did participate as members in two or more ethnic groups. It is therefore not surprising that, in a region where the small size of territorial lineages demanded considerable interdependence for survival, these units would be less stylistically different&ted than communities in the south. We can do no more in this brief sketch than to indicate some results and some expectations derived from previous research. It is clear that ethnicity in this area is situational, that material culture helps to constitute its several levels, and that the pattterns of that constitution are determined in part by relations of production and in particular the presence or absence of a transformer caste. The covariation of material culture and ethnic-linguistic groups has been seen to differ both across domains of material culture and between the northeastern and southwestern parts of our area. Although in neither region is there a marked coincidence of boundedness between different domains of material culture, this is more developed in the southeast where, despite a shared transformer caste, communities were more self-sufficient and politically independent. But it would be a great mistake to suppose that the stylistic differentiation between groups seen today as a phenomenon in space would also be recognizable in time as a series of (mini-)co-traditions. This would almost certainly not be the case. On the contrary, while at any one time there may be considerable group pressure to conform in material and other ways, intra- and inter-group social dynamics also stimulate variety. The exchange of personnel between groups through marriage and micro-migrations, not to mention regional economic, political and ritual linkages, make for continual local reworking and, from the little we can detect archaeologically, very gradual evolution of the symbolic reservoir to which all groups contribute and from which all draw in the process of elaboration of their material and non-material culture. Finally we should note that, for the archaeologist oriented towards ecological and economic questions, it probably matters very little that she will be unable to reconstruct neat, bounded ethnic groupings, but only the larger cluster that drew on the same symbolic reservoir. For this is surely the significant unit for understanding economic process on a macro-scale. Archaeologists whose interests are rather in the re-construction of frameworks of meaning must on the other hand be prepared to take an interpretive and probably generative (see Barth 1987) approach that can cope with multiple, polythetic and varied expressions of prehistoric ideas.

ACKNOWLEDGEMENTS The research on which this paper is based was carried out under authorisations from the Ministry of Higher Education and Scientific Research of Cameroon, and within the context of a Protocole d’Accord de Cooperation between the University of Calgary and ORSTOM, France. It was supported by Social Science and Humanities Research Council grants (410-86-1046, 410-88-0361, 410-89-0871) and an SSHRC Leave Fellowship (451-85-1231). We thank the Institute of Human Studies, Cameroon, and particularly the Chief of its Northern Station, M. Eldridge Mohammadou, for many kinds of assistance; also ORSTOM for logistic support.

Journal canadien d’archeologie, Vol. 15, 1991

176

REFERENCES CITED Barreteau, D. 1987 Un essai de classification lexicostatistique des langues de la famille tchadique parlées au Cameroun. In Langues et Cultures dans le Bassin du Lac Tchad, edited by D. Barreteau, pp. 43-77. Editions de I’ORSTOM. Paris. Barth, F. 1987

Cosmologies in the Making. A Generative Approach to Cultural Variation in New Guinea. Cambridge University Press. Cambridge.

Boulet, J. (with A. Beauvilain and P. Gubry) 1984 Les groupes humains. In Le Nord du Cameroun: des Hommes, une Region, edited by J. Boulet, pp. 103-157. Editions de I’ORSTOM (Collection Memoires no. 102). Paris. Conkey, M. Experimenting with style in archaeology: some historical and theoretical 1998 issues. In The Uses of Style in Archaeology, edited by M. Conkey and C. Hastorf, pp. 5-17. Cambridge University Press. Cambridge. Conkey, M. and C. Hastorf (editors) 1990 The Uses of Style in Archaeology. Cambridge University Press. Cambridge. David, N. and A. S. MacEachern The Mandara Archaeological Project: preliminary results of the 1984 1988 season. In Recherches Comparatives et Historiques dans le Bassin du Lac Tchad, edited by D. Barreteau and H. Tourneux, pp. 51-80. ORSTOM. Paris. David, N., J. Sterner and K. Gavua 1988 Why pots are decorated. Current Anthropology 29(3):365-389 Davis, W. 1998

Gavua, K. 1990 Hodder, I. 1982

Style and history in art history. In The Uses of Style in Archaeology, edited by M. Conkey and C. Hastorf, pp. 18-31. Cambridge University Press. Cambridge. Style in Mafa material culture. Unpublished Ph.D. dissertation. Department of Archaeology, University of Calgary. Calgary. Symbols in Action. Cambridge University Press. Cambridge.

MacEachern, A. S. 1988 The people of Ngolele and the question of Wandala and Montagnard origins. Paper presented at the 4e Colloque lnternationale Méga-Tchad. Paris.

Canadian Journal of Archaeology, Vol. 15, 1991

177

MacEachern, A. S. 1990 Du Kunde: processes of montagnard ethnogenesis in the northern Mandara Mountains of Cameroon. Unpublished Ph.D. dissertation, Department of Archaeology, University of Calgary. Calgary. Martin, J.-Y. 1970 Les Matakam du Cameroun. Essai sur la dynamique d’une société préindustrielle. Memoire ORSTOM 41. ORSTOM. Paris. McIntosh, R.J. 1989 Middle Niger terracottas before the Symplegades gateway. African Arts 22(2):74-83. Muller-Kosack, G. 1987 Der Weg des Bieres. Unpublished M.A. thesis, University of Frankfurt. Frankfurt-am-Main. Sterner, J. 1990

“Montage”, sacred pots and “symbolic reservoirs” in the Mandara highlands of North Cameroon. Paper presented at the Biennial Conference of the Society of Africanist Archaeologists meeting. Gainesville.

Wahome, E. W. Ceramics and history in the Iron Age of North Cameroon. Unpublished 1989 M.A. thesis, Department of Archaeology, University of Calgary. Calgary.

ABSTRACT Ethnicity involves participation in shared traditions at various scales. Multiple referents of ethnicity in the Mandara highlands, the nature of their expression in material culture, and the underlying systems of production are sketched. In these small-scale egalitarian societies, characterized by Brownian movement of people, the archaeologist can likely define only the larger population cluster, with clines and variation within it, that drew its symbols from the same reservoir. This may well be the significant unit for understanding culture process on a macro-scale.

RÉSUMÉ L’ethnicité s’exprime à différents niveaux de participation culturelle. Nous présentons sommairement les multiples marquers d’identité utilises par les gens occupant les montagnes Mandara ainsi que la manière dont il-s s’expriment dans la culture materielle et les systèmes de production qui les soutendent. Dans ces petites sociétés, égalitaristes, caractérisées par un mouvement brownien des gens, le préhistorien risque de ne pouvoir discerner que les frontières de la plus grande unite ethnique, intérieurement affectée par des variations et des gradations, puisant ses symboles dans le même reservoir. Cette unite peut fort bien être la plus pertinente pour comprende I’activité culturelle à une échelle macrosociologique.