Custom and the politics of sovereignty in South Africa.

Four eras of tribal and state formation have marked the modern history of South Africa. The conquest years of the nineteenth century saw the emergence of a colonial state and the reworking of ethnic identities tied to tribal political structures within an imperial context. The 1920s witnessed the rise of segregation and "retribalization" as set out in legislation such as the 1920 Native Affairs Act and the 1927 Native Administration Act. This and other legislation further bureaucraticized state administration of Africans and moved the country towards territorial segregation in what one scholar has described as a system of "decentralized despotism." (1) In the 1950s, with legislation such as the 1951 Bantu Authorities Act, the third era began with the Bantustan policies of grand apartheid, wherein the former reserves would become sovereign nation-states. Apartheid was many things, but most obviously it was a system of tribalist social engineering and bureaucratic authoritarianism.

The fourth era of ethnic and state formation began roughly a decade ago with South Africa's first democratic elections. The new government inherited colonial tribal structures, some well over a century old. Politicians were acutely aware that reconstructing the state in the former homelands would be a formidable undertaking. They inherited a weakened economy, diminished state resources, and a host of seemingly intractable problems. Considerable political instability and continued institutional collapse marked the Transkei and other rural areas. Just a few years after the 1994 elections, the government considered calling in the army into parts of the Transkei to restore order, a clear indication of political involution and the tenuousness of their grip on the former homeland. In one part of the former homeland, none other than Chief Matanzima, the former and considerably-hated Bantustan leader, headed a tribal court where people appeared before him as "Transkei citizens," citizens of a polity that no longer existed. Apartheid, it seems, has had many deaths, some rather more drawn out than others.

The situation in Kwa-Zulu/Natal posed acute challenges to the African National Congress. The 1994 political settlement between the ANC and the Inkatha Freedom Party, in which the ANC in effect conceded the ill-begotten IFP victory, also resulted in a recognition--however begrudgingly--of the political salience of tribalism in rural areas of the country. Earlier, in the late 1980s, "traditional" rulers had formed the Congress of Traditional Leaders of South Africa (CONTRELESA). In the 1990s, during the negotiations for a new constitution, CONTRELESA lobbied for an important role of traditional authorities within the new South Africa. More generally, polling seemed to suggest that there existed widespread support for "traditional rulers" in the former homelands. (2) Given these winds of tradition blowing across South Africa's political landscape, when Matanzima died on 16 June 2003, President Mbeki and numerous dignitaries attended the funeral, including Chief Buthelezi and other traditional rulers. Mbeki delivered the eulogy to more than five thousand mourners. Once funerals had been important sites of political imagining and resistance to apartheid. Now they have become crucial moments in the reconstruction of tribal pasts and the assertion of tribal presents. (3)

As a result of these developments and the considerable challenges facing the country, the ANC government has embarked on the most important reworking of rural political structure since the implementation of apartheid in the 1950s. There are new policies and on-going discussions concerning custom and "traditional authorities" and the transformation of communal land tenure. The renovation of custom culminated most recently in Act 41 of 2003, the "Traditional Leadership and Governance Framework Amendment Act," signed into law by President Mbeki in December 2003. The act grew out of the deliberation over the relevance of custom and traditional rule, instability within the former homelands, and struggles over the reworking of local political institutions. The legislation establishes the processes for reworking traditional rule and for providing "traditional communities" with limited sovereignty. Under the legislation, "traditional councils or traditional leaders" may have a role in issues ranging from land administration, health and the administration of justice, to economic development and arts and culture. (4) In short, the government has agreed to provide some areas of the country--in effect the older tribal reserves and the former homelands--with a degree of sovereignty. And they have provided that degree of sovereignty not to democratically-elected officials but to so-called "traditional authorities."

Given these developments, and particularly given the powerful claims that are being made on the past, this may be an auspicious time to revisit briefly the histories of culture, custom and power in South Africa. The South Africa situation may be peculiar, but it is not unique. Large swaths of the world experienced similar twinned colonial histories of state and tribal formation: parts of the United States, South Asia, and of course much of Africa under the policies of Indirect Rule. The contemporary moment has seen a re-assertion of ethnic identities, for example in Nigeria and throughout much of former Soviet Central Asia, particularly in the context of decreasing state capacity and the emergence of spaces of considerable ungovernance. In South Africa, Nigeria, and elsewhere, politicians confront the specter of ethnic assertion and the reality that many people consider traditional rule--however they define it--as something that should be preserved and implemented.

Much of the politics of traditional rule centers on the question of political sovereignty, especially historical claims around how sovereignty was lost and how it is to be reclaimed. Classically, the modern doctrine of sovereignty rested on the idea that someone was in "supreme command" and presumed the existence of a state exercising dominion over a given bounded territory. (5) The sovereign political community was an autonomous community and, quite crucially, from sovereignty flows the issue of political rights. Sovereignty lay at the center of modern state formation and, indeed, of the international system itself. Modern political theorists conceived of the indivisibility and independence of the state. By the nineteenth century the "Westphalian regime" had become firmly ensconced within European international law, at the center of which was the sovereign state. With the rise of the modern nation-state various thinkers presented sovereignty as an ontological given and as territorially defined, so that the field of international relations, for example, was seen as the interactions (often violent) of various sovereign states. (6)

What if we refuse to take sovereignty as ontologically given, but instead inquire into its history in a specific colonial setting like South Africa, asking, as well, how it is that historical actors have and continue to make such powerful appeals to what may in fact be a fiction? Foucault famously if elliptically remarked of the importance of cutting the king's head off, by which he meant revealing sovereignty for the fiction that it is as a way of exposing the dispositions of power in the modern world. (7) The issue of sovereignty, the ways power is bounded and exercised, was intrinsic to state formation as opposed to something taken for granted as a political "given." Sovereignty formed an important part of claims to power and the coming into being, as it were, of the state and, more generally, of political rights. Put another way, sovereignty is an effect of the state, one of the "categories of thought produced and guaranteed by the state," as Bourdieu would have it. (8) More generally Bourdieu has argued for the importance of understanding the ways "the state establishes and inculcates common forms and categories of perception and appreciation, social frameworks of perceptions, of understanding or of memory, in short state forms of classification." (9) What is less clear in both Foucault's and Bourdieu's formulations is the extent to which state effects, and indeed state formation itself, might be usefully seen as negotiated, and that the naturalization of the state's classificatory schemes--including sovereignty as a kind of political monotheism--emerged out of historical processes of negotiation as well as conflict and confrontation.

