Introduction: Ethnicity in Southern African History

Leroy Vail

Out of the crooked timber of humanity no straight thing can ever be made.
Immanuel Kant

Interpretations

African political leaders, experiencing it as destructive to their ideals of national unity, denounce it passionately. Commentators on the Left, recognizing it as a block to the growth of appropriate class awareness, inveigh against it as a case of 'false consciousness'. Apologists for South African apartheid, welcoming it as an ally of continued white dominance, encourage it. Development theorists, perceiving it as a check to economic growth, deplore it. Journalists, judging it an adequate explanation for a myriad of otherwise puzzling events, deploy it mercilessly. Political scientists, intrigued by its continuing power, probe at it endlessly. If one disapproves of the phenomenon, 'it' is 'tribalism'; if one is less judgmental, 'it' is 'ethnicity'.

Ethnicity's emergence as a central concern for a wide range of students of African affairs is relatively recent, and its forceful intrusion upon the dominant nationalist paradigm of the 1950s and early 1960s was both unexpected and unwelcome.^[1] At that time, it was accepted that Africans were organized naturally into 'tribes', but, as nationalist movements in Africa were then apparently enjoying great success, most observers believed that parochial ethnic loyalties were merely cultural ghosts lingering on into the present, weakened anomalies from a fast receding past. As such, they were destined to disappear in the face of the social, economic and political changes that were everywhere at work. People from all sectors of the political spectrum believed in this vision. For those on the Right and in the Centre, 'modernization' would do the job. Greater access to education, improved communications, and the shifting of people from the slumbering 'traditional' rural sector of the economy to the vibrant 'modern' industrial sector by the beneficent forces of economic growth guaranteed that ethnic loyalties would fade away. In their place would grow a new, nationoriented consciousness which would underpin progressive 'nation-building', especially if the new nation states could make good their promises of a better life for all their citizens. Africa would be a continent of new Switzerlands in which cultural divisions would be of little political importance.

For those on the Left, too, 'modernization' was the key, although it was viewed

from a somewhat different perspective. The break-down of 'traditional' societies by the forces of new, state-sponsored welfare socialism, with its expanded facilities in public education, medicine and agricultural programmes, would allow newly independent African states to 'skip a stage' in the evolution of their societies towards socialism and to enter directly into that blessed condition. In effect, socialism would then provide the material base for a pan-ethnic class consciousness that would transcend, if not negate, cultural differences. Africa would be a continent of new Yugoslavias.

The general paradigm of 'modernization', then, appealed to almost every political viewpoint. For almost every observer nationalism seemed progressive and laudable, while ethnicity—or, as it was usually termed, 'tribalism'—was retrogressive and divisive.

Ethnicity, however, failed to cooperate with its many would-be pall-bearers. It soon became clear that African nationalist movements, ideologically shaped by the basically negative sentiments of anti-colonialism and with little substantive philosophical content relevant to the day-to-day life of ordinary Africans living in post-colonial states, were simply unable to provide them with compelling intellectual, social, and political visions. Once the attainment of independence had made most of its anti-colonial message irrelevant, nationalist 'thought' was transformed into a gloss for the manipulation of the institutions of the new nation-states on behalf of the interests of the ruling political parties in a succession of one party states.^[2] Much state activity was devoted to the pursuit of variously defined forms of 'economic development', but such development proved elusive and the much-desired economic Fruits of Independence generally failed to ripen. That growth which did occur, moreover, was usually to the benefit of the dominant political classes and possessed little popular appeal.

As a result of this quick reining in of nationalism's popular thrust within the bureaucratic structures of essentially artificial post-colonial states, ethnic or regional movements rooted in the colonial era had fresh life breathed into them and came to be seen as attractive alternatives to the dominant political parties with their demands for uncomplaining obedience from the governed. In effect, the revitalization of 'tribalism' was structured into the one-party system by the very fact of that system's existence. Ethnicity became the home of the opposition in states where class consciousness was largely undeveloped. Ethnic particularism has consequently continued to bedevil efforts to 'build nations' to the specifications of the ruling party for the past two decades or more. This hard political fact has called forth ever more systematic repression of dissent by those in control of the state, thus, in effect, strengthening the appeal of the ethnic alternative. Ethnicity's future, even in countries such as South Africa, where industrialization has proceeded further than anywhere else on the continent, seems secure because it is likely to provide an important focal point for whatever opposition to the dominant political classes that might exist.

With its power to divide people politically, then, and with its sturdy resistance to erosion by the ideological forces of national or class consciousness, ethnicity came to demand close—albeit it often very grudging—attention after decades of neglect. Its source and appeal needed reasonable explanations, and interpretations of it have ranged widely, reflecting its multidimensional nature.

The most prominent explanation—if only because of its widespread use—is the one that, despite the great frequency with which one encounters it in media coverage of Africa, is plainly the least satisfactory. In effect, this interpretation is a restatement of the old assumption that Africans are by nature 'tribal' people and that 'tribalism' is little more than an irrelevant anachronism, an atavistic residue

deriving from the distant past of rural Africa. It should have evaporated with the passage of time, but, inexplicably, something went wrong, and it continues to refuse to obey the laws of social and political change. It thus remains able to motivate Africans to frequent actions of conflict and violence. Ethnic consciousness is, in this view, a form of collective irrationality.

The problems with this interpretation are clear. First, it is always dangerous to assume that people consistently act out of mass irrationality. People tend to act rationally, and there is no reason on the face of it to accept that Africans are exceptions. Second, this argument is, in effect, also a tautology with no analytical power, arguing as it does that Africans act 'tribalistically' because they are naturally 'tribal'. Third, and most tellingly, empirical evidence shows clearly that ethnic consciousness is very much a new phenomenon, an ideological construct, usually of the twentieth century, and not an anachronistic cultural artifact from the past. As an offspring of the changes associated with so-called 'modernization', therefore, it is unlikely to be destroyed by the continuation of these same processes. For all these reasons, then, this interpretation must be discarded.

Other, more scholarly interpretations have been suggested to explain the origin and persistence of ethnicity in Africa. All these interpretations have two things in common. First, they derive mainly from the work of anthropologists, sociologists and, especially, political scientists, observers who have been primarily concerned with the situation in Africa at the time they actually studied it. This has meant that their interpretations have usually been concerned with ethnicity's role at the moment of observation and its potential for the future. As such, they usually give only brief attention to its history, presenting whatever history that might be uncovered as mere 'background' to ethnicity's contemporary role.

