1In some form or another, most theories of politics in Africa rest on the assumption of a fundamental distinction between a western bureaucratic, legal form of domination, and an African domination that is much more personalised. How these elements, drawn from Weber ’s ideal types of domination, are combined in these models differs widely. Chabal and Daloz (1999) list the three dominant models of African politics in political science: the neo-patrimonial approach stresses the state’s failure to operate on the formal level of ensuing rule of law and the application of public policy. At the same time, the approach sees the African state as substantial in that it constitutes the most important arena for the establishment of patrimonial relations. What in an ideal model of a western society would constitute a public sector, the neo-patrimonial approach sees as an extension of the private. In contrast, the hybrid state model, while not questioning the existence of privatised networks, explains the current situation as the result of a mixture of pre-colonial forms and genuine African political traditions. Lastly, the transplanted state perspective depicts modern African states as neither hybrids nor modern, but as a radical altered form. A political tradition that emerged in Europe, the state, when moved to a completely different setting, changed its nature incisively.
2Against these three models Chabal and Daloz propose a fourth model, the instrumentalisation of disorder. This model seeks to understand the way in which the weak institutionalisation of the political order benefits political elites. They argue that the predicament of contemporary political analyses of Africa stem from the application of ideas about modernisation and politics derived from a western experience of modernisation. Their point is that African political actors draw on a register of resources that a western model of politics would see as a mixture of traditional and modern.
3All these models in one way or another contrast Weber ’s conception of traditional and legal domination, although Chabal and Daloz stress that this distinction might owe more the analytical models than to African realities. In this paper, I want to contribute to the debate, not by proposing yet a new model of African politics, but by examining in concrete detail the interaction of these two understandings of politics in a South African case, located in Cato Crest.
4Chabal and Daloz explicitly leave South Africa out of the analysis, but as Mamdani (1996) has demonstrated, there are good reasons for treating South Africa not as an exception, but as a typical case of the colonial state, where the population was divided into supposedly traditional subjects under traditional rule, and citizens. The case in question is a South African development project, where the developers sought to grant people’s rights to housing as citizens. In the process, however, they had to negotiate with a community leadership that operated according to another logic, that of personalised patron-client relations. The paper examines the relationship between these two perceptions of the political in some detail, before it returns to the question of the relationship between the two forms of domination.
5The area in which Cato Crest is situated, Cato Manor, has been one of the most violently contested urban areas in South Africa. It is a huge area of land within five miles of the centre of Durban. Originally owned by a white landowner, George Christopher Cato, the land was partitioned in the 1930s and sold to Indian workers who had completed their tenure as well as to other Indian immigrants (Edwards 1994). With the great influx into the city during the Second World War, many of the Indian landowners sublet to black workers and the first years after the war saw a vibrant mixed settlement emerging.
6All was not harmony however. Conflict between Indian landowners and black tenants erupted into violent conflict in 1949 and set in motion plans to regulate the area for white occupation. Cato Manor fell prey to apartheid development, and by the mid-1960s, all its inhabitants had been forcibly removed to the new townships outside Durban. The destruction of Cato Manor, which affected the lives of 120,000 people, came to epitomise the brutality of apartheid development.
7After its black and Indian occupants were forcibly removed in the late 1950s and early 1960s, Cato Manor was a wasteland for almost thirty years. Towards the end of the 1980s, when it became increasingly clear that change in the political regime was imminent, people began moving into Cato Manor, erecting shacks well out of sight from the relatively affluent neighbouring suburbs of Bellair, Manor Gardens, Sherwood and Westville. The influx of people continued, despite attempts by the authorities to halt it. In 1995 it was estimated that more than twenty thousand people lived in the squatter settlement of Cato Crest alone (Makhathini & Xaba, 1995).
8In an attempt to mobilise against the government’s demolition teams, the squatters formed a neighbourhood association, of the type referred to in South Africa as a civic. Besides the civic, other organisations also involved themselves with the plight of the squatters, among them the Cato Manor Residents’ Association (CMRA), which represented the Indian landowners who had been forcibly removed in the 1960s. Since 1979 they had successfully fought off attempts to develop Cato Manor for white housing. The CMRA, together with other organisations, established a Development Forum in 1992. In 1993, the Forum launched the Cato Manor Development Association (CMDA), with a mandate to guide the development of the whole of Cato Manor.
