Bureaucracy and Race: Native Administration in South Africa / Edition 1

by Ivan Evans

Overview

Bureaucracy and Race overturns the common assumption that apartheid in South Africa was enforced only through terror and coercion. Without understating the role of violent intervention, Ivan Evans shows that apartheid was sustained by a great and ever-swelling bureaucracy. The Department of Native Affairs (DNA), which had dwindled during the last years of the segregation regime, unexpectedly revived and became the arrogant, authoritarian fortress of apartheid after 1948. The DNA was a major player in the prolonged exclusion of Africans from citizenship and the establishment of a racially repressive labor market.

Exploring the connections between racial domination and bureaucratic growth in South Africa, Evans points out that the DNA's transformation of oppression into "civil administration" institutionalized and, for whites, legitimized a vast, coercive bureaucratic culture, which ensnared millions of Africans in its workings and corrupted the entire state. Evans focuses on certain features of apartheid—the pass system, the "racialization of space" in urban areas, and the cooptation of African chiefs in the Bantustans—in order to make it clear that the state's relentless administration, not its overtly repressive institutions, was the most distinctive feature of South Africa in the 1950s.

All observers of South Africa past and present and of totalitarian states in general will follow with interest the story of how the Department of Native Affairs was crucial in transforming "the idea of apartheid" into a persuasive—and all too durable—practice.

Product Details

ISBN-13:	9780520206519
Publisher:	University of California Press
Publication date:	09/29/1997
Series:	Perspectives on Southern Africa , #53
Edition description:	First Edition
Pages:	401
Product dimensions:	6.00(w) x 9.00(h) x 1.40(d)

About the Author

Ivan Evans is Assistant Professor of Sociology at the University of California, San Diego.

Read an Excerpt

Chapter One

Urban Administration in the Interwar Years

Urban administration in the segregation era was shaped by the liberal dislike of extensive government interventions into society. This pronounced ideological preference came under increasing strain, however, in the interwar years as segregation policy set the state on an increasingly interventionist course, placing pressure on the Department of Native Affairs to take a greater interest in the practical details of converting segregationist ideology into practical programs. The earliest stirrings of interventionist state administration are therefore to be located in this period. The department failed, however, to rise to the occasion, for three closely related reasons.

First, the first three decades after the Union was formed in 1910 were dominated by disputes over African policy. Matters of administration lagged behind the separate and combined attempts of the SAP and the NP, the two major political parties of the 1920s and 1930s, to reach agreement over the precise form of segregation. Furthermore, although success in reaching such a consensus culminated in the emergence of the Fusion Government (1932-39), a coalition of the two parties, a notable feature of African policy in the interwar years was its flexibility and ambiguity. The policy accords of the 1920s and 1930s embodied a deep-seated tension. On the one hand, they rested on the assumption that the need for racial segregation would gradually wither away as the civilizing process supplanted all vestiges of tribalism, clearing the way for the extension of full citizenship rights to Africans. On the otherhand, they presumed that until that point was reached, segregationist devices were essential to ensure that the civilizing process would be gradual and equitable to Africans and whites alike. Effective in forging an ideological consensus among whites with varying conceptions of segregation, these antithetical purposes promoted a palpable degree of confusion and inconsistency in matters of practical administration.

This ideological tension was strongly reflected within the DNA. Department officials shared the general white consensus that Africans in the urban areas should be subjected to some form of segregated administration. At the same time, many senior officials were not entirely at ease with policy developments in the 1920s and 1930s, due to their concern that vital aspects of segregation policy conflicted with the department's mission to oversee the civilization of Africans. The liberal officials who controlled the department therefore either favored one tenet of African policy over the other or veered indecisively between the two tenets in the interwar years. One consequence of this ideological ambiguity was that individual advocates of segregation were free to emphasize either of the prongs leading to two antithetical futures. Thus, in sharp contrast to its regimentation in the apartheid era, the department in the segregationist era did not speak with one voice about the nature and direction of segregation policy. It sometimes strongly endorsed the principle of African urbanization as an inevitable correlate of the civilizing process while, under the banner of protecting white interests, it subjected the process to increasingly stringent forms of control. Indeed, departmental ambivalence about much of the legislation proposed and enacted in the 1920s and 1930s illuminates its self-image as an instrument for promoting the welfare of Africans.

The second reason for the lackluster response in urban administration stemmed from a basic uncertainty in its organization. While the Act of Union held the department responsible for the overall field of Native affairs, the Native (Urban Areas) Act of 1923 specifically held local authorities responsible for urban administration. In effect, responsibility for urban administration was dispersed across numerous urban centers, where each local authority was expected to execute policies formulated by the central state. This bifurcated arrangement in urban administration undermined attempts to standardize urban controls nationwide. Local authorities refused to cooperate with the department's policy of unfunded mandates while the department, unwilling to limit the autonomy which local government had traditionally enjoyed, studiously avoided a more interventionist posture in urban administration.

Third, rounding off these weaknesses was the department's absence of experience in urban administration. Before it arose as a distinct field in the 1930s—signaled by the establishment of an Urban Areas Branch within the department in 1935—urban administration played second fiddle to the department's long experience in managing the reserves in a paternalist, rural-oriented approach which proved of little value to urban administration. In effect, African urbanization—a relatively recent phenomenon in the interwar years—caught the department off-balance. At ease with administration in the reserves, it was ill-equipped to take the lead in addressing the problems in urban administration. Accordingly, the department took the path of least resistance, leaving it to the initiative of local authorities to resolve the problems steadily multiplying on their many doorsteps. Despite three important developments—the passage of the Urban Areas Act in 1923, its significant strengthening in 1937, and the establishment of an Urban Areas Branch in 1935—the department did not generate a consistent tradition in urban administration. Although the signs of such a tradition did emerge among municipal Native administrators in the late 1930s and 1940s, these officials were united more by a sense of being neglected by the central state than by a distinct and concerted set of propositions about the goals of urban administration.

The noninterventionist compass which shaped the evolution of urban administration had important ramifications for the trajectory of the segregationist state. This was especially true in the 1940s, when concerted attempts were made to draw the department more into the regulation of the racially repressive labor market. The department's self-conception as a paternal steward concerned with maintaining a just balance between the races was sorely tested by this development. Unwilling to renege on it what considered to be its responsibilities toward both whites and Africans, its response to the unfolding of the repressive labor market was ambivalent, marked by attempts to ameliorate the most onerous legislation while promoting what it considered to be in the best interests of Africans.

The significance of this ambivalence was its contribution to two developments which would be centrally implicated in the National Party's electoral victory in 1948. First, the department's dilatory response to urban administrative issues antagonized the white farming community, particularly those in the Transvaal and the Cape, where labor-hungry farmers were decisive in giving the NP a handful of slim majorities that played a crucial role in the party's razor-thin victory in the general election of 1948. Second, the department's preference to avoid confrontations with local authorities, almost all of whom were generally indifferent to their duties in Native administration, stoked grassroots opposition in overcrowded and unsanitary urban locations, providing the material context for popular militancy to fuse with the development of an urban-based black nationalism. Together, agrarian anger and African working-class mobilization led to the unprecedented politicization of urban administration in the 1940s, placing municipal administrators at the forefront of state policy and facing the increasingly volatile mobilization of Africans in the urban areas.

Ambivalent Intervention: The DNA and the Labor Market

The impression conveyed by writers such as Lacey (1981) and Fred Hendricks (1991) that the Department of Native Affairs in the late 1920s and 1930s was systematically fashioned into a coercive institution for facilitating the superexploitation of Africans is only partly true. The mood within the department was not entirely consonant with the state's growing commitment to authoritarianism in the conduct of African affairs. For one thing, changes to the administrative structure of the department itself had not kept up with the flurry of legislation and administrative regulations. Apart from formally consolidating the minister's authority—distinguishing its own judicial powers more clearly from those of the Department of Justice, which the department perceived to be its main rival within the state—and undertaking a number of minor administrative modifications, the DNA remained as understaffed, ineffective, and overshadowed by the Department of Justice as it had been prior to passage of the Native Administration Act. When viewed as a distinct field, urban administration remained essentially an ad hoc affair in the segregation era.

A crucial upshot of the department's ideological objection to extensive interventions in the economy was the destabilization of urban administration. This destabilization was a gradual development. It assumed importance when African urbanization increased markedly immediately after World War I, and reached its high point in the course of World War II, when African urbanization increased rapidly in response to the twin forces of rural immiseration and the urban demand for cheap African labor. Urban administration ascended steadily, if uncertainly, to the forefront of the department's agenda. By the end of World War II, it was clear that urban issues had eclipsed the department's emphasis on administration in the reserves. But the formal importance of urban issues was not matched by a concerted administrative response. Accordingly, urban Native administration—or rather, its chaotic condition—become a nemesis of the segregation state.