What is striking in some of the new literature on state formation is the absence of any sense that state formation is negotiated, albeit in conditions in which differentials in coercive power could be very great. (10) States certainly make history, through the myriad actions of officials and quite literally in the production of official histories. But they seldom do so in conditions of their own choosing. This is especially so in colonial state formation that entailed, simultaneously, cross-cultural encounters with the extension of power and the creation of political institutions. (11) Social histories of state formation offer the promise of understanding these negotiated pasts. (12) In so doing, they may deepen appreciation of the histories of power and rights in South Africa and elsewhere in the colonial world.

Conquest and the Inventions of Sovereignty

An earlier generation of Africanist historians concerned themselves with establishing the existence of African states in the era before European colonial conquest. Any number of books and articles took up the issue of the state, kingship, royal prerogatives, and so on. The reasons behind this search for the state are complex and can only be touched on here. One reason certainly centered simply on the fact that many historians either implicitly or explicitly deployed the category of the state, or more precisely the nation-state, as a way of organizing historical understanding. A second reason was perhaps more political, an attempt to present on the stage of world history the fact that pre-colonial Africa also had centralized polities with strong rulers and that, therefore, the African past could not be dismissed as backward and without history. (13)

South African history, where the political issues were especially acute, reproduced these general patterns. One aspect of white supremacy represented African society as anarchic, barbarous, and lacking centralized political institutions. Or, if there was a state, it was a tyrannical one. The accomplishments of early Africanist history largely dispelled these prejudices mainly by reconstructing the political history of various ethnic groups. By the end of the 1980s scholars had amassed a considerable literature organized around the rise of states or on the history of given "kingdoms," proving, in effect, that Africa had a sensible history. In many respects this literature could be described as a kind of "rise and fall" of states, the rise of states a mark of pre-colonial African dynamism, the fall the result of European colonialism. States were usually seen as basically coterminous with ethnicity, hence for example the subtitle of one book was the "Xhosa Kingdom." The idea of territory, the spatial embodiment of the state, was usually taken for granted. (14)

The literature bears resemblance with the reporting and ethnographic representations of colonial officials, whose documentary record constitutes much of the written archive. They organized their field of representation into that which appeared intelligible and important and that which was irrational or superstitious. These officials, as Berry has written and as we shall see below, "set out to discover the boundaries and customs of 'traditional' communities and the 'original' relations between them." (15) The assumption was that political identity, rights, and political space are, or at least should be, coterminous. One job of officials was to organize knowledge of the important and the intelligible and then to institute policies that made the normative assumption the lived reality of colonial rule. (16)

This comity of official and academic representation is not surprising. Europeans measured the African world according to their own epistemologies, separating what seemed based on reason from the unintelligible and irrational. Kings and hereditary succession made sense; they provided a ready way of understanding place and time. Witchcraft, magical beings, wise lions that speak to people, and so on, made no sense, and because they made no sense served no purpose in organizing the historical past into an intelligible narrative. Likewise, for historians easily translatable concepts provided coherence to the African past. As Feierman has written in a different context, the silences produced in conquest endure in historical narrative. (17)

We now know the situation was rather more complex. Even in the case of Zulu kingdom, the boundaries of the kingdom were far less clear and far more unstable than many Europeans believed or many Africans represented; the space between the projection of Zulu power and its reality continues to be a major issue in the Zululand/Natal region. Consideration of what was once dismissed as irrational, marginal and transitory is beginning to offer new ways of recounting the African past. (18) It also suggests the centrality of colonial conquest and early colonial rule in the making and remaking of social reality.

In the Eastern Cape, early European visitors very often had a difficult time "discovering" a chief, much less a king. Considerable evidence suggests that up through the late eighteenth century there is neither the evidence of, nor the preconditions for, the rise of centralized polities in the Eastern Cape and, in fact, throughout much of what would become South Africa. To speak of a Xhosa "kingdom" is to engage a falsification and reification, reducing a linguistic category to a political order that did not exist. To the extent to which there was any political centralization, it centered on chiefs (inkosi) (19) of which, by the end of the century, there were probably over a hundred who claimed this mantle in the Eastern Cape alone. Political power tended to be localized, boundaries fluid and vague, and the authority of chiefs highly variable. The political landscape was both homogeneous and kaleidoscopic, with widely dispersed material and symbolic resources and constantly changing political domains. Even at moments of relative stasis domains of authority very frequently overlapped. (20) Political identities were multiple, with the fluidity of identities generally increasing with geographical distance from any given center of power. Domains were reworked at the death of a chief. Succession disputes were notoriously common. The "rule" was that the son of the great-house would inherit the chiefship, but this was not always the case. Moreover, all of the sons of a chief could make claim to being "inkosi" by striking out on their own and emphasizing their chiefly descent.

From the late eighteenth century, and especially in the nineteenth century, successive "great chiefs" had attempted to extend control over outlying areas. Instrumental power, however, dissipated very rapidly from their kraal. Outlying households might have some sense that one chief might be greater than another, but this recognition operated largely symbolically. In the nineteenth century especially, "great chiefs" attempted to make palpable various claims to rule; their attempts usually met with little success.

Their failures stemmed from a number of reasons. The absence of any unequal distribution of economic goods, trade, or population mitigated against the centralization of power. Second, military technology and strategy were widely democratic. Third, there were multiple nodes and overlapping domains of authority. Chiefs, for example, laid claim to the idea they could ensure rain and agricultural fertility through their relationship with autochthons, a perilous claim given the instability of the rainfall regime in the area. There were, however, numerous ways by which people could access symbolic power that circumvented chiefly claims. People could, for example, enlist the support of autochthons on their own. (21) Recognizing the authority of chiefs could shift quite rapidly given the frequently overlapping domains, and there were more localized nodes of authority based on descent that could stand in significant opposition to chiefly pretensions. This made any given territory appear to be "multi-ethnic." More generally, chiefs could lose their authority, for a variety of reasons ranging from drought through forms of protest by which people "voted with their feet."