9From the outset the CMDA set out to find alternatives to the top-down planning of the apartheid government. They chose a participatory approach which entailed close collaboration with community structures, and which rested on an assumption of the representativeness of these structures vis-à-vis the wider community. People were to be granted their rights as citizens of the new South Africa, and the community structures were to function as intermediaries between the citizens and the state, here represented by CMDA. The developers sought, in other words, to act as an extension of the new democratic government’s legal domination. But as we will see, this conception of politics came to interact with another, personalised understanding of power and the political.
10Any present-day development project in South Africa, and especially in Cato Manor, takes of course place in the historical context of apartheid development (Nustad 1996, 1999). As Murray and Williams (1994) have argued, a top-down approach to development has been made impossible by the legacy of apartheid. To succeed, the development of Cato Manor had to demonstrate that it constituted a break with top-down approaches. The ANC was favouring a bottom-up approach to development in its «Reconstruction and Development Programme» (ANC 1994), and to have any chances of success, the development of Cato Manor had to be seen as a partnership between people living in the area and the developers.
11The Cato Manor Development Association recognised that working in highly politicised settings like Cato Crest would be extremely problematic, and decided to postpone their involvement there. Instead they concentrated on the empty spaces of land in Cato Manor. But their hands were forced: because of the public outcry about land invasions, the CMDA had to start negotiations with the local leadership in Cato Crest earlier than they had wished. It turned out to be as difficult as they anticipated. People involved in the community structures in Cato Crest were extremely distrustful of the CMDA. I was repeatedly told that the CMDA could not possibly be interested in promoting development because a developed Cato Crest would mean an end to their well-paid jobs.
12Already at this stage in the narrative it is apparent that there are two different conceptions of politics at work. The developers wanted to structure their task according to a legal perception of politics. They could not relate to the whole mass of people in Cato Crest, and had to construct a segment of it, the leadership, as representative of the rest. This formal understanding derives from Weber ’s notion of bureaucratic domination whereby an elected representative acts in accordance with the rules laid down and defined by his or her office. The assumption in Cato Crest, then, was that local leadership would act as representatives for the larger community. Already in the suspicion that the employees of the CMDA were more interested in keeping their salaries than in providing houses, a different perception of the process is evident. People in the local leadership expected CMDA to mix up the personal and the official, and to act more in accordance with the Weberian ideal type of traditional leaders. This, it turned out, was how they themselves approached the task as representatives of Cato Crest.
13In this first period, the newly formed community organisation successfully mobilised against outside attempts at expelling them. However, once the immediate threat was gone, another, and more disruptive, power mechanism seems to have underscored the actions of the local leadership. This mechanism is similar to what Bayart (1993) and others have termed neo-patrimonialism. In order to hold on to the power as community representatives, the local leaders had to mobilise resources in some way, both to survive themselves and to be able to build up a group of loyal followers.
14The first such instance that members of the community structures recounted to me, and which drew the line between two antagonistic groups in the local leadership, occurred when the community organisation managed to negotiate the erection of stem pipes for water supply. It was decided that the community organisation would be responsible for collecting payment and that Cato Manor Residents Association would be the body formally responsible. This set-up resulted in allegations of mismanagement of funds, and the community organisation was divided into two factions accusing each other of bad management and crimes.
15This incident defined two groups among the local leadership who continued to fight for power throughout the period in which I followed the local political scene. One aspect is worth noticing here. The water company was willing to erect standpipes and give credit to the population in Cato Crest because the members of the community organisation were perceived as representative of the wider community. In other words, the space within which the personalised economic activity of embezzlement took place was created by the formal legalistic understanding of the water company. The company, if it was to extend its services to the community, needed a representative, responsible body with which to conduct business.