By the mid-1930s, a battery of legislation formally enjoined the department to play a leading role in the organization and management of the labor market. Three major pieces of legislation were of particular importance: the Native Labour Regulation Act of 1911, the Native Service Contract Act of 1932, and the Native (Urban Areas) Act of 1923. In theory, these acts established a triadic structure that, had it worked, would have enabled the department to facilitate the supply of labor to each sector by ensuring that each had preferential access to a designated pool of African communities. In practice, as we see below, the department fell far short of this formal goal.

A simple snapshot of this formal model would yield the following picture: the Native Labour Regulation Act would have ensured that the mines drew their labor from the reserves; the Native Service Contract Act would have guaranteed white farmers access to the African communities within or close to the white rural areas; and the Urban Areas Act would have ensured that urban industrialists benefited from a carefully regulated flow of migrants from rural to urban areas as well as from a gradually increasing urban African population.

But this model was never fully established in the interwar years. Moreover, to the extent that attempts were made to cleave to this model, its fatal flaw was the intersection of two opposing developments: the state's general preference for limited interventions in the economy and changes in the rural areas that drove rising numbers of Africans into the urban economy. The department responded to the latter development with characteristic ambivalence. Even as it explored more effective and more authoritarian measures for bringing the urbanization process under control, it refrained from imposing a coercive regime in the urban areas—principally because it did not wish to interfere with the traditional autonomy of local authorities, but also because it was in general support of African urbanization.

The DNA, Gold Mining, and White Agriculture

The Native Labour Regulation Act governed the employment and recruiting conditions in the mines and was therefore expressly designed to maintain the system of temporary male migration between the reserves and the mining areas. As its preamble stated, the central objective of the act was to ensure as far as possible a more equal division between conflicting industrial mining and agricultural interests in different parts of the Union.... The act therefore centralized controls relating to the migration of African mineworkers from the reserves, their accommodation in compound housing on the mining premises, and their return to the subsistence areas. It should be noted, however, that the actual process of recruiting workers and channeling them from the reserves to pass offices in the Transvaal (where a routine medical examination was undertaken and the service contract was registered) was left either to the industry's private labor recruiting organizations or to private labor recruiters: these were not the department's duties.

The DNA was thus never directly involved in either the recruiting or the production process of gold mining. Instead, the Chamber of Mines developed its own recruiting apparatuses—the Witwatersrand Native Labour Association (WNLA) and the Native Recruiting Corporation—going to extraordinary lengths to assert its lockhold over its far-flung sources of labor. The department's role in the industry was confined to the placing of Welfare Officers on mining premises. These officers ensured that accommodation of workers conformed to guidelines and acted as links between black workers and the Director of Native Labour (DNL), the official who (under the terms of the Native Regulation Act of 1911) managed the Native Labour Department established within the DNA. Officials in the Native Labour Department also registered the service contracts of migrant workers. Beyond such functions, however, the DNA's involvement in the gold mining industry was minimal; its principal role was limited to what it described as welfare functions. Because they coincided with and reinforced the department's paternalist responsibilities for Africans, these tasks were generally taken seriously by the sizable number of men allocated to the Director of Native Labour (three Native Commissioners [NCs], an Assistant NC, three Inspectors of Native Labour, and ten relatively senior officers in charge of pass offices). The department derived evident pleasure from the observation that Native workers had come to trust the Labour Officers who tended to their interests in the Union's labor centers.

In sharp contrast to its well-elaborated and cordial relations with the mining industry, the department's dealings with white farmers remained perfunctory and testy throughout the segregation years. The reasons for this lay in the organization of labor relations in the white countryside, and more particularly, in the central importance that the large majority of farmers attached to labor tenancy, a long-standing practice which the Native Service Contract Act sought to rationalize and protect. The department was less than supportive of this arrangement in white agriculture for two reasons. First, it argued that labor tenancy was an inefficient economic practice. It therefore persistently pressured farmers to stabilize their labor force by converting to efficient mechanized production and wage labor as quickly as possible. Not surprisingly, the department's praise was therefore reserved for the small number of more affluent farmers who, because they were better placed to take advantage of these costly innovations, enjoyed reputations as progressive farmers able to attract adequate labor by virtue of higher wages. Department officials were also uneasy with the physical abuse notoriously associated with labor tenancy, sometimes airing the view that labor tenancy was an inherently abusive institution (see below). Under these circumstances, it is not surprising that the Native Service Contract Act—which made labor tenancy the basis of capitalist agriculture and ensured that the cessation of the practice would be a conveniently gradual affair for the white farmer, drawn out over at least three decades after the act was passed in 1932—was shepherded through parliament by the Department of Justice and not by the DNA.

Ideologically impatient with the practice, the department took pains to distance itself from both the logic and the rigors of labor tenancy. It thus did little more than attempt to live up to the duties sketchily defined in the Native Service Contract Act. These were limited to the infrequent inspection of white farms and to the investigation of complaints by African farmworkers, whenever these arose. As we see below, even when it did cooperate with farmers desperately searching for ways to retain their farm labor tenants other than by improving wages and working conditions, its interventions tended to be more critical than supportive and were frequently couched in a pontificating and accusatory style which incensed labor-hungry farmers.

The department's different responses to the Native Labour Regulation Act and the Native Service Contract Act index its ambivalence about intervening in the economy. Sensitive about its paternalist obligations toward Africans, it distanced itself with notable vigor from the Native Service Contract Act, in particular. This anti-interventionist stance in white agriculture did not contradict the department's long-standing cooperation with the mining industry. The department took the position that its role in the mining industry was essentially to protect Africans, and therefore not very dissimilar to the Betterment programs it embarked on in the 1930s to stem the movement of Africans to the urban areas by shoring up the subsistence economies (see chapter 8). Having no similar powers on white farms, and being opposed to the harsh controls on which farmers relied to discipline Africans on their farms, the department consistently contended that the solution to the farm labor shortage lay in the market. Labor tenancy, it announced to various agricultural unions, should be done away with,...farm wages [should be] increased and market forces should be permitted to determine the employment preferences of Natives.

The DNA, Urban Industry, and Local Authorities

The Service Contract Act of 1932 and the Native Labour Regulation Act of 1911 were rural-based controls. In contrast, the Native (Urban Areas) Act was a measure for controlling Africans who were already in the urban areas or desired to enter these areas. Of the three pieces of legislation, therefore, only the Urban Areas Act addressed specifically urban questions of Native administration. However, if this crucial measure marked the genesis of urban administration as a distinct component of the department's activities, it also made clear that the department would not interfere with the strong pattern of municipal autonomy inherited from the colonial era. While it made the department formally responsible for formulating urban policy, the act entrusted all administrative functions to local authorities. As the act stood in 1923—prior, that is, to its repeated amendment over subsequent decades—urban policy hinged on three principles.

First, the act provided a legislative mooring to the argument that the presence of Africans in the urban areas should be limited strictly to their utility in the urban economy. As the infamous report of the Stallard Commission of 1921 expressed it, the Native should only be allowed to enter urban areas, which are essentially the white man's creation, when he is willing to enter and to minister to the needs of the white man, and should depart therefrom when he ceases so to minister. This approach gradually eclipsed the alternative urban Native policy championed by liberals drawn from the Cape, whose solutions favored converting Africans into a property-owning and politically moderate petty bourgeoisie in the urban areas. In opposition to this strategy for managing a potentially volatile urban African population, the Stallard doctrine denied Africans any unconditional rights in the urban areas; regardless of their material security, good police record, or length of residence, all Africans would be no more than temporary sojourners in the urban areas. As a number of observers have noted, the steady ascendance of Stallardism over the classic liberalism of the Cape reflected the prevalence of certain class interests in urban African policy at the time the Urban Areas Act was passed. White workers supported any plans which obstructed Africans' access to the urban labor market, fearful that the relative cheapness of African labor would undermine the civilized labor policy that enjoined employers to give preferential treatment to white workers. White commercial and real estate interests similarly viewed an expanding class of African property-owners as a threat and so supported Stallardism because it promised to preclude this outcome. Farmers also stood to benefit from the policy: by making it difficult for Africans to enter the urban areas, Stallardism reinforced the controls that farmers were demanding on the mobility of black workers in the rural areas.

A second focus of the Urban Areas Act was the policy of residential segregation it instituted for all urban areas. Capturing white unease about racial integration in the urban areas, the Stallard Commission expressed its concern about blacks and white living cheek by jowl in squalid shantytowns, striking a particularly alarmist tone about the pervasiveness of the practice in the Johannesburg area, South Africa's major industrial zone. Spurred by the outbreak of infectious epidemics in African residential quarters just after World War I, public authorities feared that the buildup of large, poor African communities in or close to urban centers would facilitate the spread of contagious diseases into the white communities. Residential segregation along racial lines was thus presented as a scientifically grounded defense against the spread of diseases, and a medical discourse about the transmission of disease from black to white bodies served as the idiom through which whites expressed their strong anxieties about racial propinquity in the Union's urban areas.