Second, all these interpretations are also marked by the fact that they have evolved out of the nationalist paradigm dominant from the 1950s into the 1970s. They implicitly accept a basically evolutionary view of human history. In this view, the future ought to be better than the past, and 'better' has been identified with improvements assumed to flow from an increase in political scale and the growth of national unity—in short, from 'nation-building'. As a consequence, most such analyses of ethnicity are concerned with the way it has traduced the promise of modernizing nationalism and are thus predisposed to negative judgments. Their emphasis, therefore, has been on ethnicity's role as a disrupter of the promising trends of secular nationalism that seemed to characterize African politics in the late 1950s and early 1960s and to promise a rosy future.

The intellectual range of these interpretations of ethnicity has been wide. One viewpoint encountered frequently—especially within Africa itself—is that ethnicity is primarily the result of a history of 'divide-and-rule' tactics which colonial governments cannily employed. European anthropologists connived at such policies by specifying 'tribes' culturally within the context of a uniquely colonial sociology, thereby giving the 'tribe' a real, but specious, identity. The element of truth in this explanation has made it superficially attractive, especially as the South African government today actively uses both approaches in its Bantustan policies and in its stress on the uniqueness of 'tribal' culture, patent efforts to promote political divisions among the country's African population.

Yet whatever its merits, it is an explanation clearly insufficient to explain the persistence of ethnic consciousness. This is so for several reasons. First, it fails to explain why, in a particular territory throughout which the colonial state employed roughly the same divide-and-rule policies, ethnic consciousness developed unevenly, strong among certain peoples but not among others, a situation common throughout Africa. Second, it tends to depict Africans as little

more than either collaborating dupes or naive and gullible people, beguiled by clever colonial administrators and untrustworthy anthropologists, a situation which empirical evidence fails to corroborate. Finally, it does not explain how, three decades after the departure of the colonialists, 'tribalism', or its close kin, 'regionalism', lives on as strongly as ever in independent African states, the governments of which have been actively trying to suppress it, and why in some places it is growing up for the first time. The clever blandishments of subtle European administrators are clearly insufficient to explain either the origins of ethnic consciousness or its continuing appeal today.

A second interpretation, especially prominent in the 1950s and early 1960s, arose from the study of urban sociology, especially in the mining areas of Central Africa.^[3] Intellectually, it was linked to the Dual Economy model of 'modernization' theory, and it located its interpretation of the development of new ethnic consciousness in the experiences of rural people in industrial workplaces. As members of various cultural groups left their isolated rural areas and interacted with each other in industrial or urban locales, they formed stereotypes of themselves and others, and these stereotypes effectively highlighted and strengthened culturally defined distinctions amongst peoples. The tendency of employers to prefer certain ethnic groups for certain types of work and their conscious manipulation of ethnic differences to keep the workforce disunited resulted in competition between ethnic groups being built into the hierarchically structured workforce. In this view, ethnicity was a recent phenomenon of the modern urban workplace in which boundaries and distinctions between people had been built up. It was not a phenomenon of the rural areas, where people were assumed to live in accordance with prescriptive patterns derived from a 'traditional' past and where they were largely isolated from peoples of differing cultures. As such, some scholars, as well as most African politicians of the time, assumed that the but recently formed ethnic identities were still malleable and that they would prove susceptible to an easy transformation into a national identity through processes of political mobilization associated with 'nation-building', especially if the labour unions representing such workers could be coopted into the national political establishment.^[4]

This interpretation is certainly valuable for its underscoring of the important point that ethnic stereotypes were indeed largely produced in work situations and in urban settings. Yet it too is unable to serve as a general explanation of ethnicity's origin or, especially, its persistence. First, by emphasizing the boundaries that the creation of ethnic stereotypes among urban Africans produced, which, in turn, created opposing notions of 'them' and 'us', it overlooks the more substantive intellectual content contributed by African intellectuals to the specification of concepts of ethnic self-identity within those boundaries. Positive views about one's history, the heroes of one's ethnic past, and the manifestations of one's culture, especially language, quite simply did not spring automatically from the work situation or the urban centre, yet they have all been central in defining ethnic identities and ethnic ideologies.

Second, by stressing the essentially non-rural nature of the growth of ethnic stereotypes, this interpretation implicitly accepts the notion that rural Africa was preserved in some sort of 'traditional' pickle, antithetically opposed to 'modern' industrial Africa and largely untouched by the forces of change associated with capitalist expansion and urbanization. Such a view of the existence of 'two Africas' with but insubstantial linkages between them has by now been convincingly discredited.^[5] Quite simply, the rural areas of southern and central Africa did not remain unchanged in a brine of 'tradition', with meaningful change

restricted to areas of obvious economic growth. Historical change affected the rural areas as much as it did the industrial and urban areas. More to the point, empirical evidence abundantly demonstrates that it is to the rural areas that one must look for most of the intellectual content of ethnic ideologies as they developed during the twentieth century in response to such change.

A third interpretation of the growth of ethnicity is that it resulted from uneven development within African colonial territories.^[6] Certain peoples were able to do comparatively well from the educational and employment opportunities that colonial capitalism presented unevenly, with aspirant petty bourgeois groups able to establish themselves in some areas but not in others. When it became clear that the colonial era was nearing its end, these petty bourgeois groups mobilized support along ethnic lines so that they would be in a position to maximize their opportunities for access to resources and power after independence. This situation led in turn to the continuation of specifically ethnic politics in many countries of Africa, resulting in a rash of coups d'état and civil wars as ethnic fragments of the national petty bourgeoisie competed for their own advantage. From this perspective, ethnicity tends to be seen instrumentally, as little more than an ideological mask employed by ambitious members of upwardly-aspiring groups as a way of papering over growing class divisions within their ethnic group so as to secure their own narrow interests through demagoguery and mystification. Ethnicity, then, when ordinary people embrace it, is the very epitome of 'false consciousness'.

