Rethinking Bundy: Land and the black middle class – accumulation beyond the peasantry

Abstract

Based on an assessment of historical data on the black middle class in Mthatha, this article argues that South Africa's black middle class has considerable time depth. It originated in Bundy's ‘peasantry’, when African farmers started producing for the market and used their surpluses to educate their children. After being educated, these children continued to accumulate land for farming. Income from the land supplemented their salaries, which allowed them to further the education of their own children and accumulate additional land and, thus, wealth. Hence the black middle class in South Africa is arguably not a post-1994 phenomenon, but is rather the result of intra-generational transmission dating back to the mid-nineteenth century.

Keywords:

• education
• farming
• land
• market
• middle class
• occupations
• peasants
• reproduction

1. Introduction

As South Africa marks 20 years of liberation from apartheid, this historic process is not only cause for celebration but also, importantly, for reflection on the changes that have occurred since 1994. One topic that has drawn the attention of scholars is the question of the black middle class. It is often argued that this class only emerged (but then grew rapidly) in the post-apartheid period (Rivero et al., 2003; Udjo, 2008). Before 1994, colonialism and apartheid had imposed numerous constraints on Africans, including violent dispossession and a battery of laws that severely curtailed their ability to accumulate wealth and form a stable middle class. The liberal view holds that the development of a middle class is critical to social stability, as it mediates between rich and poor (see, for example, Lipset, 1968; Huntington, 1992). Against this backdrop, reflection on the history of South Africa's black middle class is particularly appropriate in the 20th year of democracy.

A central question is how long South Africa's black middle class has been in existence. Is it a recent, post-apartheid development or do class continuities date back much further? In considering this question, this article moves beyond the common conceptualisation of the black middle class as based on income and occupation (Crankshaw, 1997; Seekings & Nattrass, 2006), and instead examines a combination of occupation and property (land). For the black middle class in South Africa, wealth has historically meant land ownership; this will therefore be the focus of the discussion. The notion of wealth is essential for the conceptualisation of class in general. In societies marked by racial inequality, the structuring effects of accumulated wealth are often overlooked in favour of an almost exclusive emphasis on the removal of racial barriers to occupational mobility (on the racial distribution of wealth in the United States, see Oliver & Shapiro, 2006).

Drawing on a historical case study of class formation in Mthatha in the Eastern Cape, this article argues that South Africa's black middle class is a continuation of the successful black ‘peasantry’ described by Bundy (1988). Following Newman (1993:156), the article demonstrates the importance of historical continuities in considering the class question, involving ‘family trajectory’ as well as a ‘shared identity’ encompassing the whole family. The article assesses three generations of the Mthatha black middle class that emerged at the turn of the twentieth century, and shows intergenerational continuities in accumulation and cultural transmission that persist to the present day.

The paper is organised as follows: first, the historical context of middle-class formation in South Africa and its links to the nineteenth-century ‘peasantry’ are explored. The next section traces the development of the first generation of Mthatha's black middle class at the turn of the last century, focusing on the accumulation of land. A discussion of the development of the second generation of the middle class during the apartheid era follows, emphasising the formation of an ideological community and the central role of the women in class reproduction. The penultimate section reviews the development of the third generation during the former Transkei's ‘independence’ (1976–94) and until 2010. It assesses real-estate entrepreneurship in the context of agricultural decline and shows the continued relevance of land for this class. The conclusion links the historical discussion to contemporary debates. The argument presented in this paper is based on oral history interviews of about 30 descendants of Mthatha's black middle class, which were conducted in 2010, as well as on archival data (for a detailed discussion, see Mabandla, 2013).

2. Of peasants and middle classes – Bundy revisited

The period following the extension of colonial rule in the Cape is noted not only for rampant dispossession but also for growing differentiation within African society. Bundy (1988) highlights the influential role of the mission stations in instilling capitalist social and economic norms among their black converts. Missionary engagement on the Cape's Eastern Frontier differed markedly from the belligerent approach towards independent African polities beyond the colony's eastern border. It was influenced by the emerging middle-class culture in metropolitan Britain, which emphasised a common humanity and held that freedom from repressive laws was the basis for spiritual and material progress. On the Cape's Eastern Frontier, British humanitarianism advocated that extending Christianity and British civilisation to local populations was far more beneficial to the colonial order than military conquest and suppression (Lester, 2001). This set the stage for the development of the nineteenth-century black ‘peasantry’ (Bundy, 1988), and towards the end of that century a black middle class.