The report of the Stallard Commission proffered residential segregation as the antidote to inevitable miscegenation, but also as necessary to the strict administrative controls it recommended for monitoring the presence of Africans in the urban areas. The commission recommended that the housing market for Africans in the urban areas should be confined to the construction of townships owned and administered by local authorities, with no Natives to be allowed to reside elsewhere. This body blow to the liberal strategy of promoting an affluent class of moderate, property-owning Africans bolstered by the municipal franchise was incorporated into the Urban Areas Act.

Third, giving substance to a recommendation of the Stallard Commission, the Urban Areas Act provided for the creation of a segregated municipal infrastructure for regulating urban African policy. The act provided for segregation in three specific areas. First, it made it obligatory for local authorities to record all funds used for administering townships under a separate Native Revenue Account into which all fines, rents, and profits from the sale of African beer had to be deposited. Surpluses generated under this account—derived almost exclusively through the sale of African beer, recorded under a subaccount known as the Kaffir Beer Account—could not be transferred to the city's general account; however, the act did not prevent local authorities from supplementing the Native Revenue Account with funds drawn from the general revenue account. Second, to offset the municipal disenfranchisement of Africans, the act encouraged local authorities to establish a Native Location Advisory Board, which local authorities were to consult over specific areas of urban African policy. This political structure was intended to provide a limited forum to the small African petite bourgeoisie in the Union's urban areas, a mark of recognition that might inoculate them against the alarming spread of militant rhetoric in the Union's townships in the 1920s.

The third component of administrative segregation at the municipal level was the act's stipulation that each local authority should establish a municipal Native Affairs Department (NAD). Run by a Manager, a municipal officer devoted exclusively to African affairs, who was assisted by a location superintendent who managed the day-to-day affairs of administration, the municipal NAD was intended to promote expertise in urban Native administration, establish contact between urban Natives and the Town Council and serve as a link between local authorities and the Department of Native Affairs. This development signaled the importance attributed to the new field of urban administration in the early 1920s and the perception that the field required a degree of specialization and professionalization entirely foreign to administration in the colonial era. In practice, the establishment of NADs also isolated and marginalized African affairs from the normal concerns of white city councils—as is borne out by the indifferent attention and fiscal resources that local authorities reserved for African administration well into the 1940s. Known as Non-European Affairs Departments (NEADs) in areas where African, coloured, and Indian affairs were lumped together, these municipal departments were entrusted with an expanding volume of increasingly complex work in the 1930s and 1940s, while local authorities settled for a policy of benign neglect in African administration. Unsurprisingly, NAD personnel soon came to perceive themselves as Cinderellas, or poor cousins of the state. Suspended between increasingly aggrieved African townships and, to quote Brian Porter (Johannesburg's Town Clerk), the mean and niggardly policies of local authorities, NADs were all but left to their own local devices while the DNA struggled to sharpen its competencies in urban administration. By the outbreak of World War II, however, NADs had developed into important mechanisms for managing African urbanization and for dealing with the labor requirements of urban employers.

Urban employers sounded two themes as the segregation years wore on. On the one hand, they relied on the more vulnerable and cheaper supplies of migrant labor to meet their needs for unskilled manual labor and to perform work that the more discriminating and educated urban population routinely rejected. Their official discourse, however, harped on the virtues of the free market and the superior efficiency of detribalised urban labor. But, as liberal organizations frequently pointed out in the 1930s and 1940s, these were not discrete segments of the labor market: the sheer presence of large numbers of migrant labor inevitably exerted a downward pressure on the living standards and work ethic of the settled population. Loyalty and discipline, E. Burrows observed in a memorandum submitted as evidence to the Fagan Commission in 1948, are qualities to be found only in a stable labor force. After emphasizing the problems and constant costs of retraining, the loss of working time, and the complications that migrant labor raised for the establishment of good relations based on mutual understanding, Burrows concluded that the low level of skill and the rapid job to job movement result in low output and low income. Low per capita output from African workers means an unnecessarily low national income. Moreover, low earnings are often reflected in poor housing and inadequate diet, and ultimately in reduced working efficiency....[T]he situation tends to reproduce itself. Burrows' discussion on aggregate effects suppresses the qualitative differences between settled and temporary labor as well as the effects of the latter on the living standards of the urban community. A memorandum submitted by the South African Institute of Race Relations (SAIRR) to the commission dealt at length with precisely this point:

[Urbanized Africans] are in competition with migrant laborers in the labor market, and since they are out-numbered by the latter, the customary rate of wages is determined by what the migrant laborer will make....Since the Reserves reduce the necessity of making the income received by the migrant in the town adequate to support his family all the year round, a depressing effect on wages results. This depressing effect is felt most strongly among urbanized natives and incidentally, by landless migrants, who have none of the resources of their rural brothers, who draw wages on the same scale and have to support their families in the urban areas at high costs of living....Thus, the urbanized natives have been forced below subsistence levels and even when working continuously, find that their earnings are not adequate to meet their expenses.

A growing concern thus emerged in the 1930s and 1940s about ways to limit the impact of semi-proletarianized workers on the urban African community without restricting employers' access to adequate volumes of migrant labor. The Urban Areas Act was directed toward this end. By the late 1930s, the act embodied four administrative mechanisms for controlling African access to the urban areas and for regulating Africans within the urban areas: the registration of labor (or service) contracts between employer and worker; the issuance of workseekers' permits, which authorized Africans in the rural areas to enter an urban area; influx controls and powers to evict Africans from urban areas; and exemptions from the pass system, granted at the department's discretion.

Primary responsibility for administering the regulations issuing from the Urban Areas Act was vested in municipal officials—specifically, in officials in the municipal pass office, managers of the municipal NAD, and the Superintendent of Locations. A basic tension was therefore built into the structure of urban administration. The 1923 Urban Areas Act vested local authorities with powers to carry out policies with national implications. This arrangement raised the possibility not only that the responses of local authorities might differ, but also that the narrower concerns of local authorities might conflict with the department's wider responsibilities for the functioning of the African labor market.

The Politicization of Urban Administration

Agriculture and the Labor Shortage

Even before the 1940s, a major deficiency in the state's African policy had become apparent: whereas the solution was designed to regulate the competition for labor between the mining and agricultural sectors, manufacturing industry had firmly intruded into the labor market, displacing the mining sector as farmers' greatest competitor for cheap African labor. A report of the Native Farm Labour Committee revealed the scale of the shift. In 1933, Basutoland had supplied 15,237 workers to agriculture and 25,803 to the mines. Two years later, 33,130 workseekers headed for the mines, while agriculture's share had dwindled to a paltry 3,782. What troubled farmers most, however, was the committee's ominous conclusion: The farms are now becoming what the Reserves used to be, sources of a labor supply. But whereas formerly it was the farms that drew from the Reserves, it is now the towns that draw from the farms.

The report concluded that the three-cornered competition for labor between farmers, industrialists, and miners effectively boiled down to a competition between farmers and miners on the one hand, and industry on the other. The mines, however, were not perturbed by the growing volume of workseekers heading for the urban areas: its unassailable dominance over the South African economy was buttressed by a well-developed labor-recruiting system within the reserves and throughout Southern Africa. In contrast, white farmers suffered directly. Accordingly, except for districts close to rural African communities, virtually all farmers reported seasonal labor shortages throughout the 1930s. In sometimes berating tones, meanwhile, the department justified its refusal to intercede on farmers' behalf by stressing the importance of meeting industry's labor requirements and by pointing out that low wages and poor working conditions in agriculture were the major reasons for the desertion of farmworkers.

Farmers were in an invidious position. The rise in the value of land, the shrinkage in farm size, the necessity for more intensive cultivation in the absence of initial cash, declining agricultural prices, and burgeoning debts were compelling the majority of them to rely all the more on labor tenancy. Many attempted to resolve these major problems by escalating the exploitation of their workers. Particularly in the Transvaal, farmers cut down on the grazing rights of labor tenants and required them to work the additional land released without raising cash wages correspondingly. The imposition of tribal authority structures—such as using chiefs to recruit farm labor, reducing women to the legal status of minors, activating the whipping clause that permitted the farmer to thrash any member of the tenant household, and subjecting the entire tenant household to the rigid control of the farmer by defining the labor unit as all members in the household—were harsh measures that severely intensified the farmer's domination over tenants. At the same time, tribal controls also clashed with a system in which the household head had control over stock and grazing rights, leaving younger sons with little to gain materially over the course of time. The erosion of labor tenancy therefore projected bleak prospects for sons who would one day enter into their own service contracts. In this context, the higher wages and better conditions of urban employment became all the more attractive, and as farmers complained before the Native Farm Labour Committee, in letters to the department, and in agricultural journals, permanent desertion was the inevitable result.