Again, this interpretation, with its emphasis on the pivotal roles of influential petty bourgeois intellectuals functioning as culture brokers and on smart politicians craftily manipulating popular opinion, especially in the post-colonial period, has obvious elements of truth in it. It also goes far towards explaining why some cultural groups who have had such a 'modernizing' petty bourgeoisie within them are more 'tribal' than other groups within the same country who lack such a class. Yet, on its own, it too ultimately fails to explain ethnicity's appeal. This is so because it goes too far in depicting ordinary people as being credulous, blindly accepting the ethnic party line from their devious betters. It fails to explain why, today as in the colonial period, the ethnic message should find such resonance with ordinary people. Why, in short, have ordinary people chosen so often to support ethnic politicians rather than national politicians? What is in the ethnic message that is not in the nationalist message? One must once again guard against the assumption, necessary to this interpretation, that ordinary Africans act either irrationally or sentimentally.

Finally, deriving from a Durkheimian notion of the importance of the role of the 'community', or Gemeinschaft, there is the 'primordialist' interpretation of ethnicity, an interpretation which now appears to be in the ascendancy amongst many scholars.^[7] Its attraction lies in its serious attempt to answer the crucial question as to why the ethnic message possesses such strong appeal. This interpretation seeks the explanation in the realm of psychology. Africans, it is argued, were badly affected by the disruptive socio-economic and political changes of the late nineteenth and twentieth centuries. Pre-capitalist and pre-colonial hierarchies and elements of order in social life were undermined by the growth of capitalist relations and the impact of colonialism, thereby depriving people of social and psychological security. As a result, in a hostile world they have instead sought security through the invocation of a lost past of firm values as a way of recreating a life in which they can achieve emotional and, even, perhaps, physical safety. Ethnic identity provides a comforting sense of brotherhood in a world tending towards social atomization and rootlessness. Ethnic leaders

represent and embody the unity of the cultural group. In this view, ethnicity is a kind of romantic rejection of the present. Enduring rather as religious fundamentalism or faith healing do in western societies, it is a reaction to the sterility of modern positivism and has become something akin to a civil religion with great emotional appeal.

Once again, this argument is attractive, particularly as the ethnic message, once established amongst people, does appear to be a part of the natural order of the universe. It categorizes people in accordance with inevitable, largely unselfconscious ascription: people belong to tribe X because they are born in tribe X and are, regardless of personal choice, characterized by the cultural traits of tribe X. Thus one is a member of a 'tribe' not by choice, but by destiny, and one thus partakes of a set of 'proper' customs.

Yet there are three serious problems with this interpretation. First, the mere appeal of, or belief in, a generalized idyllic past and the presumed unity of the ethnic group seem insufficiently definite to explain the relevance to people in specific historical situations of the statements that comprise constructed ethnic ideologies. Why have vague cultural statements about language or a common history or a hero from the past succeeded in 'comforting' people or mobilizing them? Does ethnicity appeal because it is intrinsically 'primordial', or is it constructed as 'primordial' in its discourse to render it more generally appealing? What specific messages within the ethnic ideology actually appeal the most and to whom? And why? In short, the stress upon the 'primordial' aspect of ethnicity tends to overlook both the actual intellectual content of the message, which can vary from group to group, and its varying appeal among different members of the same ethnic group.

Second, by stressing the backward-looking, 'primordial' aspect of ethnicity, this interpretation fails to answer the central empirical question of how the most backward-looking ethnic ideologies, with their glorification of long-dead heroes and their delight in 'traditional values', have been able at the same time to contain within them a powerful acceptance of western education and skills and a willingness to 'change with the times'. The emphasis on the primordial past does not take into account ethnicity's forward-looking aspect which, as commentators have frequently observed, gives it a Janus-like appearance. This is so, I suggest, largely because the role of class actors in creating and shaping ethnic ideologies has been largely overlooked. It is the direct appeal of fresh ideas and institutions to certain new classes that appeared in twentieth century Africa that has been translated into the progressive face of ethnic identity. The psychological appeal of primordialism and the concern for specific present-day interests of specific classes perhaps seem unlikely bed-fellows, but they are real ones nonetheless and must be explained.

Third, and directly related to the first two problems, the emphasis upon a comforting past projects upon African people's ideas an unconvincing stasis. It is simply impossible to accept that Africans, living through some of the most rapid changes that any people have lived through in all human history, have attached themselves blindly, like so many limpets, to a vision of the past that has little relevance to the present and the future just because it is 'comfortable'. As an interpretation, the 'primordialist' explanation of ethnicity, on its own, is simply too ahistorical and non-specific to convince. In analyzing ethnicity's real appeal one must instead try to relate its actual assumptions about the past to the current historical reality of those accepting them.

One may easily conclude then that ethnicity, or 'tribalism', when analyzed abstractly, is Protean, with different appeals on different levels and in different situations. In this respect, it is quintessentially situational and multi-dimensional.

It is thus only common sense to accept that no one explanation suffices to 'explain' it wholly and in every instance. But it is plainly inadequate merely to accept that all interpretations have some elements of truth within them and then try to cobble them together into an intellectual construct comprising elements of each. I would suggest, rather, that moving the analysis of ethnicity beyond the more or less ahistorical stance of the currently dominant interpretations towards a more specifically historical interpretation will shed additional light on both its origin and its continuing appeal for the peoples of southern Africa.

A History

Thus far historians have not devoted much attention to the history of ethnicity and ethnic ideologies in southern Africa. This is somewhat puzzling, especially as many have been aware for some time that ethnicity is not a natural cultural residue but a consciously crafted ideological creation.^[8] It is likely that the explanation for this relative neglect lies in the fact that historians were, like other scholars, caught up in the nationalist paradigm that dominated the entire range of African studies in the 1950s and 1960s. They thus saw studies of the growth of ethnic consciousness as parochial, misconceived, and largely irrelevant to their main concerns at that time: the recovery of Africa's pre-colonial past and the exploration of the growth of anti-colonial resistance and its flowering into progressive nationalism. In the optimistic nation-building mood of the time, studies of ethnicity were also extremely unpopular with African opinion-makers, embarrassing even to mention, and they exerted pressure against studies that might further divisiveness in the new nation states they thought they were 'building'. Thus, the history of ethnic identities largely remained to be written.