Europeans and especially early colonial officials very often found African polities to be exasperating and scarcely intelligible. One thing seemed reasonably comprehensible, that is most easily translatable into their own political epistemologies: that there were some men of elevated status who wore and laid claim to the skins of leopards and lions. These men often practiced polygyny, lived in larger communities, usually possessed more livestock than others, and were referred to and used the title "inkosi," but beyond that seemingly little differentiated chiefs from most everyone else. Particularly challenging was determining the relationship between status and territory. Yet, identifying African rulers formed an important part of the work of European military leaders and colonial administrators. This followed from the idea that to assume sovereignty required at minimum some location of it in the African context. European administrators came with the idea that there was--or should be--a person who had authority within a specific territory. There might be border disputes between two or more sovereigns, but these were the concerns of those involved.

Transposing Westphalia to the Eastern Cape and indeed elsewhere in Africa was more difficult than it might at first seem. A few sovereigns seemed relatively easily identifiable, particularly those chiefs who claimed the mantle, and were recognized by some as being, of paramounts or "great chiefs," though the ostensible domain of their control was exceedingly unclear.

Conquest and early colonial state formation permanently altered this complex pre-colonial landscape of power, authority and identity. Three developments are especially notable. First, ideas on and around sovereignty came to be fused to what was seemed most comprehensible to Europeans: hereditary chiefship. Among other things this silenced alternative forms and nodes of authority and powerfully shaped the colonial production of knowledge into that which was intelligible (hereditary chiefship) and that which was irrational (especially witchcraft). This brought chiefs to the forefront of the making of the colonial order. Second, political authority became more fixed in the figures of individual people. This nicely dovetailed with the preoccupations of the British, who where concerned with identifying sovereigns and their areas of jurisdiction. Finally, chiefs and paramounts and those who laid claim to office typically attempted to both solidify and extend their control by using the semantic ground of the colonizer to their best advantage.

In the 1870s, for example, in the context of expansion across the Kei River colonial officials tried to make sense of the relationship between paramounts Kreli and Gangelizwe and what is glossed in the archival record as the Galeka and Thembu tribes. Jurisdiction became a burning issue. In 1871-2 war broke out between Kreli and Gangelizwe as the result of the attempts by each of the "great chiefs" to make real their claims to rule certain territories. Officials believed that much of the "safety of the Colony" depended on resolving the competing claims of the two chiefs. (22) The conflict erupted at precisely the same time officials were addressing the "boundary question" in the region. (23) The officials concluded "that all that has taken place may be traced to a dispute as to the ownership of the coast country east of the Bashee" River. (24) The officials, however, recommended hedging on the issue, for "although we would decide that Gangelizwe has no right in that country we do not propose that Kreli's claim to it should be permitted by Government. We would say that Government knows of no authority possessed by Kreli beyond the Bashee." (25)

These sorts of issues emerged across the Eastern Cape, as they did throughout Africa in the early colonial period. Conquest and early state formation created a far more competitive political environment. Africans who laid claims to chiefship fought against others with regard to territory. They also sought to instruct colonial officials as to what comprised their sovereign territory. At the same time, there was the near-obsession of colonial officials to fix power spatially and bodily--identifying rulers and their areas of control--and, in the very same breath, the avowed aim to destroy chiefship. It was, as it were, important to bring clarity to the sovereignty issue as part of the very process of taking it away. In short, sovereignty was invented as an ineluctable part of colonial conquest because of the necessity of identifying the political subjects of empire. The colonial state giveth and taketh away; indeed the process was very much a part of the colonial state's very formation. In Gangelizwe's case, and indeed in the case of most other paramount chiefs, the officials proposed that his "authority should not extend beyond his own section." (26) Pretensions to kingship were dashed on the rocks of colonial conquest.

Colonial officials had little time for paramount chiefs, these men who would be king, and found that their pretensions frustrated colonial efforts at bringing clarity to the sovereignty issue because they generated overlapping claims to territory that seemed to create more, not less diverse, populations. At the same time there was an often frantic jockeying for power on the part of men who represented themselves as being a chief. Many of the same issues that had unfolded in the case of paramounts Kreli and Gangelizwe unfolded in the realm of chiefly politics. Chiefs represented to colonial officials the areas under their control and historicized their rule by deploying an ideology of descent. On the one hand Africans themselves argued that they were chiefs and that they were so on the basis of descent; the person of the chief embodied the political principles of the tribe. Elided in this representation was the issue of legitimacy and that the principles of authority were not necessarily synonymous with the body of the chief. In the precolonial period, for example, chiefs who failed to bring rain in effect ceased being chief, or people simply voted with their feet. The issue of genealogy might be important, but politics could not as it were be reduced to a schematic. But a schematic was precisely what colonial officials wanted. Moreover, and importantly, the men who would be chief knew that Europeans could not fathom African epistemology and so they discussed politics on a semantic ground intelligible to, and indeed desired by, the official mind.

From this schematic flowed political rights based in the idea of residence, indigeneity and ethnically pure administrative spaces. Europeans confronted a situation in which authority was not circumscribed by bounded space. As the Chief Magistrate complained to a chief in 1884: "It is impossible that a Chief can be responsible for the conduct of men in his ward, who don't acknowledge or obey him.... It is quite impossible that either judicial or fiscal administration can be satisfactory so long as the several sections of tribes ... remain intermixed as at present." The chief had pointed out the fact that there was "no fixed boundary" and that people were "intermixed." He argued that there were people living eighteen miles from his kraal that were under his authority, but that the people living in the "space between us" belonged to two other chiefs. (27) In other words, space and rule did not overlap as predicated by the modern conception of sovereignty.

The answer to these predicaments was of course to define some people as residents and others as strangers or, in the colonial parlance of the time, "squatters". Residence rights and the schematic of authority would thus become coterminous; in short the state created bounded territories based on ethnicity and defined them as once sovereign. Various resident magistrates and especially the 1872 Commission set out creating districts and sub-districts. Magistrates began removing "squatters" to their respective districts. At the same time they made it clear that the residents' rights rested with the chief, magistrate, and headmen of their district, and not with someone who might be, as it were, some distance from their location or even in another district.