Complaints to the department generally solicited the response that desertion, although illegal, was also understandable. For example, an appeal to the department's Land Section (in 1936) to coerce Africans residing on land owned by the South African Native Trust to enter into farm labor elicited the biting response from Smuts that the appeal smacked too much of slavery to be accepted in a free country where every man is at liberty to sell his services in the highest market or to use them to the best advantage to himself. While a circular dispatched by D. L. Smit in 1936 instructed Native Commissioners to clamp down on deserters by withholding travel permits, it did so largely because many of these Natives arrive in Johannesburg, Pretoria and elsewhere without passes or documents of any kind and in a destitute condition....The result is that the work of our officers [in these urban areas] is greatly increased. Urban administrative problems, not those of farmers, were the primary concern.

The urban bias in administration prompted some farmers to devise their own means to retain labor without increasing wages. Farmers in the Northern Transvaal, it seems, initiated an experimental system of Native farm labor bureaus well before the Native Service Contract Act was passed. A bureau consisted of farmers, Native Commissioners, and native Councils—that is, chiefs and indunas [tribal supervisors] chosen by the different tribes in collaboration with the Native Affairs Department. The main object of these bureaus was to provide regular labor for farmers, to combat recruiting agencies for the mines, and, as the Native Commissioner for Potchefstroom informed the Secretary for Native Affairs, there is the further laudable object of absorbing all available ‘picanin' labor and preventing as far as possible the drift of these youths to the larger centers. Since farmers proved hostile to negotiating with chiefs and indunas, the Native Services Contract Act extended this idea into a system of Native Farm Labour Advisory Boards, but eliminated African representation. These boards, however, did not develop into a real live force, as the Secretary for Native Affairs (SNA) had expected them to. Because they lacked authority to restrict the labor tenants' movements or to make their own decisions binding, and possibly because Native Commissioners persistently raised the issue of wages and working conditions, the boards were stillborn and did little to prevent the departure of rural Africans for the urban areas.

Existing side-by-side with the eroding system of labor tenancy and the farm labour shortage was the phenomenon of large communities of Africans squatting on the land of absentee landlords. African access to the land as autonomous subsistence producers incensed farmers. Precluded from owning land outside of the reserves, beleaguered African peasants clung tenaciously to the land, and their resistance delayed plans to subject them to employment on white farms. A 1939 departmental survey of Transvaal farm districts initiated by the Controller of Native Settlements revealed that the majority of farming districts were in favor of enforcement of Chapter IV of the Native Land and Trust Act, the provision that enabled the department to uproot squatter peasants and either disperse them across white farms or remove them to the reserves. The department, however, was compelled to refrain for two reasons. The first was the lesson drawn from the Lydenburg affair of 1937-38, when, in response to farmers' requests, the Lydenburg district was proclaimed under the terms of Chapter IV. Less than a year later, the proclamation was suspended in the wake of the violent response of squatters and their accelerated movement into adjacent areas not affected by the proclamation. Smuts, however, declined to respond with anything that smacked of an element of coercion.

Thus, the Service Contract Act and Chapter IV were unable to generate a satisfactory and regular supply of agricultural labor or to subject labor tenants adequately to farmers' control. More than any other sector, therefore, farmers developed an abiding and obsessive concern with forms of identification that would easily distinguish farm laborers from the labor force in the urban areas. Correspondingly, the deliberate destruction of identification documents by farmworkers or their circulation among needy counterparts in the urban areas were simple but effective forms of resistance that jarred urban controls. The inefficiency of the process of bringing pass law violators to book is designed to frustrate the farmer, a correspondent to the Farmers Weekly complained in 1937. The discovery of a passless Native in an urban area initiated a lengthy correspondence between the municipal Pass Officer and the Native Commissioner in the labor tenant's declared district of origin. While the department determined the details of a particular case, the passless Native was allowed to seek and take up urban employment. The problem was that in many cases the Native Commissioner of the home district intimates in his reply that the Native is required in that District to fulfill his contractual obligations. The Native, if found in the urban areas, is then sent for and warned to return to the farm, but if he fails to do so the department is powerless to take further action and it then lies with the farmer to initiate proceedings for a contravention of the Masters and Servants Laws and/or the Native Service Contract Act. . . . Farmers loathed this cumbersome procedure, which called for their continuous intervention in numerous individual cases. The resultant need to travel to town, where the Native Commissioner would hear the case, appeared to farmers to be a sort of double penalty, since urban employers already had abundant labor. For transfixed farmers, replies to the effect that desertion could not be stopped by the mere introduction of a regulation conjured up the nightmare of permanent competition with urban industry for African labor. Matters came to a head in the World War II period when Native Commissioners were instructed by the SNA to relax the application of the pass laws.

Although the department was not completely without sympathy in these times when food production is of such vital importance, its responses incensed rather than appeased farmers. A memorandum drawn up by W. J. G. Mears (SNA, 1944-49), then Under-Secretary for Native Affairs, outlined the ways in which the Department has endeavored to be of assistance to farmers: a list containing the addresses of known labor recruiters had been compiled; Native Commissioners had been instructed to help farmers where possible; farmers had been advised to improve wages and working conditions; labor gangs were to be made available to farmers under conditions of work and remuneration prescribed by the Department, and a farm labor scheme was to be put into operation soon. In short, the department was not prepared to move beyond the stricter application of existing provisions, as Piet van der Byl, Minister of Native Affairs, was instructed by Smuts to reply.

The farms, therefore, continued to expel labor out of the rural areas for the rest of the 1930s. Desertion, the most effective form of resistance to the direct domination of the farmer, escalated in the 1940s, making the farms the single largest source of labor for the urban areas. The act of desertion was relatively easy: as Smit observed, farmworkers could simply remain undetected in the urban areas, walk or catch a train to the nearest urban area, or, in some extreme cases (as the Cape Town City Council complained), even catch a boat from one port city to another. Once in the urban area, deserters became enmeshed in the general confusion in urban administration in the 1930s and 1940s. It was relatively easy to slip between the many cracks in urban administration, find employment in the wartime conditions of high employment, and blend with the thousands of other passless Natives living in squatter camps, shantytowns, and overcrowded municipal locations.

Africans and Urban Administration

Despite growing concern that the unsanitary conditions prevailing in African residential areas posed threats to whites and that these conditions were fueling working-class opposition to the state, the department failed to devise an effective administrative infrastructure for absorbing Africans into the urban areas. Its policy of permitting local authorities every latitude in the control of the Native population essentially permitted local authorities to ignore their civic responsibilities toward the black working class. Similarly, its disinclination to become actively involved in the urbanization process even after the passage of the Urban Areas Act undermined the emergence of a coherent administrative approach to the urbanization of Africans. Consequently, notwithstanding the introduction and repeated strengthening of the influx control and expulsion provisions of the Urban Areas Act between 1923 and 1945, urban administration remained unsynchronized and incapable of detecting or absorbing the streams of Africans who entered the increasingly squalid and febrile urban areas, either as migrant workers or as permanent settlers.

With thousands of Africans in the urban areas trapped between a rising cost of living and the absence of housing accommodation, the inadequate provision of social services in the urban areas became a focal point of resistance. Of the 816,456 Africans on the Witwatersrand in 1946, a mere 18.1 percent were housed in townships constructed by local authorities; the overwhelming majority lived either as squatters on vacant land or as lodgers (subtenants) in the houses of others. The Manager of Johannesburg's NEAD, L. Venables, estimated in 1946 that the city alone had a waiting list of 17,000 families requiring accommodation. Since at least 3,000 families entered the city every year, the Johannesburg City Council (JCC) was informed that it would have to construct 60,000 family houses over the next five years to clear up the immediate shortage, while thousands more would have to be built annually to cope with incoming families. It was the housing shortage that African grievances clustered around in the 1940s: it generated waves of protest against the lack of adequate accommodation, high rental fees, and availability and cost of transport to and from the city center. Even before the war ended, the Union's major urban areas—inevitably the Witwatersrand and Reef areas in particular—remained in a state of constant turmoil over issues basic to the reproduction of the black working class.

In August 1943, a bus boycott in Johannesburg's Alexandra township indexed the scale of opposition to inroads in the already tenuous income of the township's sixty thousand residents. Rather than submit to an increase in bus fares from 4d. to 5d., thousands of workers daily trudged the eighteen miles back and forth to the city center. The success of this boycott carried over into a second bus boycott in November 1944, when, aided by the Alexandra branch of the Communist Party of South Africa (CPSA) and the Transvaal ANC, residents once again walked the distance, this time for seven weeks. Struggles between African workers and the city council over transport continued well into the 1950s, marked by sporadic violence, deaths, and the large-scale involvement of location inhabitants dependent on cheap mass transport. Residential segregation and the siting of locations away from the city center made transport a central issue. Alexandra inhabitants, for example, spent three and a half hours per day commuting to and from the city center, facing an unpleasant transportation staff, courting danger in unsheltered terminals, and walking considerable distances to central terminals in the township and the city—all for a service that often consumed the second largest chunk of the weekly budget, after rent. Although transport protests were pervasive throughout the country, T. Lodge notes that there appears to have been a correlation between boycotts and locations in which freehold property rights existed for Africans: Sophiatown, Lady Selbourne, Alexandra, and Cato Manor all experienced bus boycotts during the 1940s and 1950s. These townships—older, more established, and accommodating more skilled workers and the petty bourgeoisie—possessed a class base whose moral and physical resources were directly threatened by the JCC's plans to relocate them to the perimeter of the city, as well as by the pressure of newcomers searching for accommodation wherever they could find it.