The essays in this volume attempt to remedy this situation by placing the study of ethnicity within the unfolding history of a set of societies which are genuinely comparable. This approach is an alternative to the usual one of attempting to analyze the phenomenon on its own, as a subject to be considered sui generis and in a grandly conceived comparative framework. Such abstraction risks removing ethnicity, at least partially, from concrete historical process and blurring specific local factors contributing towards its development and acceptance. The various studies in this volume, when taken together, suggest a basic model that helps us to understand the processes involved in the creation of ethnic ideologies—including those of non-blacks—and sheds some light on why they have had genuine appeal for ordinary people both during the colonial era and in the post-colonial period of national independence.

The area chosen as the setting for these case studies is southern Africa, a region extending from Cape Town to south central Zaire on a south-north axis, and from Namibia to Mozambique on a west-east axis, but with Angola largely excluded. Despite the cultural variety present in this extensive area, it has constituted a coherent regional unit over the past century or so. The event which served as the catalyst for the melding of diverse peoples into such a unit was the discovery of gold on the Witwatersrand in 1886. This initiated the building of Africa's single most potent economic force and attracted capital investment to other, less important focuses of investment, such as the copper mines of Zaire and Zambia, the farms and ranches of Zimbabwe, and the plantations of central Mozambique and southern Malawi. The links that were rapidly constructed to weld together the various territories of this region—and their societies—included ties of finance, trade, political influence, and, especially, migrant labour.

Yet the creation of such ties was necessarily differential, and great variation is

to be found from one area to another within the region. In some places, such as Lesotho, the Transkei, southern Mozambique, northern Malawi, and western Zambia, links with the Rand's mines were direct and obvious: large-scale and persistent male labour migrancy organized through the Witwatersrand Native Labour Association demonstrated clearly the dependence of these regions. In other places, such as the Zambian and Zairian Copperbelts, central Mozambique, and southern Malawi, local capitalist interests were able to dominate and the influence of the Rand was less obvious and less direct. In still other locales, such as parts of central Malawi, southern Zambia, and parts of Zaire and Swaziland, successful peasant production permitted local Africans to avoid both long-distance labour migrancy and working for local entrepreneurs. Nonetheless, the Rand's influence was everywhere present, if only as a model of labour relations and a distant, but powerful, economic presence. Although certainly uneven, the Rand's influence knitted the region's territories together.

As a consequence of the growth of capitalist relations of production both on the Rand itself in the 1890s and in other centres of capitalist endeavour that were established throughout the region shortly afterwards, the people of virtually all its societies experienced pervasive social, economic, and political change. The range of such change was broad, and many of the changes were clearly disadvantageous to the people affected. The capitalist enterprises of the region were all highly labour-intensive, requiring large and constant supplies of cheap African labour. To push Africans into the service of these enterprises, colonial governments imposed taxes, which in many areas could be paid only through men leaving their homes to participate in labour migrancy. These taxes were imposed during, or immediately after, a series of ecological disasters during the 1890s and the early 1900s that greatly weakened the fabric of local African societies. These disasters included drought, locusts and famine, but perhaps the key one was the great rinderpest epidemic of the mid-1890s, which killed livestock through the whole of southern Africa. Because livestock was widely reckoned as the embodiment of wealth, rinderpest's impact effectively constituted a gigantic mass bankruptcy for many societies. Moreover, as the exchange of cattle for women through the system of bridewealth (lobola ) payments was the principal way in which many of the region's societies regulated marriage and the establishment of new families, it became socially necessary to work for money, using it either to restock the herds or as a substitute for cattle in the making of bridewealth payments.

Later, widespread alienation of African land, the establishment of overcrowded 'native reserves', and the entrenchment of patterns of labour migrancy resulted in both impoverished villages and strained relationships within divided families. The labour demands of mines, plantations, and industries, coupled with governmental tax and land policies and the rising needs of people to purchase discretionary goods, pressed men out of the rural areas as workers, especially after the outbreak of World War I in 1914. It should be stressed that this process of rural transformation was not restricted to black Africans. The commercialization of agriculture in large areas of South Africa also undermined an Afrikaner society that had hitherto been characterized by paternalistic relations of clientage between Afrikaner grandees and poor Afrikaner tenants. This commercialization forced from the land white Afrikaners who had long had direct access to it. They moved into the growing cities of South Africa, where, because of their lack of education or marketable skills, they came to constitute a 'poor white' problem of startling dimensions.

What was common for all the region's peoples—blacks and whites alike—was

that many of them were gradually losing control over their lives as control over that most basic factor of production, the land, slipped from their grasp. No longer were rural communities—whether black or white—able to exist autonomously, beyond the reach of capitalism and colonial administration. At the same time that this rural transformation was occurring, the region's mixed-race groups, such as the 'Cape Coloureds' of South Africa and the Luso-Africans of Mozambique, were suffering an erosion of their positions. Earlier, through possession of language and other skills, they had enjoyed relatively secure social and economic positions as intermediaries between whites and blacks. After the 1890s, however, these positions were successfully challenged by poor Afrikaners in South Africa and immigrant Portuguese whites in Mozambique, both of which groups increasingly benefited from the support of racist state institutions. Thus they, like blacks and some white Afrikaners, were also caught in a process of declining control over their lives and destinies.

People of all these groups fought against the erosion of their positions. For many involved in this struggle land, and access to land, came to stand at the very centre of their consciousness, being fixed there not only at the beginning of the process of the undermining of rural autonomy, but also in succeeding decades. For white Afrikaners, land ownership was also important, kept alive as the ideal Afrikaner way of life even among the poor whites of the cities and towns.

For Africans, however, access to land remained a central issue for a more pressing reason. This was because, from the very start of the industrialization process, employers and government officials alike were determined to create a system in which unskilled workers would oscillate from the rural villages to work sites and then back to the villages and in which skilled positions would be held by whites. In this way, their wives and children would remain permanently behind in the rural areas, while the men would dwell in bachelor dormitories at the work sites for the duration of their contracts. Such a system had many advantages for both capitalist entrepreneurs and European administrators. For the employers it helped keep the working class fragmented and unorganized, and it allowed them to pay wages that were less than what would have had to be paid if the whole of a worker's family migrated and settled permanently as fully proletarianized people. For the officials it assured that there would be at least some money brought into the rural areas to help sustain village life there. In some cases, moreover, the migrant labour system also enabled governments to collect capitation fees for each worker recruited.