In the late 1870s, for example, there was lengthy discussion around the issue of "Pondo and Pondomise Squatters, who were reciprocally occupying each other's ground" in the area that was becoming Qumbu district. The officials struggled to have each removed to "within the boundaries of their respective territories" as set out in the 1872 Commission which "fixed" the boundaries of the various districts and sub-districts. Qumbu was defined as a Mpondomise district, and so the magistrate "accordingly took steps to remove the Pondomise from Pondoland into my District," and vice versa. The Mpondo squatters were forced to vacate some "twenty two villages" which were then re-occupied by Mpondomise "who had been living in Pondoland." As the magistrate summarized the developments,</p> <pre> it may fairly be taken as an indication that the two tribes concerned .... are not prepared at present to shew any serious disinclination to obey the orders of Government .... there was a strong wish on the part of the Pondos that their people should remain there for various reasons; and as regards Mhlonhlo [the Mpondomise chief] I believe that if he did take any part in the negotiations Mr. Davis set on foot, he did so with the object of establishing a precedent ... (28) </pre> <p>In fact the chief had accompanied the officials when they had set about their work of establishing boundaries. This was not at all unusual. People were intensely concerned about the emerging colonial organization of space. Chiefs and others, especially lineage heads, accompanied commission members and magistrates, making claims to territory and, in effect, providing one explanation of the political landscape that made sense to Europeans.

The earliest information has the tenor of ethnography. Officials produced genealogies and narratives of chiefly descent in their attempt to understand African political society and, especially, as part of their efforts to create ethnically pure magisterial districts. The law of succession, as they saw it and as Africans themselves typically represented, was that the son of the "right hand house" inherited the position of chiefship. This principle in effect became enshrined practice. Evidence to the contrary--succession not following the line of the "right hand house," chiefs being deposed, commoners becoming chiefs, and so on--was usually simply explained away or at least not engaged with in any substantive way. The other sources of authority ended up on the cutting-floor of colonial state formation.

Early colonial officials used genealogical reckoning in their attempts to bring some clarity to the political landscape. Early colonial power was weak. The initial position of the Native Affairs Department was that resident magistrates were to rule Africans "principally through their own Chiefs and in accordance with Kaffir laws and customs." (29) Therefore, resident magistrates "exercise[d] judicial functions by the grace of the Paramount Chief." (30) According to one legal opinion at the time, officials thus had "no legal right to exercise jurisdiction." (31) On the one hand, it was as if the department was simply creating a series of agreements between two sovereigns. On the other hand, it was clear to all involved that the department was in the business of bringing once independent peoples under its control.

Genealogy represented a crucial instrument in these developments, serving as a kind of survey instrument of political authority. Officials were especially interested in creating hierarchy, first by identifying the "royal line" and who the paramount chief was and then, from there, working down to chiefs, sub-chiefs and, finally, to the newly-created office of headmen. As a mark of their status as once sovereigns now subalterns, chiefs were to receive subsidies from the colonial state in lieu of fines. Ideally, each district would have a chief assisted by paid headmen who, along with the magistrate assisted by police were responsible for enforcing the law. In creating districts chiefs, often glossed as "sub-chiefs," became headmen. Not surprisingly, these developments unleashed considerable political competition as various men jockeyed for power within the new colonial system, offering officials representations of the past as part of their attempt to provide legitimacy to their claims.

These claims were very often considerably inflated. Paramount chiefs were invariably weak. To the extent that people followed anyone it was the local chief, not the paramount. Even here it would be a leap of faith to agree with the pretensions of some chiefs that people remained "completely under my control," (32) as Kaiser Matanzima's nineteenth-century descendant represented. Chiefly power waxed and waned and political boundaries were constantly in flux and overlapping. British officials found this political landscape unintelligible and untenable. What unfolded in the early years of colonial rule was a flattening out and simplification of this landscape. In theory at least, the power of the colonial state was to be distributed homogeneously and so also was the power of individual chiefs, sub-chiefs, and headmen who were now state employees. The authority of a given chief, headman or official should not be any different at any point within his jurisdiction. And, of course, a chief's authority should not extend beyond "his "own section." (33) This requirement that authority and space would be coterminous recognized--if also rejected--the fact that it was not, that authority was not reducible to and enclosed by space and that power was exceedingly unevenly distributed across the landscape.

Colonial state formation had a number of immediate effects. By aligning space and authority it presented a new conceptual map of power that dovetailed with the colonial definition of "custom". It located authority in the body of a person who could be fixed within schematic political diagram. And it created and valorized upward links to the colonial state. Together the changes represented a political revolution. At the most basic level it made the conditions for colonial governmentality possible by defining the political subject. That subject could now be named, located, and enumerated, in short known. State formation at the same time produced a radically simplified or flattened political world. The other sources of authority entered into the prison house of irrationality, locked into the official discourse of superstition. Within what was becoming the formal domain of politics, many lesser chiefs who had enjoyed authority over others now found themselves effectively deposed and transformed into headmen and functionaries of the colonial state. Their domains were incorporated into the various magisterial districts in which a colonial official might rule over just one or a few chiefs. For example, in Engcobo, Thembuland, some thirty chiefdoms were reduced to just four; only two received subsidies so that, in reality, the other two chiefs had been utterly emasculated. The nineteen chiefs under the putative control of Dalasile were reduced by more than half to seven; these in turn became headmen. (34)

The invention of sovereignty in the context of colonial conquest thus entailed four general processes, sovereignty emerging as an effect of colonial state formation, one way the state "saw," one way of rendering legible the confusing world of European conquest. The first centered on embodiment, the ways by which a set of political principles came to inhere in persons as holders of office. This embodiment can be seen most obviously in the genealogical diagram of chiefs and their descendants, a product, I would suggest, that formed a part of state formation and a very distinctive map of the past. A second process entailed spatialization, or more precisely a territorialization of culture. Colonialism was all about flags and boundaries, in such a way that space and culture would, as an object of the state, be coterminous. In doing so it created classes of insiders and outsiders, residents and strangers. Indirect rule was premised on this notion, though its history predated indirect rule policies. Both embodiment and spatialization were not foisted upon a passive population; they appeared instead from the give and take of conquest and early rule. A third process involved codification, jurisdiction and policing, in short the law and its prosecution. Finally, fragmentation and normalization represented the fourth process. The invention of sovereignty naturalized some political principles, most obviously chiefship. At the same time other geographies of power and authority came to be fragmented and marginalized. (35)

Sovereignty and Segregation

One of the ironies of the conquest period is that the invention of tribal sovereignty seemed necessary for one of the avowed goals of early colonialism: the destruction of chiefship and the "civilization" of the native. The state invented sovereignty as an ineluctable part of conquest. The situation was especially complicated in the Cape, where Africans technically fell under European civil and criminal law. Neither chiefship nor native law and custom was ever abolished, however, and in King William's Town a special court was established to adjudicate disputes according to native law. In short, the situation was confusing at best. In the Transkei, however, despite the rhetoric of destroying chiefship and civilizing the native, the use of native law and custom was "accorded the widest possible measure of recognition." (36)