The resulting friction between established residents organized in civic associations such as Standowners, Tenants, and Subtenants Associations was sometimes of a high order, and competing claims for leadership or for official recognition were common. Alexandra alone appears to have sported, in addition to a number of social clubs, at least ten Standholders, Landlord, and Tenant Associations whose divergent interests—for example, in rent control—meant that disputes and quarrels between the various leaders of organizations and associations in Alexandra Township were rife, as the Secretary of the New Standholders Committee complained to the Minister of Native Affairs.

The urban African community therefore was not homogeneous. Tensions existed, for example, between the established urban community and the incoming migrant population. More securely ensconced within the urban areas, the older sections of the urban population did not feature prominently in the more militant forms of opposition that emerged in the 1940s, limiting their involvement to high-profile but carefully organized events such as bus boycotts. In contrast, the squatting epidemic of the 1940s chiefly involved the stratum of subtenants. Over the years, the de facto existence and continuing growth of this category had compelled many city councils to ignore location regulations that permitted only family members to occupy accommodations in their locations. Regulations formally prohibited persons who are not owners or occupiers of location property nor members of the families of such people from residing in locations. As a rule, however, lodgers in municipal housing tended to be related to the tenant, in contrast to shantytowns, where the majority of tenants and subtenants were likely to be unrelated.

Native Commissioners in the Witwatersrand identified subtenants—the loafer-lodger type—as the prime actors in the squatting movement. This was not entirely inaccurate. More prone to periodic unemployment and more vulnerable to influx controls, subtenants played a role in meeting industry's need for unskilled or casual migrant labor. They proved to be a militant category, attracted by the flamboyant and militant leaders who emerged in the shantytowns in the early 1940s. Apart from the eccentricities that characterized the personal styles of leaders such as A. Mohase, A. Ntoyi, and the doyen of the squatter leaders, James Sofasonke Mpanza, the squatter movements that erupted in Johannesburg, in particular, quickly displayed an insouciance and rudimentary organizational structure that alarmed administrators and held the police at bay. Parallel administrations run by dangerous fanatics had begun to emerge in the vacuum caused by the breakdown of civil administration in Johannesburg's shantytowns, the Commissioner of Police urgently warned the Secretary of Native Affairs in 1944. He had, he noted, squatter leader James Mpanza foremost in mind. In 1944, Mpanza orchestrated the first large-scale squatter movement, exiting from Newclare and Kliptown onto vacant land in Orlando. In return for an initial 2d. entrance fee and a weekly rental of 2.6d., the Sofasonke (We Shall All Die Together) Party permitted those willing to swear the obligatory oath of allegiance to Mpanza the privilege of erecting any form of shelter on an allotted stand. Several squatter leaders duplicated Mpanza's innovative organizational structure of unlicensed trading, designating leaders and committees who appointed camp guards and provided crude sanitary conveniences, and levying charges on occupants for the maintenance of these services.

It was not surprising that subtenants found such initiatives attractive. Subtenants were highly vulnerable to exorbitant rents and to summary eviction by registered tenants of municipal houses and homeowners in areas where freehold rights existed for Africans. Despairing that municipalities would ever clean up the horrendous conditions in squatter camps, let alone construct new housing that they could afford, subtenants were drawn by Mpanza's successful confrontations with Johannesburg's NEAD and the South African Police. After squatters had seized partially completed and temporarily vacant housing in Pimville and Alexandra, C. Alport (Director of Native Labour) and Graham Ballenden (Manager, NEAD, Johannesburg) inspected Mpanza's camp and reported to the city council that Shantytown is an established fact: 95 percent of the families resided in extremely small, roughly made shelters constructed merely of poles and hessian....The balance are made of mielie meal stalks and old pieces of tin. Later, Mpanza threatened the outraged city council with the diversion of municipal water and the erection of windmills. More squatter camps developed, several of which originated as rival or splinter groups. In 1947 a rent boycott emerged in Jabavu and in the Moroko Emergency Camp—established to relocate and centralize squatters from the various camps—leading to the killing of three policemen in the ensuing violence.

Squatter camps began emerging in peri-urban areas (i.e., the outskirts of urban areas) throughout the country, where the Urban Areas Act did not apply. The Cape Town City Council's confidence in the strategy of restricting entry into the municipal area by soliciting the intervention of the railways, who agreed not to issue tickets to travelers without prior proof of employment in the Cape Peninsula, evaporated as workseekers devised alternative routes. The high profile of Johannesburg's seventy thousand squatters and the burgeoning squalor of shanty settlements and locations filled beyond capacity publicized the breakdown in urban administration. By the late 1940s, local authorities in the industrial zones of the Transvaal had abandoned hope of gaining immediate control over the housing issue.

One consequence of grassroots volatility was the invidious position in which Native Advisory Boards found themselves. Advisory boards were vested with just two functions: to consult with city councils over location registrations and to elect a representative of the urban areas for each province to the Native Representative Council (NRC). The nebulous status of advisory boards may be gauged from the failure of the 1924 (and subsequent) location regulations to elaborate on these two functions. With the support of the liberal establishment, who viewed advisory boards as a vehicle for nurturing the small class of African property holders and traders in the townships, the annual Location Advisory Boards Congress restricted itself to its tepid constitution, pleading with unwavering regularity for official municipal recognition of the advice and recommendations it proffered in accordance with regulations published under the Urban Areas Act. The establishment of alternative residential organizations reflected the general rejection of these bodies. With the support of the CPSA, Subtenants Associations developed strong antipathies toward advisory boards. The Joint Council of Europeans and Africans, a liberal body with no interest in exaggerating the gulf between white city councils and Africans, informed W. J. G. Mears, the SNA, in 1946 that it was no use blinking the fact that Advisory Boards do not command the respect and confidence of the African people. Embattled Native Commissioners fruitlessly attempted to subject Subtenants Associations and shantytown inhabitants to location regulations. Major F. Rodseth (Under-Secretary for Native Affairs) distributed a circular to NCs in the Witwatersrand area advising them to give their support discreetly to the more reasonable ones who are open to discussion and...accept the authority of the NC.

Several boards, such as Johannesburg's Joint Advisory Board (representing all boards in the city's locations), appear to have adopted a more aggressive stance in their dealings with local authorities. In 1946 and 1947, in keeping with its decision to exploit prevailing sympathies to representative structures, the CPSA either actively supported Subtenants Associations in their electoral stands against board members or advocated boycotts of advisory board elections. As early as 1939, the Joint Advisory Board had passed a resolution exhorting boards to go ahead vigorously with the fight against the gross greed of local aggression and defiantly asserting its refusal to become a henchman for oppression. According to J. E. Mathewson (Manager, NAD, Benoni), who was a staunch protagonist of the advisory board system, participation in advisory board elections varied enormously, involving anywhere from 1 1/2; to 70 percent in different localities. Johannesburg, according to the president of the SAIRR, appears to have averaged about 10 percent participation.

In the postwar period, several advisory boards recognized the NRC's decision to adjourn indefinitely, in reaction to the brutal quashing of the mineworkers' strike, and the Orlando Advisory Board actually adjourned sine die itself. At this time, the Johannesburg City Council had recently suffered a series of losses in court when the validity of location regulations had been contested. Gloomily assured of a hostile reaction from the Joint Advisory Board, it hung back from promulgating new regulations because any amendments will, in turn, only become the object of further litigation. But carelessly worded location regulations made it difficult to secure convictions for violations of proclamations prohibiting squatting and the uncontrolled erection of shelter, prompting the Johannesburg City Council to complain that the Courts tended towards a sympathetic view of the plight of squatters. An official report cited two additional reasons for the lack of any concerted state response: the costly legal burden of proving that each individual squatter was guilty of a violation and the stolid refusal of local authorities to accept responsibility for the accommodation of even legally employed Africans.

The radicalizing of the Union's black residential areas was reflected in similar changes in national political organizations. A. B. Xuma's election to the presidency of the ANC in 1940 marked the eclipse of that organization's concerns, under Selope Seme, with the African chieftaincy and administration in the reserves. This development signaled the rise of an urban-based program of opposition spearheaded by the younger generation of urban intellectuals. Disillusioned with institutional protest and the failure of the anti-pass campaigns of 1944 and 1946, the ANC turned to a policy of noncollaboration with state institutions. Under the guidance of young intellectuals such as Anton Lembede, Nelson Mandela, and Oliver Tambo, the Congress Youth League, established in 1942 within the ANC, brought these tendencies to the forefront of national black politics and pressed for the rejection of segregation at all levels. The NRC—the advisory body established as a compensation for the disenfranchisement of Africans in the Cape in 1936—also reflected the militant postwar mood: in 1945, it abandoned its focus on the African vote in the Cape and resolved instead to concentrate on the legitimate needs and aspirations of the permanent rather than temporary urban Africans.