Migrant labour had less appeal for the workers themselves, but they had little choice in the matter. The need for money and the official pressures upon the men to work as migrants on contract, coupled with the establishment of effective recruiting agencies, resulted in the rapid institutionalization of the system of oscillating migrant labour as the standard mode of labour mobilization. But because the system was one in which workers were to move back and forth, even rural areas that were little more than unproductive rural slums necessarily remained of central concern for the migrants. On the one hand, they could not remain at home to supervise life in the village and oversee their wives and children. On the other hand, they could not abandon their rural homes. Laws prevented the relocation of families to work sites and strictly regulated the length of contracts a worker could assume. Thus, it was in the rural areas that the workers' long-term interests necessarily lay, for they would eventually return there when their working life was over. Even while absent for decades from the rural areas, then, the workers' concerns typically remained sharply focused on what was occurring at home. This situation could not but produce profound

apprehensions in the migrants, and the capitalist era for them was—and still is—truly an age of anxiety.

While the majority of people were affected adversely by the changes produced by industrialization and capital investment, not everyone suffered. Indeed, the establishment of capitalist enterprises and colonial administrations provided a range of opportunities that many whites and some Africans could seize. Certain people were able to respond to the growing markets for produce, becoming peasant producers or even small-scale farmers, while in South Africa Afrikaner agriculturalists on medium-sized and large farms prospered. Others, especially those able to gain an education or useful skills, were able to take up places in the social interstices that the changing economy opened up, becoming relatively well-rewarded teachers, ministers of religion, artisans, government clerks, or even small businessmen. In effect, then, the economic changes that followed on the establishment of the Rand's gold industry and the binding together of the far-flung areas of southern and central Africa into a regional economic unit were accompanied by a rapid and increasingly sharp differentiation of the region's peoples into more favoured and less favoured societies and of the societies themselves into more favoured and less favoured classes-in-the-making.

Such rapid social and economic change eroded earlier political relationships based on clientage both within and outside of lineages, social patterns, and religious beliefs, all of which had characterized societies during the nineteenth century. This erosion in turn opened the way for new forms of consciousness throughout the region. Worker consciousness amongst both whites and blacks appeared spasmodically in situations of localized stress on the work site. Evidence of such class solidarity was shown at times of rapid socio-economic change, appearing in such events as the Rand Rebellion of 1922, the strike of copper miners on the Zambian Copperbelt in 1935, and the African mineworkers' strike of 1946 on the Witwatersrand gold mines. But class consciousness remained exceptional for as long as the working class was weak and fragmented and difficult to infuse with a sense of community.

New types of popular religious consciousness also appeared in the form of mainline Christian churches as well as separatist churches such as Watch Tower and a myriad of Zionist sects, and these shaped their adherents' evolving new self-identities. And among the educated clerks, teachers, clerics, and businessmen who emerged in the black, 'coloured' and mixed race communities a petty bourgeois consciousness, with an acceptance of Victorian notions of respectability, progress and individual uplift through hard work, gained prominence.

It is crucial to the argument of this book that one of the most far-reaching and important of these new forms of consciousness was a new ethnic—or tribal—consciousness that could and did encapsulate other forms of consciousness. Ethnicity could coexist with other types of consciousness without apparent unease because it was cultural and hence based on involuntary ascription, not on personal choice. People were members of a particular ethnic group whether they liked it or not. It was simply a fact of existence. As such, ethnic identity could inhere in both petty bourgeois and worker, in both peasant farmer and striving politician.

A Model

The case studies in this volume suggest a model for the development of ethnic consciousness, or tribalism, in the southern African region. This model stresses the historical creation of such ideologies over time. It is also flexible, in that it can

be detected operating within different societies at different points in time from the late nineteenth century down to the present—and probably into the future—depending upon the impact of a range of variables. It has the advantage of incorporating many of the valuable insights of the various interpretations that have dominated the analysis of ethnicity up to now without being a mere combination of their best points. Finally, it has the additional attraction of suggesting concrete reasons why ethnic consciousness and its near relative, regionalism, were able to develop a special allure for the people of southern Africa during the colonial era. Within the historical perspective, one can begin to understand why it was so easily revitalized after independence, when one-party systems of government were established, and why it can still mobilize popular opinion today, albeit in new ways.

The creation of ethnicity as an ideological statement of popular appeal in the context of profound social, economic and political change in southern Africa was the result of the differential conjunction of various historical forces and phenomena. It is the very unevenness of their co-appearance and dynamic interaction that accounts for the unevenness of ethnic consciousness in the region. One may discern three such variables in the creation and implanting of the ethnic message. First, as was the case in the creation of such ideologies elsewhere, for example in nineteenth century European nationalism, it was essential to have a group of intellectuals involved in formulating it—a group of culture brokers. Second, there was the widespread use of African intermediaries to administer the subordinate peoples, a system usually summed up in the phrase 'indirect rule', and this served to define the boundaries and texture of the new ideologies. Third, ordinary people had a real need for so-called 'traditional values' at a time of rapid social change, thus opening the way for the wide acceptance of the new ideologies.

What emerges perhaps most clearly from these studies is the fact that intellectuals carefully crafted their ethnic ideologies in order to define the cultural characteristics of members of various ethnic groups. These intellectuals could be European missionaries, as the studies by Harries, Ranger, Vail and White, and Roberts make clear. Or, as Harries, Vail and White, Jewsiewicki and Papstein show, they could be European anthropologists and historians. And, in all cases in this volume, local intellectuals—whether Afrikaner, 'coloured' or African—were intimately involved in the process and, where it was possible, they were ready to work hand in hand with their European counterparts.

The role of missionaries was especially crucial in at least one—and sometimes all—of three ways, and it is evident that their influence upon the development of African history in the twentieth century has been far greater than they have been given credit for over the past two decades. First, missionaries themselves were often instrumental in providing the cultural symbols that could be organized into a cultural identity, especially a written language and a researched written history. Samuel Johnson long ago recognized that 'languages are the pedigree of nations', and missionaries accepted this dictum wholeheartedly. They had the skills to reduce hitherto unwritten languages to written forms, thereby delivering the pedigrees that the new 'tribes' required for acceptance. Again and again, in the chapters by Giliomee, Butler, Harries, Ranger, Vail and White, and Penvenne, to mention only some, language stands as the central item in the assemblage of a cultural package. Where these languages have been African languages, it was the missionaries who chose what the 'proper' form of the language would be, thus serving both to further unity and to produce divisions by establishing firm boundaries.^[9]

In addition to creating written languages, missionaries were instrumental in creating cultural identities through their specification of 'custom' and 'tradition' and by writing 'tribal' histories, a process discussed in the chapters by Ranger, Vail and White, and Jewsiewicki. Once these elements of culture were in place and available to be used as the cultural base of a distinct new, ascriptive ethnic identity, it could replace older organizing principles that depended upon voluntary clientage and loyalty and which, as such, showed great plasticity. Thus firm, non-porous and relatively inelastic ethnic boundaries, many of which were highly arbitrary, came to be constructed and were then strengthened by the growth of stereotypes of 'the other', as the essays by Siegel and Papstein show.