The usual way scholars have represented the colonial history of the region is the shift from direct to indirect rule, the latter based squarely on the idea of limited sovereignty. (37) The political road map then leads from indirect rule--seen most obviously in the 1920 Native Affairs Act and the 1927 Native Administration Act--to the 1951 Bantu Affairs Act and the beginnings of apartheid. It is certainly true that a main aim of early colonial policy was for magistrates to "occupy the position formerly held by the Chiefs." (38) Much of the everyday functioning of colonial rule circled through the magistrate's office and the location headmen. But it is a mistake to think that the transition from one sort of rule to another was so clear-cut. In the official mind, to transform African society required detailed ethnographic descriptions and quite explicit identification of and engagement with chiefs. Chiefship was, paradoxically, both attacked and elevated. The territorial claims of some chiefs were valorized and their domains greatly expanded. Certainly the locus of power had shifted towards resident magistrates and headmen, but chiefs were in a sense always waiting in the wings for a return to the political center stage of the colonial state.

Their return, as we now know, began in the twentieth century with the application of indirect rule to the native reserves. Embodiment, spatialization, and the law, the historical processes that shaped sovereignty and the political culture of the conquest period, sat at the ontological center of indirect rule. At the same time, the other loci of power and authority were further fragmented and marginalized, whether in terms of the privatization of healing or the emergence of "religious" social movements, as if the categories of the religious and the political were so easily distinguishable. (39) Legislation passed in the 1920s attempted to grant some limited sovereignty to the reserves at the very same time as creating a far more authoritarian order whereby the colonial state could rule "by decree" and thus conveniently by-pass much of the colonial parliamentary process. (40) This allowed, among other things, for the Governor-General, as the "supreme chief," the power to create, "divide" or "amalgamate" tribes. (41)

One of the most important shifts centered on the council system and the creation of chiefly courts whereby chiefs would apply "native law and custom." These new policies in effect returned to the situation of the 1870s, when the position was that rule would be orchestrated "principally through their own Chiefs and in accordance with Kaffir laws and customs." (42) Chiefs were well-aware of what was going in within the Department of Native Affairs. The shift towards segregation and indirect rule in the 1920s generally created a more competitive political environment in the native reserves. It seems that at least some chiefs invoked the early colonial period in their claims to control particular areas. In other words, the representations of political power and jurisdiction that had taken place in the context of colonial conquest were revisited in the twentieth-century politics of Indirect Rule. It is highly likely--though again much research remains to be done--that they drew on documentary evidence as well as upon chiefly histories that conflated genealogy, political succession and sovereignty. What is clear is that the new policies--the return of the chiefs to the center of "traditional" power--lead to various claims and jockeying for position.

The Great Depression and the Second World War delayed the full realization of the political world envisioned by the legislation of the 1920s. The turn to segregation and the idea of limited sovereignty accelerated dramatically in the 1950s with the introduction of apartheid. In 1956 the Bantu Affairs Act passed five years earlier was implemented in the Transkei, followed by the Bantu Self-Government Act past three years later, in 1959. The granting of self-government proceeded in fits and starts--mainly fits--until the 1970s and 1980s when the Transkei and the Ciskei became "independent."

The making of apartheid has been chronicled elsewhere. Three points are apposite here. The sorts of simplifications and the sorts of claims that had first emerged in the early colonial period returned with a vengeance in the 1950s, both on the part of NAD officials and traditional rulers. In many respects apartheid might be usefully viewed as a second colonial conquest, a new round when the identification of traditional rulers and their jurisdictions loomed large in the reworking of the political landscape. Men already identified as chiefs or paramounts represented to the state their claims to rule over their respective chiefdoms or tribes. Others submitted "claims for chieftainship," (43) claims that were then processed within the NAD on the basis of customary law and the official ethnographic record. Again, genealogies loomed large. The ethographical section of the NAD expended considerable efforts looking "into the history, ethnology and present day conditions of the tribes" (44) so as to restore native custom and the power of chiefs. For under apartheid "The Chief of a tribe .... is the political, administrative, executive and priestly head of the tribe." (45) The fiction, then, is that a single node of authority existed within any given chiefdom. Under apartheid, the territorial "units of administration" ostensibly returned to the original boundaries of the chiefdoms that had existed prior to colonial rule. (46) This completed the process of returning chiefs back to the center of the political state begun in the 1920s. Apartheid also continued a process that preceded indirect rule policies, that of creating clear, ethnically pure administrative boundaries. (47)

The second point is that "custom" and traditional leaders moved to the center of the most coercive aspects of colonialism and apartheid. Under apartheid especially, economic control and labor regulation formed a central part of tribalization. Not surprisingly, chiefs and headmen often found themselves targets of popular resistance; in the 1950s and early 1960s not a few were murdered. But they were not stooges either, at least not all. Many tried to walk a quintessential political tightrope and did their best to grapple with their position of controlling and allocating scarce resources and in mediating local conflict, on the one hand, and, on the other hand, recognizing that they were, in the end, employees of the state.

A third point centers around the politics of paramount chiefs. In the context of early colonial state formation, paramount chiefs had attempted to represent themselves as kings only to be demoted to mere chiefs. In the 1950s these politics resurfaced. The Thembu paramount Sabata Dalindyebo took on the pretensions of chief Kaiser Matanzima. Sabata may have enjoyed considerable popular support and certainly largely rejected the introduction of Bantu Authorities, but he was still committed to an assertion of "royal" control in Thembuland. Elsewhere, paramount chiefs largely sided with the new policies, which placed them at the head of the Territorial Authorities and through which passed state monies and other support. Paramount chiefs invoked "royal" (the word becomes increasingly common throughout the 1950s) prerogative which they used in their attempt to amass power and rework the political landscape. They were, in a sense, attempting to create kingdoms where none really had ever existed. One official rightly had observed that in the past chiefs had really "paid" paramounts "lip service." (48) With apartheid this all changed as paramount chiefs sought to become kings. As the official concluded, "matters became different." (49)

Apartheid culminated a policy and inaugurated a new era in which chiefs were beholden to the state. Chiefs's crucial linkages lay not with local communities but with the colonial administration, then the apartheid state, and now the post-apartheid ANC-led government. Certainly different traditional leaders have done different things at different times for various reasons; to portray them as mere stooges is a mistake. At the same time, it is abundantly clear they depended on the state for their personal income, for their access to police power, and very often for their personal safety. It is also clear particularly in the late 1950s and the 1960s that "traditional leaders" bore the brunt of popular resistance, resulting in the deaths of a number of chiefs and headmen. A state of emergency was proclaimed over the Transkei in the early 1960s and remained in effect for nearly three decades. Matanzima's rule became increasingly corrupt and authoritarian. By the late 1970s some politicians within the apartheid state had begun to realize that their dream of separate development had collapsed into a nightmare of poverty, instability, and illegitimacy.