At a meeting which included the SNA, the Director of Native Labour, and the Commissioner of Police, L. I. Venables (Manager, Johannesburg NEAD) argued that the problems of urban administration, and principally the housing crisis, were deeply implicated in these developments: Housing is the dominant need....Communism is being eclipsed by a growing nationalism which in my view is a dominant influence in widening the gap between the Natives and the Europeans....The movement is out of control today. They are recruiting men for a guerrilla army based on the plan of the Jews in Palestine. These youngsters are fanatical and don't fear death. The Commissioner of Police amplified on the link between housing shortages and political unrest, claiming that squatter leaders were coercing homeless Africans into paying as much as £6.10.0 for a lorry to transport their goods to the site and then they are required to pay £3 in advance for their supposed rights to the land. He summarized developments: I have thought over the question of leaders and who can be approached. Certain Natives [i.e., advisory boards] are supposed to be leaders but they are not backed up by the Natives themselves. Some have lost prestige and have been thrown out. There is a type of business leader growing up. Some of these have not been fiery enough and they have been rejected. The African Youth League is to my mind a very dangerous organization. It is imperative that the housing issue should be solved as soon as possible.

The role of the municipal Location Superintendent in the day-to-day administration of locations became a serious issue. Caught between the agility of their city councils and of the DNA in evading final responsibility for location administration and housing construction, superintendents were indeed in a difficult position. One superintendent was moved to complain, Local authorities continually complain that they are ‘the Cinderellas of the State', but we [Superintendents] are the real sacrificial lambs of urban administration. At the very bottom of the administrative hierarchy, superintendents were also at the forefront of the battles between Africans and the local state, and so they deeply resented the low status attached to their occupation. They invariably had no knowledge of the vernacular, possessed an unreliable grasp of legislation and the regulations that clothed them with extraordinary powers over location inhabitants, and battled constantly with the advisory boards whose meetings they chaired—to the displeasure of board members, who regularly requested that the superintendent should be replaced by a member of the town council. Under the terms of regulations issued under the Urban Areas Act, location superintendents regulated the issuance of site, residential, tenant, subtenant, and visitor permits, which conferred the respective forms of legality to Africans in their locations. Such duties were inherently provocative and so inevitably pitted the superintendent against his wards:

In addition to the management of locations, the Superintendent may have to enforce regulations dealing with the registration of service contracts, the housing accommodation of Africans outside locations, municipal beer halls or the control of domestic brewing. Much of his work, therefore, consists of carrying out a national rather than a municipal policy, and one of his difficulties is that his functions as civil administrator are not easily distinguishable from his role of policeman and supervisor of restrictive legislation.

Complaints against superintendents escalated during the war years as attempts to impose order in locations resulted in the technique of nighttime pass raids and imperious searches for evidence of illegal beer-brewing.

Local Authorities and the DNA

Agrarian dissatisfaction and African oppositionfirmly focused national attention on the unsatisfactory state of urban administration. Sporadic at first, this attention became acute as the Union's largest urban centers entered into a period of political turmoil, with the country's urban African population doubling from 587,000 to an estimated 1,150,000 between 1921 and 1936. By the mid-1930s, the Urban Areas Act had gravitated away from the social and medical anxieties that had informed its passage in 1923; thereafter, as its amendments in 1930 and 1937 made clear, urban administration was increasingly dominated by influx control concerns. But, despite manifest evidence that urban administration lagged far behind these legislative accomplishments, the department did little to address a central problem in the organization of urban administration: the ambiguous and overlapping concerns of the central and local agencies in the management of urban Native administration.

Although local authorities were clearly subordinate to the Department of Native Affairs, in the 1930s and 1940s the department took the position that local authorities were primarily responsible for the administration of urban policy, while its own major focus was the administration of the reserves. Speaking in his capacity as Acting Minister of Native Affairs in 1937, Smuts informed a gathering of municipal officials that the problem of influx controls was essentially an urban one and your locations concern you directly more closely than they concern the rest of the community.... Furthermore, administration threw up technical and practical difficulties which only municipal officers know about.... Local authorities cited a lengthy history of autonomy from the central state, but at the same time demanded greater assistance in the area of financing, especially to provide housing for their African populations. With the National Party pressing the case for white workers and white farmers—two crucial conservative constituencies who were demanding that African access to the urban areas should be choked off by firm administrative means—the department came under pressure to review its relationship with local authorities in the 1930s. By 1937, it hesitantly reached the conclusion that effective control over Africans in the urban areas would be achieved only if local authorities behaved more responsibly towards their Native populations and took advantage of the powers which the Urban Areas Act places at their disposal. For the first time, talk about compelling local authorities to implement various components of the Urban Areas Act surfaced with increasing regularity, and such measures eventually featured prominently in the reforms advocated by the Fagan Commission.

But the Urban Areas Act was itself implicated in the department's more critical perception of local authorities. The act harbored a host of provisions that complicated the problem of controlling Natives in urban areas. A major problem was the voluntary nature of Section 12. This provision required employers to register every contract entered into with an employee, to pay a registration fee, and to report the termination of the contract to the municipal pass office, which in turn forwarded the information to the local police. These provisions were designed to yield an accurate picture of employment levels and to keep track of individual workers in the urban areas. The ability to distinguish legal from illegal Africans therefore derived from the ability to register service contracts under the terms of Section 12. However, Section 12 came into effect only after a local authority had exercised its discretion to seek proclamation of the terms of the act. The voluntary nature of Section 12 thus became a major problem for the department in the 1940s. Most local authorities did not request to be proclaimed under the terms of Section 12. Local authorities agreed that Section 12 was indispensable to the proper control of Natives in the Urban areas, but protested that the provision saddled them with expensive administrative costs—buildings had to be purchased, constructed, or rented; administrative and clerical personnel had to be recruited; and the logic of influx controls meant that the size of the municipal police force had to be increased to ferret out the idle and dissolute. A further problem dissuaded a number of local authorities in the Witwatersrand from regulating the registration of contracts. Here, local authorities complained that the Urban Areas Act held them responsible for the administrative costs of registering service contracts while diverting the registration fees to the Provincial Authority.

Moreover, virtually all local authorities refrained from exercising their rights to limit Africans' entry into the urban areas. In 1937, a mere eleven had done so—again a consequence of the fact that influx control provisions were not mandatory, even after they were pointedly included in a 1930 amendment to remedy a major weakness in the parent act of 1923. The original act did, however, authorize local authorities to expel Africans deemed to be idle and dissolute. Suspects detained under this section were brought before a magistrate, Native Commissioner, or Native Sub-Commissioner, who could then mete out one of two penalties: the African could be expelled to his rural residence, or he could be detained on a farm or labor colony for a period of up to two years. But the legal and financial burdens associated with proving that the accused was indeed idle and dissolute dissuaded local authorities from vigorous action.

Regulations governing the issuance of workseekers' permits were another problem. Incoming rural workseekers were required to apply for a workseeker's permit allowing them to remain in an urban area for fourteen days; theoretically, this meant that migrant workers were subjected to greater controls in the labor market than were members of the settled population. At the same time, the permit could not be denied to any African who wished to enter an urban area. Thus, this provision prodded law enforcement authorities to police the validity of workseekers' permits constantly, for this was the only way to detect the presence of illegals in the urban areas. But local authorities did not possess the administrative resources to undertake such a high volume of continual surveillance, so workseekers had virtually unfettered access to the Union's urban areas.... A further weakness of the act was that it concentrated on African males, leaving their families free to enter and remain in an urban area, whether or not the male secured employment. Local authorities in the Cape Province were faced with an added problem: because the pass laws were held by the courts to be ultra vires there, the only way the department could dissuade Africans from departing the Cape's reserves for the urban areas was, as D. L. Smit put it, to spread propaganda amongst them.

Apart from these administrative problems, two further issues loomed large in the disarray of administration. One was the contentious nature of the Stallardist doctrine. Not all local authorities accepted the Stallardist contention that the urban areas were essentially the white man's creation; they therefore exercised their rights not to seek proclamation of the terms of Section 12, principally for ideological reasons. J. Beet (Kimberley City Council)—who, like all the officials present at a 1937 conference between municipalities and the department, supported the view that greater state control over Africans in the urban areas was required—reacted viscerally to Smuts' insistence that all redundant labour should be expelled to the reserves: ...[W]hat about the children? what about old people? Must they be cast out to die like a lot of dogs as has been suggested?