Second, and of considerable practical importance, European missionaries, assuming that Africans properly belonged to 'tribes', incorporated into the curricula of their mission schools the lesson that the pupils had clear ethnic identities, backing up this lesson with studies of language and 'tribal custom' in the vernacular. The importance of such education is made clear in the chapters by Ranger, Vail and White, and Jewsiewicki. Thus, mission education socialized the young into accepting a tribal membership, and to be a member of a 'tribe' became 'modern' and fashionable through its close association with education.

Third, and finally, missionaries educated local Africans who then themselves served as the most important force in shaping the new ethnic ideologies. These people—usually men—were keenly aware of the forces that were pulling apart their societies and, with the examples of nationalism in Europe derived from their own mission education before them, they sought to craft similar local movements as a means of countering these problems. Despite their own western-style education, they realized that such a construct would best be understood and accepted if it were put in a cultural idiom easily accessible to the people. Thus, in formulating their new ideologies, they looked to the local area's past for possible raw material for their new intellectual bricolage.^[10] Like their European predecessors during the initial stages of nineteenth century nationalism, they 'rediscovered' the 'true values' of their people and so defined the 'ethnic soul'. Their cultural strongbox was the 'customs' and 'traditions' of the people, identification with which they saw as giving an automatic, ascriptive cultural unity to 'their' people as they confronted the challenge of colonialism and the impact of industrialization. Virtually every study in this volume demonstrates the role of educated people as key actors in the creation of such ideology.

In those societies where missionaries did not work, or where they did work but did not introduce education along western lines, or where African intellectuals emerged only at a late period or not at all, the development of ethnic ideologies was either stalled or never occurred. The unevenness of education in southern Africa largely determined the unevenness of the development of ethnic consciousness, as is brought out in the essays by Vail and White, Macmillan, Siegel, and Papstein. In many locales it is only today, after the post-independence expansion of education and the emergence of local intellectuals, that the process of creating such ethnic ideologies and 'forging traditions' has emulated what happened earlier in other societies.

It was not sufficient, however, that there should be local intellectuals—white or black—interested in the recovery of the ethnic past. A second, more instrumental factor was also required, and this is explored most fully in the chapters by Harries, Vail and White, Marks, Jewsiewicki, and Papstein. All of southern Africa was under direct European administration of various types, and by the period after World War I, virtually all administrations were engaged in implementing systems of indirect rule, using African 'traditional' authorities as intermediaries between

the white administration and the ruled. Thus, if language in the form of written discourse was central in specifying the forms of culture, indirect rule provided the institutional framework for articulating these forms. Communication between the European administrators and subordinate Africans was distinctly tribal in its tone and content. Africans were talked to in terms deemed suitable, and these terms were ethnic. In the cases of the 'Cape Coloureds' and the Luso-Africans of Mozambique, and, to some extent, the Afrikaners, for whom the conventions of indirect rule were not suitable, they were simply denied representation, as the studies by Goldin, Penvenne, Giliomee and Butler indicate.

There were several reasons for the European policy of indirect rule. First, there was the realization that the use of so-called 'traditional' African leaders could be markedly less expensive than the employment of expensive European officials. Second, administrators assumed that Africans were naturally 'tribal' people. If the natural ethnic units could be strengthened, it would help ensure their continuation as discrete 'tribal' groups and prevent the emergence of 'detribalized' Africans of whom whites were deeply suspicious. This, in turn, would slow the emergence of any potentially dangerous territory-wide political consciousness that might develop. The remarks of a British War Office official in 1917 reflect these divide-and-rule tactics:

[The] spirit of nationality, or perhaps it would be more correct to say, of tribe, should be cultivated and nowhere can this be done with better chance of success than in British East Africa and Uganda, where there are numerous tribes ethnographically quite distinct from one another. It is suggested that in each ethnographically distinct district the schools should, as far as possible, form integral parts of the tribe and centres of folk-lore and tradition. . . .

. . . a method may also be found whereby the efforts of missionaries may also assist in the cultivation of national spirit. This it seems might be done by allowing only one denomination to work in each demographic area and by not allowing the same denomination to work in two adjacent areas.^[11]

Third, by the end of World War I it was becoming increasingly evident that the chronic absence of men from rural societies was producing great social stresses. The administrators became convinced that the rural disintegration occurring before their eyes could be slowed, if not stopped, by the encouragement of 'traditional authorities' to use 'traditional sanctions' in exercising control over the rural areas to counter the forces of social decay.

This acceptance of indirect rule by European administrations obviously gave opportunities to African political authorities to augment their personal power. More importantly, I suggest, it gave opportunities to the intellectuals of the areas concerned—both European missionaries and African members of the educated petty bourgeoisie—to implement their ideological programmes through alliances with the newly recognized chiefs. This process is a theme of several essays in this volume, including those by Vail and White, Marks, Jewsiewicki and Papstein. In this way the cultural ideals contained in their new ideologies could be at least partially actualized in the day-to-day workings of African administrations under indirect rule. Ethnic identity, thus, came to be specified not only by the written histories, grammars, and accounts of 'traditional customs' produced by local culture brokers, but also—and in many respects, far more importantly—by the actual operation of the administrative mechanisms of indirect rule. This aspect of the development of ethnic identity was the consequence of the dynamic interaction of African initiative with the expectations of European administrators and forward-looking missionaries. It should be remembered, however, that the

subordinate peoples did not have a free hand in their work as they had to operate within the severe constraints imposed by racist administrators who were ever alert to check initiatives deemed either unseemly or dangerous, something brought out clearly in the chapters by Marks, Macmillan, Jewsiewicki and Siegel, among others.