16This example was relatively innocent, but the same mechanisms where involved later with much graver consequences. Events took a dramatic turn in mid-1993 when there was a huge influx of people into the Cato Crest as a result of the pre-election violence in KwaZulu/Natal. In a few months, the population increased from a few hundreds to more than twenty thousand. As the settlement increased in size, it became the focus of targeted efforts of improvement. The Cato Manor Development Association was established and given the sole responsibility for coordinating development efforts in the area.
17There was a mismatch between a formal understanding of the local leadership as representative of the local population, and the actual practice of the leadership. The factions in the leadership that were established right from the start continued to fight for the control over the positions as community representatives in negotiations with outsiders. As a result, there were constant struggles over the positions of the community organisations and the establishment of rival organisations.
18Why was it so important to gain access to these positions? The obvious answer is access to resources. Whereas for the developer, a community organisation represents a manageable manifestation of «the community’, for people living in the squatter settlement the organisation it represents a link to the world of government and NGO funding and equipment. Such positions become extremely valuable resources in themselves, resources that exist because of the way the development process is structured.
19Converting a position in a community organisation into material gains and power took a variety of forms. One group of strategies was to allocate jobs given by the developers to dependants and followers; another to appropriate money and equipment from development organisations. A third very common strategy was the trading of registration numbers by people involved in the community organisations. Regardless of whether they were private developers, local or provincial authorities, or NGOs, the external agents involved in Cato Crest have been concerned with imposing some kind of order on the perceived chaos of the settlement. Making plans for the area made it necessary for the planner to have some control over those to be planned for. Two closely connected strategies of formalisation were employed by the developers to achieve this. The first involved the transformation of persons into objects in the form of numbers. This was done through the allocation of registration numbers to the residents in the area, with the aim of establishing the size of the population. The second strategy entailed the extension of recognition of the right to stay, symbolised by the registration number, on condition that the recipients managed to keep others out. The idea, then, was that those given the right to stay would see it as in their interest to keep other people out. This strategy would then stabilise the situation and enable the planners to estimate the number of people to include in their plans. There were four different attempts at ordering Cato Crest in this way. Each time new shacks were built, and when the number system lost all resemblance with the reality on the ground, the developers devised new formulae.
20These futile attempts to establish control underwent a transformation through their application. The numbers, in themselves, became commodities that obtained an exchange value. Ownership of a number not only gave freedom from police harassment, but also had an additional potential value. If the development of formal housing commenced, it guaranteed the number-holder a part in that process, and eventually a formal house. While the trading of registration numbers had taken place to a smaller degree when Cato Crest consisted of a handful of houses, trade became organised after the mass influx of 1993. As part of the agreement between the then active community organisation and the provincial authorities, the organisation was charged with controlling the influx of people into the area. The civic initiated a system of «marshals», people who were to allocate registration numbers to those already settled and make sure that no one arrived after the cut-off date. This put the marshals in a position where they had, on the one hand, a valuable commodity to offer (the number which symbolised legality), and on the other hand a sanction; they had the power to inform demolition teams from the provincial authority that a shack had been erected after the cut-off date. The marshals were quick to exploit the opportunity opened to them; they approached people building shacks and told them that they needed a registration number to stay. Prices for a number ranged from R 200 to R 500 (USD 37-95). The marshals could demand a relatively high price because they had the sanction of informing the provincial authority that the house was being built illegally and therefore had to be destroyed.
- 1 Interview, 26 October 1996. The interview was interesting; after answering my question at first in (...)
21I spoke to the CMDA Chief Executive Officer (CEO) after competition between the rival fractions had fuelled a taxi war that killed fourteen people and destroyed all organised community politics1. He conceded that many of the problems they had encountered were the result of struggles between different factions of the leadership over positions of representation. He began his analysis of Cato Crest, and of what he called the structural problems of the CMDA’s position, by saying that the main problem was one of legitimacy. The CMDA lacked the authority to make decisions about development that would be accepted by people in the area. There is always a question, he said, of whether any other structure than an elected local government could have sufficient authority and legitimacy to do that.