Disaffection with Stallardism, however, was by no means confined to the ranks of local authorities; the department itself was divided over the matter. Heaton Nicholls used his influential dual position as a member of parliament in Smuts' SAP and chairman over the Native Affairs Commission (NAC) to champion Stallardism and tribalist policies at every opportunity and appears to have been the only senior Native affairs man willing to threaten local authorities for not complying with their statutory duties. On the other hand, senior Native administration men such as Denys Reitz (Minister of Native Affairs, 1939-1943), Piet van der Byl (MNA, 1943-1948), D. L. Smit (Secretary of Native Affairs, 1934-1944), and W. G. Mears (SNA, 1944-1948) all expressed strong moral reservations about the pass laws they were duty-bound to uphold, and none of them appears ever to have come down in favor of the Stallardist position.

Three senior department officials actually repudiated this doctrine in 1935. Appointed by Smuts to reexamine the recommendations of the Stallard Commission's report, C. Mould Young (a member of the NAC, and formerly Chief Native Commissioner [CNC] in Natal), A. L. Barrett (CNC, Cape Province), and Fred Rodseth (then Inspector of Urban Locations) recorded their opinion that we see no ground for the claim that the towns are an exclusive achievement and possession of the White man. They therefore opposed policies that would secure the wholesale eviction...of numbers of families who have been resident in the urban areas for generations. Contrary to the claims made by farmers and echoed by Smuts, unemployment levels among blacks were indeed low throughout the 1940s. In 1942 a departmental officer rejected claims about labor surpluses with the observation that There is no influx of Natives along the Reef—and even recommended that pass laws should be more leniently applied. Other Native Commissioners reported that labor shortages existed in their districts and saw no reason to apply urban controls more forcefully. An NC in Johannesburg provided another reason for the lax application of the law: Magistrates were reluctant to commit people to an institution because it was conducted like a prison. It was intended to be a labor colony but was actually conducted on prison lines. The NC had the power to commit an idle and dissolute person for a period of up to two years. If one visited the colony and saw how these people were disciplined, one felt that for merely being idle and dissolute it was not fair. This was a sympathetic administrative view that stands in graphic contrast to what was to come later.

However, Native affairs personnel were inconsistent in their liberal critique of the Stallardist doctrine, and the vacillations of recognized liberals such as D. L. Smit and General Smuts—not to mention the tribalist proselytizing of Heaton Nicholls—did much to perforate the liberal aura the department sought to maintain in the segregation era. That Smuts would ride shotgun for General Hertzog in the 1930s, advocating influx control provisions that Smuts admitted were specifically geared to meet farmers' demands for greater urban controls on Africans, comes as something of a surprise. But even more surprising is D. L. Smit's own support of these measures, particularly in the teeth of opposition from a number of local authorities—all considerably less liberal than Smit—who were adamant that stricter influx controls were not needed. (In 1942, however, Smit decisively recovered his liberal keel by authoring a report recommending that the pass laws should be abolished.) Also, the trenchant rejection of the Stallardist vision that J. Young and A. L. Barrett articulated in their 1935 report was compromised by the practical recommendations they proffered for increasing the ability of local authorities to compile information concerning redundant Natives. This information easily lent itself to justifications for limiting or evicting surplus labor from the urban areas, and would later be approvingly cited by Verwoerd on precisely these grounds.

The second reason for the disarray in urban administration resided in the virtual absence of any empirical information about most issues of administration. Due to the chaotic housing situation, unregulated urbanization, and the burgeoning population of subtenants, estimates of urban populations tended to be unreliable, while the voluntary nature of Section 12 undermined the ability of urban officials to determine accurate local unemployment levels. Also, the inability of local authorities to congregate incoming workseekers in reception depots (as regulations required them to do) reflected the more or less complete lack of any centralized coordinating apparatus administered by either the local authority or the department. Whereas reception depots did exist in some areas, in the Transvaal they were either highly inadequate or virtually non-existent, according to a 1944 departmental memo whose very tone of uncertainty suggests the department's hazy grasp of its own administrative institutions. General Smuts justified his strong support for more effective influx controls by citing numbers almost certainly plucked from the air. His claim at the 1937 conference that the Witwatersrand area alone was being menaced by the evil of ninety-three thousand unemployed Africans was therefore met with open disbelief by representatives from local authorities across the Union. By the same token, the vague guesses that municipal delegates offered to prove that unemployment levels were low in their areas also underscored the absence of reliable data about the labor market.

Given these weaknesses, it is not surprising that a departmental review of urban policy in 1934 judged the Urban Areas Act to be practically toothless and highly uneven in practice and held local authorities responsible for the many weaknesses in urban administration. But, hinting at the principal motivation behind calls for a more muscular urban machine, the memo also lamented the consequent complaints which the Department receives from the Union's farmers about Native farm workers who disappear in the urban areas. As D. L. Smit, General Smuts, and Heaton Nicholls made clear, political considerations—specifically, the obligation to honor the compromises reached over Hertzog's Native Bills—and not the bogey of African urban unemployment lay behind the authoritarian modification of the Urban Areas Act in 1937. Indeed, the new clauses governing influx and expulsion controls were so stringent that municipal administrators appeared genuinely puzzled and uneasy when departmental officials outlined the measure to them at the conference held in 1937.

Apart from significantly escalating the powers of local authorities to exercise these powers, the Amendment Act of 1937 is noteworthy for two other reasons. First, it placed an unprecedented emphasis on the importance of empirical data about Africans in the urban areas. The Amendment Act sought to overcome the pervasive statistical ignorance in urban administration: if the Government has to step in and restrict the influx of Natives into towns, it must have facts to go by. . . . The act therefore provided for the taking of a municipal census and made it obligatory for local authorities to enumerate their African population once every two years and to submit census returns to the DNA. Another statistical survey provided for by the act was intended to yield information that could be used to deter workseekers from leaving the rural areas; this information would be made available to the labor exchanges established by the act. By first checking with the labor exchanges, the Native who wishes to seek employment in towns...will be able to ascertain from the officer in charge of the bureau whether labour is required in particular centers and probably obviate the influx which is taking place to-day.

The other innovation introduced by the Amendment Act was that it pointedly reemphasized the department's overall control over urban policy and its statutory dominance over local authorities. Pivotal sections of the Urban Areas Act—those dealing with the registration of service contracts, influx control, and residential segregation—were altered to vest discretionary power in the Minister of Native Affairs, authorizing him to make their application compulsory if particular municipalities proved recalcitrant. Section 18 increased the minister's powers generally, empowering him to compel any local authority to enforce any section of the act. With the approval of the Native Affairs Commission, any section of the Urban Areas Act could be implemented by the DNA over the objections of a local authority and at the latter's cost. Much to the irritation of municipal officials, these powers were activated immediately. Brushing aside heated objections from municipal officials, Smit and Nicholls insisted that half the financial costs of taking the biannual census would be borne by local authorities whose African population numbered more than five thousand, and that the full amount would be paid for by those with smaller populations.

Much as the 1927 Native Administration Act had done, the Amendment Act provided the department with the opportunity to expand and stake out its authority within the state. Whereas the Native Administration Act had strengthened the department's profile in relation to its perennial competitor, the Department of Justice, the Amendment Act performed a similar service in bolstering the department's authority over local authorities. The Amendment Act brought to urban administration a new discourse centering around technical information, centralization, and bureaucratic hierarchy. Formerly, important aspects of administration such as the need for influx controls had been portrayed variously as social, political, or economic necessities. These perspectives persisted throughout the segregation era, but in the wake of the Amendment Act, influx controls also surfaced more clearly in the department's discourse as a problem of administration and technical management. The importance of empirical information, managerial control, and the internal cohesion of the overall machine thus detached themselves from the pristine fears that whites harbored about racial mixture, their medical anxieties about socially transmitted diseases, and their amorphous anxiety about unregulated Natives in the urban areas. While these latter considerations—the immediate motivations behind the Urban Areas Act in 1923—continued to produce reliable effects and so were faithfully trotted out at election time, urban administration in the 1930s became distanced from the dynamics of ideological debates, policy accords, and the deliberations of secretive parliamentary joint select committees. Administration was increasingly rendered into a bureaucratic realm ruled by Native administrators, who, in turn, were increasingly governed by the technicist considerations of operations and methods.