The presence of intellectuals, the socialization of ethnic ideas through mission schools and through the actual operation of administrative systems under indirect rule to strengthen 'tribal' rule were, however, by themselves inadequate to produce a broad acceptance of an ethnic ideology. The ideology itself needed a raison d'être and an appeal, and it was this appeal that constitutes the third factor in our model of the growth of ethnicity in southern Africa.

The ideologies of nationalism have often been described as 'Janus-like'. They are in one aspect profoundly reactionary, looking backwards to a Golden Past: they concentrate upon its heroes, its historical successes, and its unsullied cultural purity, and are decked out with the mythic 'rediscovered' social values of that past. In Africa, the explicit association of such ethnic ideologies with chiefs and headmen whose position was often firmly rooted in the past was an additional factor in accentuating the backward-looking face of ethnicity. Yet these ideologies were also clearly products of the present, concerned with current conditions, and they typically exhibited a forward-looking concern for the future. Nationalism—and tribalism—have thus appeared uncertain and ambiguous to many observers.

Yet when one looks closely at the situation in southern Africa, one comes to realize that the ethnic message's backward-looking aspects and its forward-looking concerns have been in no way contradictory. The emphases on past values, 'rediscovered' traditions, and chiefly authority were truly conservative—that is, they were calculated to conserve a way of life that was in the process of being rapidly undermined by the forces of capitalism and colonialism. Forward-looking members of the petty bourgeoisie and migrant workers alike attempted to shore up their societies and their own positions in them by embracing ethnicity and accepting tribal identities.

As the chapters by Vail and White, Marks, and Jewsiewicki show, ethnicity appealed to the petty bourgeoisie because its forward-looking aspects ensured them a leadership role in the newly defined 'tribe' as the well-informed interpreters of 'tribal tradition'. Their position as allies of chiefs further legitimized their role, blunting consciousness of the class divisions that were then appearing in local societies. In this situation, it was generally accepted that they also had a duty to improve their own social and economic positions 'for the good of the tribe'.

Far more importantly, ethnicity appealed strongly to ordinary African men, not primarily because it gave them a sense of psychological comfort, as the primordialist interpretation argues, but because it aided them in bringing a measure of control to the difficult situations in which they found themselves in their day-to-day life. The word 'control' is crucial. It was the element of control embedded in tribal ideologies that especially appealed to migrant workers, removed from their land and families and working in far distant places. The new ideologies stressed the historical integrity of the tribe and its land and, especially, the sanctity of the family and its right to land.^[12] Land stood at the very centre of ethnic ideologies.

The place of women was also a central issue dealt with in ethnic ideologies. In the early decades of the century bridewealth steadily inflated in value, and women thus represented a greater 'investment' by men in cattle or money. With most men absent as migrant labourers, women were also becoming more important to the day-to-day survival of the family through their work on the land. Yet such

valuable women naturally often sought to act independently, even to the extent of seeking divorces or leaving the rural areas illegally to move to industrial and urban areas. This produced acute conflict between the genders. Therefore an emphasis on the need to control women and a stress on the protection of the integrity of the family came to be intrinsic to both ethnic ideologies and the actual institutional practices of indirect rule. The studies by Vail and White, Marks, and Jewsiewicki show this with special clarity. Ethnicity's appeal was strongest for men, then, and the Tswana proverb to the effect that 'women have no tribe' had a real—if unintended—element of truth in it.

Ethnic ideologies helped to provide the control necessary to minimize migrants' natural anxieties about what occurred at home. In the system of indirect rule, the chiefs were of central importance. It was they, with their new official histories, their new censuses and lists, their new courts and records, all of which employed for the first time that most fundamentally powerful invention, writing, who were now able to exercise a greatly increased degree of surveillance over both women and land in the absence of the men. It was they who brought into daily practice those 'rediscovered traditions' which emphasized control in the name of 'custom'. The old dictum that 'all politics is local' was especially valid throughout southern Africa. African men and their lineages accepted that it was in their essential interest to support the new structures of chiefs, their courts, and their educated petty bourgeois spokesmen and agents. It was also for this reason that men, when returning at the end of their contracts from the mines or farms or plantations, gave chiefs the gifts that constituted one of their most important sources of income. The good chief was a proxy who protected the interests of the migrant workers and, for that, they were ready—if not eager—to reward him materially. In effect, the bureaucratized chief of the newly constituted 'tribe' had replaced the lineage head or independent patron of earlier times, and the old language of kinship came to be employed as metaphor to sustain and legitimize this new, obviously non-kinship relationship.^[13]

It was for very real reasons of exercising at least a measure of control over land and women, thereby bringing at least a measure of peace to their minds, that African men welcomed the new ethnic ideologies which involved augmenting powers of chiefs in a situation of rapid social decay. Ethnicity, insofar as it was a mechanism of such control, may be interpreted, then, as a form of popular male resistance to the forces that were reshaping African lives throughout southern Africa. It was for this reason also that the appeal of ethnic ideologies was strongest amongst those who were migrant labourers. The ethnic identity that was rooted in the realities of the countryside was, rather incidentally, strengthened in the workplace, where migrants found themselves in the company of, and often in competition with, workers from other cultural groups, a situation which generated sets of largely negative ethnic stereotypes and is explored in the chapters by Jewsiewicki, Siegel and Papstein.

Men came to think of themselves as belonging to particular ethnic groups, then, not because they especially disliked their fellow workers, nor because being a member of the group made them feel good, but rather because the ethnic apparatus of the rural area—the chiefs, 'traditional' courts, petty bourgeois intellectuals, and the systematized 'traditional' values of the 'tribe' as embodied in the ethnic ideology—all worked to preserve the very substantial interests which these men had in their home areas. Without ethnicity—or tribalism—the migrants would have been less able to exercise the control that was necessary for them to assure the continuation of their positions in rural societies and their ultimate retirement in their home areas.