A careful reading of Bundy's work shows the complex intersection of cultural and economic institutions, simultaneously preventing and enabling access to land by black people. Colonial conquest disrupted the stability of the precolonial mode of production through the dispossession of African land. This was followed by colonial land reforms, which saw large tracts of productive land privatised and concentrated in European hands. On the other hand, the mission stations facilitated limited land accumulation by Christian converts throughout the 1830s. By the 1840s, the Cape colonial government had issued title deeds to about 70 000 acres of land to mission stations (Bundy, 1988). Colonial land reforms had both economic and political implications for African social formations. First, the dispossessions disrupted traditional landholding patterns and production structures; second, the subsequent land redistributions disrupted the chiefs' power to allocate land. Land-poor adherents could now turn to missionaries (and conversion to Christianity) as an alternate option for accessing land, thus bypassing traditional structures.

The combination of cultural and economic factors meant that social mobility was a question not only of access to land but also increasingly of education. Mission stations contributed to education-based stratification by training converts in new agricultural techniques, literacy and numeracy in order to facilitate economic transactions. This improved the productive capacity of African farmers and contributed to the formation of a class of agricultural producers who could participate fully in the emerging capitalist agrarian economy. Increasing demand for commodity production, the imposition of colonial taxes and the general penetration of the rural economy by consumer goods compelled them to expand beyond subsistence farming to produce a surplus for the market. By adopting a mixed model of subsistence and commercial agriculture, they laid the foundation for the development of a small black middle class in the Eastern Cape. They reinvested their profits in equipment, livestock and crops, as well as in the education of their children, who became clerks, teachers, priests and even interpreters. This education represented a qualitative shift from the functional training their parents had received. The children had marketable skills – in the sense of Weber (1968) – that could be exchanged on the emerging capitalist labour markets, and they reinvested their salaries from these occupations in agriculture.

In addition to improving the productive capacity of local farmers, the missionaries established stores that sold clothing, household items and agricultural implements. Mission stations thus contributed to both the transformation of the mode of production and the establishment of new forms of consumption. They encouraged new ways of living, including the move towards square houses (instead of the traditional round houses or rondavels). The new type of house was symbolic of new patterns of consumption and tied African societies firmly to the British colonial economy. As elaborated in the missionary journal, the Kaffir Express: ‘With a proper house, then comes a table, then chairs, a clean cloth, paper or whitewash for the walls, wife and daughters dressed in calico prints, and so forth’ (cited in Bundy, 1988:37). The logic of consumerism embedded in the design of living space was again evident a century later, with the introduction of ‘matchbox’ houses in African townships. Upon receiving a house, a first-generation urban resident is reported to have remarked: ‘This kind of house whispers to you that it needs more furniture’ (Bonner, 1995 cited in Seekings & Nattrass, 2006:103).

In summary, mission stations played a major role in colonial social transformation. Access to land, education, a combination of subsistence and commercial agriculture, and lifestyle and consumption changes contributed to social stratification within African society and led to the rise of new, post-traditional classes, the ‘peasantry’ (Bundy, 1988) and the black middle class.

By and large, African society adapted to these changes. The training they obtained gave black agricultural producers a competitive edge and they were soon able to increase their output significantly. However, the transition into agrarian capitalism also exacerbated inequalities between rural households. Higher demand for food production meant that those who could afford to purchase ploughs, wagons and other equipment took full advantage of market opportunities and competed successfully with white farmers.

However, this progress was soon crushed through the adoption of the Natives Land Act of 1913. Bowing to pressure from the commercial farming and mining industries to address labour shortages, the state introduced a singularly controversial piece of legislation. Under this Act, Africans were forbidden from accessing and owning land outside the reserves. Black sharecroppers and labour tenants, who had worked the land as independent producers in areas now reserved for white occupation, were most affected. They were forcefully removed to the reserves created by the Glen Grey Act of 1894, which amounted to only around 7% of the country's land area.