Drawn from the repertoire of administrative science and organizational theory, such developments abstract bureaucracy from the social relations in which it is embedded and are, one source notes, crucial expressions of the development of the instrumental rationality of capitalist society. In organizational theory, the broader socioeconomic social context is taken as a given so that bureaucratic rationality appears removed from the unstable variability of immediate interpersonal and class relations, marked by divergent interests and struggles. Because bureaucracies process inputs (information) into outputs (policy/administration), state bureaucrats also appear detached from the private interests of civil society, as servants only of the public interest that the state is seen to protect and embody. A striking example of this depoliticizing effect in South Africa is the debate over responsibility for the costs of the biannual municipal census provided for in the Amendment Act of 1937. The purpose of this census was to ground the exercise of controversial influx and expulsion provisions on a scientific empirical basis. In 1937, however, departmental and municipal officials bickered fiercely over which branch of the state would benefit most from the expulsion of Africans from the urban areas—for, in accordance with the philosophy of Native affairs, the costs of administration would be borne by whoever benefited most. With this fiscal sword hanging above them, even those officials who were uneasy about the escalation of influx control in the Amendment Act abandoned their moral protest to rally around the argument that the taking of a census being a national benefit, the Government is respectfully asked to bear the cost thereof.... As Beet's poignant plea (What about the children?) suggests, unless continually resisted, the effect of administrative discourse generally is to place the actions of administrators at a remove from the accumulation process, leaving it to the banality of routinization to depoliticize the technicist consequences of administration. In the administrative lexicon, administration therefore boils down to problems of control and law and order—or the character of idle or dissolute Africans. In sum, the administrative discourse that consolidated within the ranks of urban state cadres in the 1930s diminished the gravity of the assault on African liberties. It presented the rationalization of urban administration, not as the deeply political process it was, but as an administrative imperative on which the welfare of the Union as a whole depended.

The national conference held in 1937 between DNA and municipal officers to discuss the ramifications of the Amendment Act was a signal development in this process. With the basic parameters of segregation policy finally resolved and the disenfranchisement of Africans accomplished, the department was free to turn its attention to the nuts and bolts of practical administration. The 1937 conference of urban administrators, which was almost certainly the idea of the department's new SNA, D. L. Smit, is thus significant in two senses. First, it signified the supersession of the broad disputes around the Native question by an emerging consensus in the more narrowly defined field of urban administration. Furthermore, it prefigured the ascendance of a stratum of state cadres fully apprised of their distinctive and important function within the state and armed with a specific language of legitimation for dealing with the business of monitoring and controlling African urbanization.

If these two developments marked an important change in the character of urban administration, they were at the same time largely formal and embryonic in the late 1930s and 1940s. A cluster of events insured that they would not congeal into either a definite bureaucratic philosophy or a distinct administrative style in the segregation years. First, liberal Native affairs men themselves undoubtedly pulled back from the full implications of the Amendment Act. Having just rammed the Amendment Act down the throats of municipal officials in 1937, Smuts remained confident that the act would be implemented in a milder and fairer form than most people think, largely because the Native Affairs Department is on the whole wiser than the legislators of this country. Second, the outbreak of World War II almost immediately confounded the authoritarian spirit embodied in the Amendment Act. Far from restricting Africans' access to the urban areas, the department all but abandoned the pass system in practice—and even briefly suspended the pass laws in the Witwatersrand in 1942—in order to meet the wartime demand for African labor. Third, the eruption of urban struggles in the 1940s, principally around the issues of housing, rents, and transports, severely jolted municipal administration, taxing NADs at a time when much of their personnel and fiscal resources were already absorbed by the war effort. Last, despite the passage of the Amendment Act, an array of legal loopholes pockmarked the Urban Areas Act, making it difficult for administrators, for example, to evict and prosecute participants in the squatter movements or to deal effectively with the phenomenon of illegal subtenants. Poorly drafted regulations therefore provided inadvertent legal protection to Africans, which (to the irritation of the Johannesburg City Council) the courts all too frequently upheld.

Thus, in 1945, municipal administrators in the Witwatersrand area were sufficiently frustrated with the vacillations in urban policy to reiterate verbatim a request made to the Urban Areas Branch in 1935 that municipalities should know what the definite policy of the government is upon various matters such as segregation, land tenure, influx control, etc., in which respect the Urban Areas Act is inadequate.

The slippage of responsibility for urban administration would be sharply condemned by the Fagan Commission in 1948: It is, therefore, a disturbing thought that in accordance with policy we have inherited from pre-Union days when the problem had not yet assumed this aspect...the regulation of the contacts is not in the hands of the central government, but in the hands of hundreds of municipalities. Local authorities responded to the lacunae in policy and administration in two ways: they combined into regional associations and explored ways to rationalize and update urban Native policy independently of the department. At least two organizations were formed in the early 1940s (and it is likely that more were established, possibly as early as the mid-1930s). First off the mark were local authorities in the Cape Peninsula. Here, fourteen local authorities banded together in 1941 to form a Coordinating Committee re Natives in the Cape Peninsula. Although this committee aspired to do more, its principal concern was to synchronize efforts to combat the ease with which rural arrivals from the Transkei entered (many of them on the mbombela train) and settled on the outskirts of municipal areas. The major accomplishment of the coordinating committee was to press the South African Railways (SAR) successfully into refusing to issue tickets to Transkeian Africans who could not first provide proof of employment in the Western Cape. Although it remained dubious about the success of this tactic, the SAR later extended the policy to other parts of the country.

Much more significant was the establishment of the Association of Administrators of Non-European Affairs (AANEA) in 1944 by Managers of NADs on the Witwatersrand, South Africa's crucial industrial zone. There, the dense configuration of urban centers generated confusion about which local authority could be held responsible for the costs of implementing policy, especially the daunting expenses associated with the provision of housing and public utilities. Dominated by the Johannesburg NEAD, the AANEA developed into a counterweight to the imperialist tendencies of the central state, articulating the collective interest of local authorities in maintaining their relative autonomy, particularly from the DNA. It also afforded NADs on the Witwatersrand a regular forum to explore and compare solutions to the various administrative problems they encountered, giving strong impetus to the consolidation of urban administration as a technical, nonpolitical field managed by a peer group of bureaucrats. As the licensing of managers by the Minister of Native Affairs from 1937 onward confirmed, these bureaucrats were acknowledged to be experts on the problems of urban Native administration. AANEA meetings in the 1940s focused on the lack of uniform regulations, the lowly status of municipal NADs, and the ineffectiveness of the department's Urban Areas Branch.

But, perhaps inevitably, the AANEA also engaged the broad issue of the Union's Native policy. Between 1945 and 1948 it responded to both African resistance and the lethargy of the Urban Areas Branch by exploring ideas that centered around greater state control over Africans in the urban areas. A letter (dated 12 June 1948) that Venables sent to the apartheid government's first Minister of Native Affairs, E. G. Jansen, summarizes the extent of the AANEA's concern with national policy—as well as its similarities to Verwoerd's urban policies in the 1950s. On behalf of the AANEA, Venables requested the standardization of administration regulations, the proper planning and siting of locations, cheap mass transport between segregated locations and city centers, and the strengthening of influx and expulsion controls. In addition, the AANEA called for

the immediate establishment by the Union Government in Johannesburg of a national labor and information office to coordinate and publish information on the labor requirements and surpluses of rural and urban centers throughout the Union...[and] machinery...to authorize the direction of labor to places where employment is available.

Venables concluded that the rapid reorganization of urban administration along these lines was necessary to avoid further squatter movements and possibly civil disorders on a larger scale. The AANEA's reasoning communicates the tenor of the debate between local administrators. For managers of NADs, the virtual absence of a national urban policy had contributed to the numerous practical problems which had come home to roost on the municipal doorstep. Their response, therefore, was shaped by the immediate exigencies of local administration, where the impact of African resistance was most direct, dramatic, and jarring to the administrative machine.

Thus, although all municipalities supported segregation in some form, their calls for the central state to take a greater role in the final years of segregation had more to do with restoring immediate law and order than with grand solutions to the Native problem. They conceived state control over African urbanization not as a grand strategy of domination but as a bureaucratic problem in need of bureaucratic solutions: improved efficiency, coordinating the gathering and distribution of information, and clarifying the administrative hierarchy. But if a number of them sought ways to ameliorate the general impact of these laws on Africans, as W. J. P. Carr claims, they also ensured that the administrative system functioned smoothly: it was not the fundamental principle of racial segregation but the severity of the laws that disturbed them. In a move that crowned the steady ascendance of urban administration to the apex of state policy in the 1940s, the AANEA was superseded in 1950 by the establishment of the Institute of Administrators of Non-European Affairs (IANEA), a national body of Native administrators drawn from across the country, which would develop close relations with the department in the 1950s.

---

Excerpted from Bureaucracy and Race by Ivan T. Evans. Copyright © 1997 by the Regents of the University of California. Excerpted by permission. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

---

Table of Contents

	Preface	ix
	List of Abbreviations	xi
	List of Ministers of Native Affairs, 1910-60	xiii
	Introduction	1
Part 1	Urban Administration
1.	Ambivalent Intervention: Urban Administration in the Interwar Years	25
2.	Reviving the Department of Native Affairs	56
3.	Corrupting the State: Urban Labor Controls	86
4.	The "Properly Planned Location"	119
Part 2	Administration in the Reserves
5.	Ideology and Administration in the Transkei	163
6.	The Bastardization of Authority: Administration and Civil Society in the Transkei	190
7.	From Native Administration to Bantu Administration	224
8.	The Vulgarization of Authority and Rural Revolt: The Transkei, 1955-60	246
	Conclusion: Native Administration and State Formation	277
	Notes	307
	Selected Bibliography	363
	Index	383