In those situations in which labour migrancy was not a pressing reality (the Afrikaners, the 'Cape Coloureds', the Luso-Africans of Mozambique and, to a lesser extent, contemporary Swaziland and Ciskei—see Giliomee, Butler, Goldin, Penvenne, Macmillan and Anonymous, respectively) or in areas from which men did not emigrate in large numbers, such as southern Zambia and central Malawi, the ethnic message has clearly had less popular appeal, reaching no further than the petty bourgeoisie in most cases. In the case of the Afrikaners effective class alliances between the bourgeois elements of society and the 'poor whites' were brought into being only in the 1940s and afterwards. In the case of the 'Cape Coloureds' and the Mozambican Luso-Africans—and possibly Swaziland and the Ciskei—the gaps between well-off and poor were too great to be easily overcome by appeals to ethnicity, as is suggested in the chapters by Goldin, Penvenne, Macmillan and Anonymous. In these situations, class identity—or at least class tension—has tended to overshadow ethnicity.

The Situation Today

For large areas of southern Africa, independence came in the 1960s and 1970s. But the condition which stretches basic economic, familial and welfare concerns between rural residence and work site endures down to the present. Migrant labour is still a dominant form of labour mobilization throughout the region, and the mental attitudes intrinsic to it continue. Even in situations where men have been permanently resident in the urban areas with their families for decades, these attitudes are widely found. This is so not because Africans are inherently rural people or are in close harmony with Nature, but because housing and living expenses are far lower in the rural areas than they are in urban areas. This lower cost of living serves as a constant reason for those dwelling in urban locales to keep the rural areas always in mind and to view their urban sojourn as only temporary.^[14] Thus, because at the end of one's period of employment retirement benefits are usually given in the form of a single lump sum of money rather than in monthly payments, if they are given at all, a person—whether unskilled migrant or educated white collar worker—has little, choice but to return 'home' to live out the rest of his days, spending as little money as possible.

The preoccupation with one's connection to the land has been overwhelming, with virtually everyone either possessing a piece of land in actuality or desiring it in his or her fantasies. This continuing fixation on land, I suggest, has resulted from decades of the existence of an oscillating workforce that has only partially proletarianized workers and from the failure to establish the sort of welfare measures that would support a fully urbanized population after retirement. The concern for land as an ultimate fall-back means of survival is clearly an economic concern, then, and, in the circumstances, it is quite understandable. Even in South Africa, the most industrialized of all the countries of the region and one in which complete proletarianization on a substantial scale in secondary and tertiary industry has existed at least since the 1940s, lack of adequate welfare and retirement measures keeps alive deep concern about access to land. Thus the African National Congress still finds that to contend publicly that one of the fundamental roots of the political conflict between black and white in South Africa is the Native Land Act of 1913, and to talk about a land reform that would give dispossessed blacks renewed access to land, have great appeal to their constituency.

Added to this economic concern is the fact that the nation states that have appeared since the 1960s have suffered profoundly from economic weakness, a

weakness which has grown more serious with each passing year. Quite naturally, this has had a negative effect on the possibility of the creation of broad loyalty to the nation state itself. The nationalist message before and immediately after the end of colonialism was that the new dispensation would result in economic improvements and much increased welfare benefits. Unfortunately, this progress has not occurred, and instead the nation state's administrative structures have faltered and shrivelled.

There are thus further economic reasons why sentiments which would be described as 'nationalist' do not converge with citizenship in a new nation state, as it has come to be identified as at least the occasion, and sometimes as the cause, of a declining standard of living for the majority of people. People perhaps accept that they are citizens of the country in which they live, but this acceptance of civil status does not produce the same loyalty as does their ethnic identity. There has therefore been an increased concern with ethnic identities over the past two decades, and with it has come a great acceleration in the 'rediscovery' of culture for more and more ethnic groups, as the essay by Papstein explores in detail.

For economic reasons, therefore, as well as for reasons of psychological satisfaction, it seems clear that ethnic loyalties will continue in southern Africa for the foreseeable future. The exact forms of future ethnic identities are still cloudy, largely because conditions related to certain variables have changed since the development of ethnic consciousness in the colonial period, a process which has for the most part provided the model used in this volume. Education, for example, is now almost wholly under the control of the nation state, and, hence, will not be as easily employed to bring about acceptance of specific ethnic identities among children. In some countries—such as Malawi and Swaziland—the chiefs remain as influential figures in the rural areas. In others, such as Zambia, the chiefs remain, but most of their power has been taken from them. In yet others, such as Mozambique, chiefs have been abolished totally. Therefore it is likely that the symbols of ethnicity will vary from place to place and from country to country depending on the nature of local government and the way the state communicates with ordinary people.

Furthermore, the potential culture brokers are far more numerous now than sixty years ago, and they have been exposed to a far wider variety of thought, usually not associated with missions. This means that while the backward-looking aspects of future ethnic phenomena—concern for the glories of past history, culture heroes, the central importance of language, and the like—will remain pretty much the same as for examples in the past, the forward-looking aspect of the Janus of ethnicity has the potential of wide variation across the political spectrum. In contemporary Zambia, for example, a main focus of ethnic identity for the Bemba-speaking people who see themselves cut off from state power is the predominantly Bemba miners' union.

The unevenness of development that has marked southern Africa since 1886 shows no sign of ceasing now. Therefore it is likely that the content of the ethnic message itself will continue to vary from people to people, as the culture brokers craft messages that will resonate with their own clienteles. For the serious student of political history in the region, then, it will not be adequate to approach ethnicity, or 'tribalism', as if all examples were essentially the same. Concern with the content of the message will be of ever greater importance if we are to understand it.

Finally, as ethnicity and parochial loyalties within the borders of nation states are likely to continue, it is important to cease approaching them from the perspective of the nation state itself. Ignoring them as embarrassing

epiphenomena that should have long ago disappeared will do no good. Condemning them as 'reactionary' or 'divisive' will accomplish very little. Instead, granted that it is virtually certain that the nation states of southern Africa are going to continue as institutionalized governing states in tension with those whom they govern, it will be necessary for the region's politicians and scholars alike to work towards accommodating ethnicity within these nation states.^[15] States like Lesotho and Botswana, where the nation state and ethnicity are largely coterminous, are exceptional. Multi-ethnic states like Mozambique and Zaire, Zambia and, most crucial of all, South Africa are typical. The western model of the nation state which sees it as identical with the cultural nation itself simply does not obtain in such situations and to insist upon its superior claims to legitimacy and loyalty is simply myopic. Instead, accepting that ethnicity does exist as a potent force, Africans will have to produce political solutions derived from African experience to solve African problems, and this is clearly of great importance in the evolving situation in South Africa, the political and economic centre of the region.