22The CEO said that he felt «played» by the different interest groups. His description of the situation parallels my own. He saw the leadership in Cato Crest as being in a position where they could only entrench their power bases by being perceived as those who delivered the resources. Their power derived from their ability to say that they were the people who had access to CMDA resources, that the CMDA was working for them. This was, according to the CEO, the situation that would have to change if the development of Cato Crest was to succeed. There had to be a mental shift, whereby people saw themselves as citizens with inalienable rights, rather than as the dependents of local strongmen. The vehicle for this transformation could only be the newly elected local government:
«The advent of democratic local government and metro government is a major watershed in one’s ability to deal with these issues. What is crucial I think is to integrate this; is to treat the people, the individuals in Cato Crest, the shack-dwellers, as citizens, as citizens of this city. And to, in a sense, to transfer that knowledge, that psychological shift – that we are citizens – we don’t have to work through local power brokers, local warlords and other con artists. It is part of another culture really. As citizens we have rights, and we have rights, and we have rights to access resources, the rights to be integrated into the body politics, into the city. The medium for that ultimately, the legitimate leadership, are the elected politicians. And I think we have started, we actually got a good leadership»2.
23The CEO went on to clarify the differences between the two forms of community leadership: that which existed at the present and that which would have to come into being:
«I think the elections throw up an opportunity, a real opportunity for legitimate political leadership, with authority, with legitimate authority, to actually start to provide a leadership that is abstracted from the immediate material struggles in the area, and to start to draw those people into, lead those people into – I don’t know, I am not an anthropologist – but [from a] very primitive form, to a another form of social organization, to a more modern [form]»3.
24Having made the distinction between the old and the new forms of leadership, he explains the current problems as a result of the juxtaposition of these forms:
«The leadership format has come from an environment where there is no government and never has been a government really. Where people, and it emerged out of apartheid, where people weren’t citizens, they didn’t have rights, they didn’t have potential access to resources. So you had one particular mode that is inappropriate to a context where they are recognized as citizens, where they do have rights, but they don’t know it. And the only persons that can tell them that are the politicians …. Politicians will have to do this, got to get the message to the people, right down to the grassroots, to each individual, to each household, that you are a citizen, you have a subsidy, you have other rights. As politicians our responsibility is to make sure that those assets are available to you, that you do have these assets. They are yours. And your local development committee ... are not bringing you the resources, you have got them. They are the people to ensure that you get the most out of your things. So you are no longer now dependant on those people, you are entrusting your assets, ... the management of your assets, the articulation of your assets, to the local development committee. They are there to ensure that the agent bringing those assets utilize them to the most effective way possible. They are your representatives, and you are investing your shares with them or you are entrusting your shares with them .... There is a very big difference between that and some shack-lord saying I am bringing you ... you don’t get access to those resources except through me»4.
25According to the CEO, then, the development of Cato Crest can only succeed once a change in the consciousness of the population has taken place. The people of Cato Crest must begin to see themselves as citizens with inalienable rights rather than as dependents of local strongmen. In this scenario the development committee are there to look after the interests of the people, they are merely representatives.
26The dichotomy that the CEO employs – the leadership as the guardians of peoples’ already-existing rights versus the leadership as the supplier of resources – is similar to a distinction that Weber deployed when he characterized the difference between two ideal types of domination: «traditional domination» and «legal domination». By using Weber ’s analytical tools the assumptions that the CEO, and many others with him, base their argument on will be easier to explore. What follows is not an attempt to force the reality of the development of Cato Crest into a model constructed for the analysis of the Prussian state. But by applying the tools that Weber developed to the two leadership models that the CEO described – that which existed at the time and that which had to come into being before any development could commence – the premises of participa- tory development can be made apparent.