Past and Future Imperfect 

What is striking about the politics of the Transkei over nearly the 150 years is that the issue of sovereignty has emerged at crucial moments of instability and state formation. Sovereignty, and the political rights that ostensibly flow from it, has become part of the common currency of South African politics. The state presents itself and the categories of its thoughts as "objective" and "real," and in so doing erases the arbitrariness of its creation: sovereignty is transmuted from historically constituted to ontologically given. The reification of political principles, new ways of locating power in bodies and space, had led to a situation in which political identities and political rights proceed power.

Today, as in the past, sovereignty is presented ahistorically, by various actors. As a recent COSATU document put it, "prior to the advent of colonialism political structures ... referred to as Traditional Leadership Institutions constituted the actual state formation .... based on particular ethnic identities or affiliations." (50) In a publication by the Helen Suzman Foundation, ANC MP and CONTRELESA President Patekile Holomisa argued, in September 2003, for the "[t]otal and meaningful recognition" of no less than 2,426 traditional leaders. "There is still the matter," he continued, "of those kings who were deposed or not recognized by the colonial governments. This has to be sorted out ..." (51) Elsewhere the African past is presented as the history of "over 10,000 states" where "traditional leadership"--read kings, chiefs and so on--had an "uncontested predominance." They were the "sole leaders with no other authority above them." (52)

Ethnicities were thus so many sovereign states. Here and elsewhere there is little attention to the invention of sovereignty and political rights within the context of early colonial state formation. Instead, by locating political rights in a very specific, singular and closed reading of the past, the pluralities of the past--and with them the possibilities of envisioning multiple futures--have been drowned out by the singular din of ethnic politics within the modern state.

Following the 1994 elections the South African government began making real its commitment to local government and especially to democraticizing rural areas, including the former homelands. Early policies centered on the creation of democratically elected councils that would oversee local government and administration, including land and development. Under this system traditional authorities were to have only an ex officio status. (53) CONTRELESA, representing traditional rulers, naturally balked at this initiative since it meant that state funds would be run through the councillors and not the chiefs. Traditional leaders threatened to boycott the 2000 elections as part of their protest at any legislation that might marginalize their position. Ultimately the Eastern Cape House of Traditional Leaders endorsed the new system once they realized that it "would broaden" their "influence." (54)

What is especially striking today is the reworking of custom and the state within a democratic order. South Africa is confronted with the issue of reorganizing the state and its relationship to political identities within a neo-liberal political economy and, ultimately, the conception and practice of citizenship within a plural, transitional society. The state's instrumental capacity is, however, relatively weak. In much of the Transkei no institution has stepped in to take the place of traditional rulers. Quite the opposite has happened. The region has emerged as a space of state ungoverance.

In the context of a neo-liberal order and state ungovernance the government initially appeared to pursue two strategies at once: to move forward with the creation of a new system of local government, and to placate traditional rulers and begin to address the question of "traditional rule." Discussions over traditional rulers and local government took place within the context of quite considerable violence in some areas (especially Tsolo/Qumbu), institutional collapse nearly everywhere, extraordinary levels of poverty and a horrific HIV/AIDS crisis. To speak of governance in much of the Transkei is something of a misnomer; the region might better be described as a space of ungovernance. It is probably fair to say that the state is confronted with its own weakness in the former homelands, either in terms of sheer political will or institutional capacity. For whatever reasons, what is clear is that the ANC has decided to co-opt traditional leaders.

The April 2000 "A Draft Discussion Document towards a White Paper on Traditional Leadership and Institutions," had at its center an imagining of the pre-colonial political past as comprised of clearly delineated "absolute monarchies," in other words sovereign states that lost their independence in the colonial period. Within this historical representation the document valorized traditional rule in the struggle for democracy, as if early resistance anticipated later political movements. It argued that "not withstanding oppression by successive colonial and apartheid regimes, the institution of traditional leadership pioneered resistance and led numerous struggles against colonialism. The advent of democracy in South Africa is also due to that pioneering role, which traditional leadership played." The document also relied on the 1948 Universal Declaration of Human Rights and its pronouncement that all people have "the right to participate freely in the cultural life of the community." Hitching representation of past culture to a discourse of prospective rights, culture emerges as if it were a kind of private property. Culture was represented not conceptually but as a thing, something that could be stripped away but which all people had an inalienable right to, in short culture as property in the singular and not as a "plural resource." (55) And, given the putative connective historical tissue tying early resistance to later democratic struggles, it naturally followed that traditional leaders would continue to have a role in a new South Africa; this, in fact was precisely what had happened under Chapter 12 of the South African constitution which had come into effect three years earlier in 1997.

Recent legislation would seem to indicate that, to the extent that the government is committed to building local institutions, it is casting its weight with traditional leaders. In so doing it is foreclosing other ways of thinking about history, politics and citizenship in a post-apartheid South Africa. The Traditional Leadership and Governance Framework Amendment Act became law at the end of 2003. Earlier, in August 2002, the government published the Communal Land Rights Bill which addressed, among other things, the hoary problem of communal land in the former bantustans. The legislation, which became law in early 2004 and which will begin to be implemented in 2005, vests authority primarily in the Minister for Land Affairs and in traditional leaders. One of its main objects is to "legally recognise and formalize African traditional system [sic] of communally held land." Another object is to establish a process "in which traditional leaders and local and national government actively participate and support communities in the administration of their land." By "community" the framers of the legislation mean "traditional" and it was but a small step to the conclusion that rights derived from one's membership within such communities so defined. Custom, in short, becomes the political trump card. (56)