The 1913 Act caused massive dispossession of black farmers. According to Bundy (1988), it led to the ‘fall’ of the ‘peasantry’, as overwhelming numbers of farmers were forced into wage employment for survival (also Mafeje, 1988; Plaatje, 1995). However, the grand narrative of proletarianisation cannot adequately explain the persistence of landownership within certain sections of black society. The literature is relatively quiet on the fate of the descendants of the early commercial farmers. Did all of them enter the ranks of the proletariat or other forms of wage labour? Did access to land disappear entirely? As the following case study of Mthatha's black middle class demonstrates, land accumulation for these descendants continued, in an urban context, beyond the ‘fall’.

3. Elisha Mda and the first generation of Mthatha's black middle class

The story of Elisha Mda, one of the early black landowners in Mthatha, is important for understanding the colonial history of the black middle class and the interplay between landownership and the opportunities afforded by mission education. Described in Mabandla (2013), this story sheds light on the lives of African landowners following the restrictions on land accumulation in the Cape and demonstrates the continuity of the middle class beyond Bundy's ‘fall of the peasantry’. In this article, Mda's life history illustrates the linkages between the ‘peasantry’ and the first generation of the black middle class.

Elisha Mda's date of birth is unknown but is generally associated with the great drought of 1860. It could well have been earlier, as he had qualified as a teacher by the late 1870s. His family had settled in Dutywa after their expulsion from the mouth of the Buffalo River following the Nongqawuse cattle-killing disaster (1856–58). Mda grew up in a traditional context; his parents were peasant farmers similar to those described by Bundy. They invested in his education and sent him to live with his aunt, who stayed near a missionary school. His grandson, Mda Mda, explains:

We were red people from Ndlambe's territory. Elisha went to stay with his aunt in Tsomo with the amaZizi clan. A missionary school had been established there by the whites. That land was under white rule. So, he attended school. And because the boy was clever, he was well liked by the teachers and missionaries. (M Mda, interview, Mthatha, 9 July 2010)

After completing his education at Lovedale, Mda became one of the first teachers at a newly opened school in Ntshatshongo (Fort Malan) in the 1870s. He also managed to acquire land, by saving part of his salary and investing his savings in wagon transport. He then combined teaching with transport riding until he had accumulated enough money to buy land. Bundy (1988:77) suggests that this was common practice. Many ‘peasants' engaged in transport riding once their crops were harvested, taking their produce to the markets in Queenstown and King William's Town and carrying goods for local traders on the return journey to Fingoland.² Transport riding, according to Bundy, was a means of earning money to buy or hire land, and many did this on a full-time basis. Mda used the proceeds from transport riding to buy land in British Kaffraria.³ However, like many others, he lost his land after being suspected of aiding and abetting enemies of the British Crown in the last frontier war, the War of Ngcayichibi (1877–78). It is not clear whether he or any of the other landowners received compensation, but he reinvented himself as a landowner in Mthatha soon after, at the turn of the twentieth century.

The first generation of Mthatha's middle class illustrates the historical complexity of black landownership and agricultural production. The early adoption of education and the combination of farming and other occupations render Bundy's term ‘peasant’ inappropriate for people like Elisha Mda. It also shows that the ‘many Africans' engaged in transport riding (Bundy, 1988:77) were in fact a highly differentiated group. The few who could afford land were drawn mostly from the limited ranks of professionals (Peires, 1989).

The emergence of a substantial and well-defined first generation of the black middle class in Mthatha is closely linked to the Umtata Water Scheme of 1906, when municipal land had to be auctioned to finance the building of the then Kambi (now Mthatha) Dam. After many attempts to restrict land sales to whites, black farmers were eventually allowed to take part in the land auction of Plot E, later known as Ncambedlana, in 1908. This laid the foundation for land accumulation by Mthatha's black middle class, a process discussed at length in Mabandla (2013).