27When Weber set out to investigate forms of power, domination, and legitimacy, he started by distinguishing between three «pure» or «ideal» types of domination: charismatic, traditional and legal. This typology was meant as an intellectual apparatus for the investigation of concrete historical manifestations that would, Weber stressed, always be a combination of the three types. His concern with forms of legitimacy (the idea that justifies domination) did not imply that he favoured an idealist explanation of domination. Rather, it was an important element to be considered together with other elements: «Experience shows that in no instance does domination voluntarily limit itself to the appeal to material or affectual or ideal motives as a basis for its continuance. In addition every such system attempts to establish and to cultivate the belief in its legitimacy» (Weber 1968, 213). As Marx also stressed in his earlier works (e.g. The German Ideology), these ideas are not only a reflection of an underlying base, they have consequences in themselves. Thus, Weber continues, «according to the kind of legitimacy which is claimed, the type of obedience, the kind of administrative staff developed to guarantee it, and the mode of exercising authority, will all differ fundamentally» (Weber 1968, 213). It is this interdependence between forms of obedience, administration and type of authority that I will draw upon to examine the CEO’s assumptions.
28According to Weber, a bureaucratic organization operating in a situation where the rule of law prevails is governed by certain principles:
«I. There is the principle of fixed and official jurisdictional areas, which are generally ordered by rules, that is, by laws or administrative regulations.
1. The regular activities required for the purposes of the bureaucratically governed structure are distributed in a fixed way as official duties.
2. The authority to give the commands required for the discharge of these duties is distributed in a stable way and is strictly delimited by rules concerning the coercive means … which may be placed at the disposal of officials.
3. Methodical provision is made for the regular and continuous fulfilment of these duties and for the execution of the corresponding rights ...». (Weber 1995, 196).
29Weber ’s first criteria of «rational legal authority», then, are concerned with the rule-bound administrator whose work is governed by impersonal criteria and who has strictly defined authority and means of compulsion. All this serves to separate the powers vested in the office from the person holding it at any given time. Weber continues, «II. The principles of office hierarchy and of levels of graded authority mean a firmly ordered system of superand subordination in which there is a supervision of the lower offices by the higher ones. Such a system offers the governed the possibility of appealing the decision of a lower office to its higher authority, in a definitely regulated manner … III. The management of the modern office is based upon written documents (‘the files’), which are preserved in their original or draught form …. in general, bureaucracy segregates official activity as something distinct from the sphere of private life. Public monies and equipment are divorced from the private property of the official» (Weber 1995, 197).
30A further separation of person and position is described here. Not only is the authority vested in the office strictly defined, but also the «means of administration» is strictly separated from personal property. Likewise, the office is subject to checks in an administrative hierarchy and the person holding an office cannot appropriate it.
31Contrast this with Weber ’s description of the system of administration under traditional or patrimonial rule:
«The masters are designated according to traditional rules and are obeyed because of their traditional status (Eigenwürde). This type of organized rule is, in the simplest case, primarily based on personal loyalty which results from common upbringing. The person exercising authority is not a “superior ”, but a personal master, his administrative staff does not consist mainly of officials but of personal retainers, and the ruled are not “members” of an association but are either his traditional “comrades”… or his “subjects.” Personal loyalty, not the official’s impersonal duty, determines the relations of the administrative staff to the master» (Weber 1968, 226-227).
32Instead of an impersonal system, domination here is manifested through a person. He derives his authority through personal loyalty rather than through written laws. This has consequences for the type of administration that sustains traditional domination:
«In place of a well-defined functional jurisdiction, there is a conflicting series of tasks and powers which at first are assigned at the master’s discretion …. These competing functions originate particularly in the competition for sources of income which are at the disposal of the master himself and of his representatives» (Weber 1968, 229).
33A prerequisite for a bureaucratic organization is that it is conducted on a continuous basis, which means that it must be a full-time occupation. Combined with the prerequisite that compensation must not be derived from the revenue of the office, some other compensation must be sought. The bureaucratic alternative is to reward the official with a monthly allowance in money or kind. Alternatively, if one sees the leadership in Cato Crest as politicians, Weber ’s distinction between «living off» and «living for» politics could be applied. The point is the same: the leadership in Cato Crest are expected to perform their functions as politicians or bureaucrats without compensation. Their situation resembles much more what Weber describes as non-bureaucratic administrations, which depends «upon the fact that those who outrank others in social, economic, or honorific terms are associated in one way or another with the performance of administrative functions. This usually means that the incumbent is rewarded for his assumption of administrative duties by the economic exploitation of his position, which he may also use for purposes of social prestige» (quoted in Bendix 1960:423). This is an alternative description of the «corrupt» leadership in Cato Crest. Indeed, it is an ethos which was explicitly expressed by some among the Cato Crest leadership – one must help oneself while helping others. Weber ’s description of traditional domination is similar to the CEO’s description of the current situation in Cato Crest, where the leadership is not «abstracted from the immediate material struggles in the area’.