Taken together, the Traditional Leadership and Governance Framework Amendment Act and the Communal Land Rights Bill mark a decisive turning-point in this new era of tribal and state formation. Recent policy has not been uncontested, both within and without parliament. However, the land bill, which passed unanimously, may be in violation of the constitution. The legislation is in effect locating rights in custom and recreating the politics of indigeneity that had first begun with colonial conquest. These political turns obviously have an impact on rights and citizenship, and, most immediately, on rural women whose position already is highly vulnerable and who are unlikely to secure stable access to land under the new legislation. (57) The new laws also have an impact on local development, where, for example in Pondoland, policies have been implemented without "any clear local development plan" and with relatively little participation of the communities involved. Where local institutions remain weak, as in much of the Transkei, observers now fear that the conditions exist for the development of "political patronage on a grand scale." (58) In short, there is some reason to be concerned that the new policies granting limited sovereignty to areas such as the Transkei may represent both a recognition of the state's weakness and its willingness to begin assembling clientalist networks. Certainly directing resources to tribal leaders, and running the apparatus of "development" through them, is a recipe for clientalism and raises troubling questions about transparency and democratic accountability in the new South Africa.

Beyond the South African case, however, there is the additional challenge of the rediscovery of the state and politics by social historians in a world after Foucault, Derrida and Lyotard. The linguistic and cultural "turns" within the historical discipline often marginalized the study of the political to the analysis of the discursive. Poststructuralism largely annihilated the social. There is now fortunately a return to the social and to the material that is crucially shaped and informed by the enormous theoretical advances of the last two or three decades. Phenomenologically-based analyses of issues such as state formation may offer social historians new ways of understanding the histories of power in the making of the modern world.

Department of History

Atlanta, GA 30322

ENDNOTES

1. Mahmood Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism (Princeton, 1996).

2. See, for example, Barbara Oomen, "'Walking in the middle of the Road': People's Perspectives on the Legitimacy of Traditional Leadership in Sekhukhune, South Africa," African Studies Centre, Leiden, 17 January 2002. Oomen has a book in press on the politics of traditional rule in contemporary South Africa. Thomas Koelble and Edward LiPuma are doing important work on chiefship and liberal democracy. See, for example, their "Chiefs and Democrats: On the Compatibility of Political Democracy and Traditional Leadership in South Africa," unpub. Paper, 2003, and "Limits to Liberation: Cultural Politics in the Age of Globalization," in Steven Robins, ed., Limits to Liberation: Citizenship and Governance after Apartheid (Cape Town, 2004).

3. See Daily Dispatch, for coverage of Matanzima's death and funeral.

4. Act. No. 41 of 2003: Traditional Leadership and Governance Framework Amendment Act, 2003," in Republic of South Africa, Government Gazette, v. 262, 19 (Dec. 2003). This legislation is discussed in the paper's conclusion.

5. David Held, "The Changing Structure of International Law: Sovereignty Transformed?" in Held and McGrew, eds., The Global Transformations Reader (Cambridge, 2000, 2002), 162-76, 162-4. See also, Graham Burchell, Colin Gordon and Peter Miller, eds., The Foucault Effect: Studies in Governmentality (Chicago, 1991); Jens Bartelson, The Critique of the State (Cambridge, 2001) and idem, A Genealogy of Sovereignty (Cambridge, 1995).

6. Bartelson, Genealogy of Sovereignty. For Hobbes, the sovereign brought domestic order but created anarchy internationally. On the Hobbesian dilemma see Keohane, "Sovereignty in International Society," in Held and McGrew, The Global Transformations Reader.

7. Michel Foucault, Power/Knowledge: Selected Interviews and Other Writings, 1972-1911, ed. Colin Gordon (New York, 1980), 121.

8. Pierre Bourdieu, "Rethinking the State: Genesis and Structure in the Bureaucratic Field," in George Steinmetz, ed., State/Culture: State Formation after the Cultural Turn (Ithaca, 1999), 53. See also Jens Bartelson, The Critique of the State (Cambridge, 2001) and idem, A Genealogy of Sovereignty (Cambridge, 1995).

9. Pierre Bourdieu, "Rethinking the State: Genesis and Structure of the Bureaucratic Field," in Steinmetz, ed., State/Culture, 68.

10. See, for example, Steinmetz, ed., State/Culture: State-Formation after the Cultural Turn

11. See Clifton Crais, The Politics of Evil: Magic, State Power, and the Political Imagination in South Africa (Cambridge, 2002).

12. See, for example, Thomas McClendon, "The Man Who Would be Inkosi: Civilising Missions in Shepstone's Early Career," JSAS, 30, 2 (June 2004): 339-58; Clifton Crais, The Politics of Evil.

13. See, for example, J. B. Peires, The House of Phalo: A History of the Xhosa People in the Days of their Independence (Berkeley and Los Angeles, 1981) and, more recently, Norman Etherington, The Great Treks: The Transformation if Southern Africa, 1815-1854 (London, 2001). The great exception of course was the Zulu kingdom. However, for an important revision of how we understand the kingdom see Carolyn Hamilton, Terrific Majesty: The Powers of Shaka Zulu and the Limits of Historical Invention (Cambridge, Mass., 1998)

14. Peires, House of Phalo; Etherington, Great Treks. Cf. Igor Kopytoff, ed., The African Frontier: The Reproduction of Traditional African Societies (Bloomington, 1987).

15. Sara Berry, No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub-Saharan Africa (Madison, 1993), 27.

16. See Crais, Politics of Evil. A closer reading of this historical writing (and here I include my own early work) does not reveal much of a hermeneutical analysis of official texts or much attention to the sociology of early colonial knowledge, particularly its negotiated nature. This naturally led scholars to analyses of ethnicity as largely colonial inventions mediated only by African intellectual representations. In one sense the historiography has reproduced the gaze of early officialdom, at times assuming the categories that shaped the production of knowledge and which constituted the archive as reflecting social reality. They were also largely accepting unquestionably African elite representations of the past, for example Zulu royal assertions around what defined the political boundaries of the kingdom. As Glassman has written, following Nietzsche, "in defining" the "phenomenon" of ethnicity "we deny it a history." Jonathon Glassman, "Slower than a Massacre: The Multiple Sources of Racial Thought in Colonial Africa," American Historical Review, 109, 3 (June 2004): 720-54

17. The Creation of Invisible Histories," in Victoria E. Bonnell and Lynn Hunt, eds., Beyond the Cultural Turn: New Directions in the Study of Society and Culture (Berkeley and Los Angeles, 1999), 182-216.