34Importance adhering to the person is another characteristic of the leadership in Cato Crest which Weber defines as being contrary to a bureaucratic form of domination. In patrimonialism the leader achieves great personal authority whereas the bureaucratic leader has authority only through his or her office (Weber 1995:198). Weber ’s traditional ruler has to contend with the forces of decentralization – the tendency for his subordinates to usurp his privilege in their own names. Again, this is a process that has been evident in Cato Crest. The various groups contesting for power have been centred on certain strong individuals who have managed to command direct personal obedience. The two main antagonistic groups arose after the break-up of one group, the first civic, which was, in Weber ’s terms, subject to the forces of decentralization.
35So far, what I have outlined is nothing more that the incongruence of two perceptions of politics. The CMDA failed because they based their programme on an understanding derived from Weber ’s legal domination, where the people in Cato Crest already were citizens with inalienable rights. In this scenario, the community leadership were to act as the people’s elected representatives, confined to mediating between the individual citizen and the state. Instead, the community leadership acted as patrimonial leaders. The power of the leadership in Cato Crest, then, parallels that which Weber described as «traditional domination’. The CEO’s ideal, that the people see themselves as citizens and their leadership as providers of inalienable rights, parallels Weber ’s «legal domination’. The leadership, in this scenario, is based on bureaucratic principles where the means are entrusted to them by the people or the people’s elected representatives. The more interesting aspect of the case arises when it is subjected to closer scrutiny. Then it appears that the two perceptions of politics were less dichotomical than it first would appear. In fact, the patrimonial relations that were established were conditioned by the legal framework. The resources that went into establishing power as community leaders were made available because the developers interacted with the community leadership as if they were legal representatives.
36The majority of the leadership in Cato Crest had no stable income and hence they used their positions on the community structures for what they were worth. Their activity followed the logic of Weber ’s traditional leader. Positions were used to build up contacts with the outside world and to create a following of dependents inside Cato Crest. What emerges, then, is a picture of a situation where people who were subjected to a legal domination managed to turn the instruments of that subjection into resources in a patrimonial network.
37The conscious creativity involved in that translation was best expressed by the first person I got to know in Cato Crest, who was assigned to me as an assistant by the community organisation then in control. We were sharing a few beers, the situation was still very new to me, and I asked him why he had become involved in the community structures. «People who do the projects do not like individuals’, he said. When I asked him what he meant, he explained that if he, as a black person, wanted to do something for himself, he could not just go to the NGOs and ask for money or support. «The people who do the projects» think of the black man as a group. In effect, membership of one of the community organisations was the only way to access beneficiary organisations.
38In most cases, the linking together of these two political forms were straightforward, as when the community leadership appropriated resources earmarked for community development. More theoretically significant is the case of the housing numbers, where the same object, numbers, were able to achieve totally opposite functions within the two forms. Conceived to order the social field so as to guarantee the citizens their inalienable rights in an as effective way as possible, the numbers were transformed to become valuable resources in the patrimonial network.
39This takes us back the models that political scientists use to make sense of politics in Africa. The case strengthens Chabal and Daloz’s point that contemporary politics draws on a wide spectre of resources that cut across scientific categorisations. In addition, it demonstrates that different political forms can be both in conflict and function syncretically depending on the perspective from which it is viewed. From the point of view of the CMDA, the politics of the local leadership must appear as an alien form, as disruptive to what it set out to do. But from the point of view of the local leadership, the legal bureaucratic conception of politics opened up a space for a different politics. This should caution us about operating with too rigid conceptual boundaries between different forms of politics. As the example from Cato Crest demonstrates, different political form can be dependant on each other, and be linked in creative ways by the people who are attempted subjected to them.