18. See Feierman, "Creation of Invisible Histories"; Luise White, Speaking with Vampires: Rumor and History in Colonial Africa (Berkeley and Los Angeles, 2000); Crais, Politics of Evil.

19. Inkosi is a wonderfully plastic term. Its linkage to status and authority is clear and ancient for large areas of Eastern, Southern, and South-Central Africa. It has still earlier roots relating to the word for lion in West and West-Central African languages. Initially, it appears to have meant a "strong or mature person, particularly (though not necessarily only?) male, having authority over a small group of kin," in short, the head of an extended family. In South Africa the word is retains its honorific history. The word's root forms part of the common word for "thanks." In areas where states did emerge, the word could mean both chief and king, the latter modified by the adjective "great". In other words a "king" was a "great chief." Such also was the case for so-called paramount chiefs; what distinguished a paramount from a king would become the subject of considerable and ongoing discussion and debate. According to Ehret, "*-kosi enlarged its scope to the leader of a larger local grouping of people, metaphorically invoking but no longer literally connoting the role of a head of a family." Inkosi is, finally, used as "God", as in "Inkosi Sikulele Afrika", "God Bless Africa," South Africa's national anthem. What exactly inkosi meant, then, may be less apparent than usually assumed. The very plasticity of the word is suggestive of the ambiguities of power and authority in the pre-colonial era. Ehret, An African Classical Age, 148-9, 251. Among the Shona the word could mean "senior wife," while among the Manyika, also in Zimbabwe, a form of the word means "headman." See Ehret, African Classical Age 148; Guthrie, Comparative Bantu, vol. 3 (Farnborough, 1967), 289, C.S. 1102*-koci.; a history of the word's meanings could offer a richer understanding of South African political history.

20. See also David Lan, Guns and Rain: Guerrillas and Spirit Mediums in Zimbabwe (Berkeley and Los Angeles, 1985), for another example.

21. In a similar fashion Feierman has written of the alternative forms of authority in East Africa. See "The Creation of Invisible Histories," in Bonnell and Hunt, eds., Beyond the Cultural Turn, 182-216.

22. BC 293, B2731, Edmonstone, Judge and Grant to Secretary of Native Affairs, 30 Dec. 1872.

23. BC 293, B 263.17, Stanford, statement of Xelo, 8 Feb. 1872.

24. BC 293, B2731, Edmonstone, Judge and Grant to Secretary of Native Affairs, 30 Dec. 1872.

25. BC 293, B2731, Edmonstone, Judge and Grant to Secretary of Native Affairs, 30 Dec. 1872.

26. CO 4521, Office of resident with Gangelizwe, 28 Oct. 1875.

27. CMT 1/143, minutes of meeting, 24 June 1884.

28. CMK 1/94, Hope to Brownlee, 4 Sept. 1879.

29. BC 293, D10, ms. of Rev. E.J. Warners, n.d., t.s.

30. CO 4521, Graham, Unannexed Tembuland, 16 Aug. 1881

31. CO 1156, Leonard, Opinion, 23 Apr. 1881.

32. 1/COF 9/1/44, Extracts from the Report of the Government Commission on Native Laws and Customs, Jan. 1883.

33. CO 4521, office of resident with Gangelizwe, 28 Oct. 1875.

34. CMT 1/27, list.

35. See Crais, Politics of Evil; Mamdani, When Victims Become Killers.

36. Howard Rogers, Native Administration in the Union of South Africa (Johannesburg, 1933), 220.

37. See, for example, Mamdani, Citizen and Subject.

38. NA 158, Shaw, Report, Jan. 1878.

39. Feierman. On social movement see Crais, Politics of Evil.

40. Mahmood Mamdani, Citizen and Subject: Contemporary Africa and the Legacy and Late Colonialism (Princeton, 1996).

41. Quoted in Mamdani, Citizen and Subject, 71

42. BC 293, D10, ms. of Rev. E.J. Warners, n.d., t.s. We do not know to what extent and in what manner officials, in devising legislation such as the 1927 Native Administration Act, consulted the documentary record beyond that of laws and proclamations; to my knowledge there is no study of the kinds of historical claims officials were making as part of the law-making process. However, it seems highly likely that they did precisely this, in effect trying to re-imagine the sorts of political arrangements that had been put together in the early years of colonial rule and their bearing on the political and economic context of the twentieth century. Rogers, Native Administration in the Union of South Africa, for example, provides precisely this documentary history of policy. The most obvious change center on the 1910 Union of South Africa and the increasing reliance of the mining industry on labor from within South Africa.

43. (UMT) CMT 66/177, Secretary for NA to CMT, notes by CMT, 9 Mar. 1953 [1956?].

44. CMT 3/1451, report for 1955.

45. 1/ALC 10/11, "Bantu Authorities," n.d. [1963].

46. Evans, Bureaucracy and Race, 251.

47. In practice of course expedience outweighed allegiance to ethnography. The NAD was more than willing to create chiefs where no existed or appoint chiefs purely on the basis of their collaboration.

48. 1/BIZ 6/47, Ramsay to Secretary for Bantu Administration and Development, 21 Nov. 1958

49. 1/BIZ 6/47, Ramsay to Secretary for Bantu Administration and Development, 21 Nov. 1958

50. "COSATU Submission on the Draft Paper on Traditional Leadership and Governance," 30 Nov. 2002.

51. Patekile Holomisa, "Traditional Leadership in a Post-Colonial SA," Focus 31 (Sept. 2003).

52. Prince Mashele, "Traditional Leadership in Historical Context," Municipal Talk (May 2003).

53. Municipal Structures Act No. 177 of 1998.

54. Daily Dispatch, 5 Sept. 2000.

55. Mahmood Mamdani, When Victims Become Killers (Princeton, 2001), 277.

56. Communal Land Rights Bill, October 2003.

57. PLAAS/NLC Community Consultation project on the CLRB, "Submission to the Portfolio Committee for Land and Agriculture," 10 Nov. 2003.

58. Edward Lahiff, "Land Reform and Sustainable Livelihoods in South Africa's Eastern Cape Province," IDS, March 2003. See also Koelble and LiPuma, "Chiefs and Bureaucrats."

By Clifton Crais

Emory University