The town of Mthatha was an important commercial and administrative centre. Its well-developed railway network linked the Transkei hinterland with the industrial centres of the Republic. The town attracted both labourers en route to cities (such as the industrial metropolis of Johannesburg) and the aspiring middle class who held professional positions in the colonial civil service. Many members of the first-generation middle class came from the ranks of the latter group. Apart from Elisha Mda, noted members of this generation included AC Zibi, Luke Yako, David Noah and Tennyson M Makiwane (Redding, 1992). Most were descendants of Bundy's ‘peasantry’, who had obtained an education and established themselves in white-collar occupations in town, using their salaries to accumulate land. Bundy's ‘peasantry’ could thus be described as ‘Generation Zero’, who used profits from agriculture to educate their children. They, in turn, became ‘Generation One’ – Elisha Mda and others like him who had professional jobs and accumulated land in places such as Mthatha. This continuity with the activities of their ‘peasant’ parents is not captured by the ‘fall’ thesis.

The contrasting backgrounds of TM Makiwane and Elisha Mda illustrate the complex history of this class, in which some were closer to their traditional/peasant roots, while others had a longer family history of education. TM Makiwane was the son of Elijah Makiwane, a mid-nineteenth-century leader of the church and the African education movement. Hence, unlike Mda, Makiwane did not come directly from the ‘peasantry’; he was essentially second generation, since his father had been educated. A graduate of Lovedale College, he was a teacher until 1910, when he bought land in Mthatha and left teaching to take up a senior clerical position at the Transkeian Territories General Council – the Bhunga (caucus, isiXhosa).⁴ Makiwane was active in local politics and was an elected member of many civic organisations (Redding, 1992; for more details on his life, see Mabandla, 2013). His opposition to white domination at the Bhunga (M Mda, interview, Mthatha, 9 July 2010) and his civic spirit are celebrated in the poem Izibongo zika Gambu (The Praises of Gambu, isiXhosa) by St J Page Yako.⁵ Makiwane was also the first editor of the bimonthly English–isiXhosa agricultural journal, Umcebesi Womlimi Nomfuyi/Agricultural and Pastoral Guide (founded in 1925 and published by the Agriculture Department of the United Transkeian Territories General Council; see Switzer & Switzer, 1979) as well as of the local weekly Umthunywa (The Messenger, isiXhosa). While maintaining his clerical position, he was also a highly successful farmer (M Makiwane, personal correspondence, 1 August 2010). His salary and agricultural profits enabled him to educate not only his children but also his wife Virginia, who established a career as a social worker. In 1943 he bought six more plots at a cost of £800, at a time when an experienced African clerk earned only around £160 per year (Redding, 1993).

The success of the first generation of Mthatha's black middle class lay in their unique combination of education, employment and land ownership. A mixed model of subsistence and commercial farming, combined with salaries from professional jobs, formed the basic mode of reproduction for this class. They continued to farm as their parents did, raising both crops and livestock. In addition to their own land, they had access to the municipal commonage for grazing their livestock. Such access was crucial to their success, and the decline of farming in later generations is partly explained by housing development on the Mthatha commonage.

Mthatha's first-generation black middle class shows that there was little discontinuity in their lives from being successful ‘peasants' in the rural context to combining professional jobs with landownership and cultivation. Thus, in contrast to Bundy's ‘fall’, there was continuation and even an ongoing ‘rise’.

4. The second generation, 1950s to 1976

Between 1950 and 1976 the ruling National Party severely tightened segregation laws and implemented various apartheid policies. Among these was the Group Areas Act of 1950, which demarcated separate residential and business areas for blacks and whites. In the traumatic forced removals that followed, large numbers of African people were herded into locations far from their places of work and business. Yet while people were forcibly being removed from areas such as Sophiatown and District Six, the second generation of the Mthatha black middle class continued to acquire land and urban property.

Many scholars hold that the development of the black middle class in this period followed an occupational trajectory, resulting from black urbanisation in the 1940s–60 s (Kuper, 1965; Crankshaw, 1997; Seekings & Nattrass, 2006). Even in the Bantustan context, the one place where blacks could own land, research has largely focused on the role of the Bantustan state in enabling class formation (Southall, 1982; also Josana, 1989), while the role of land has largely been overlooked. One of the few studies addressing land (Redding, 1993) also argues that commercial farming in Mthatha, which had protected the middle class from sliding into migrant labour, had declined by 1950. However, the part of the middle class that combined professional employment and landownership continued to exist after the 1950s.

The ongoing accumulation of urban land by this middle class was unusual. Towns like Mthatha, with significant European commercial interests, were not part of the reserves. They were regarded as ‘white spots' and were thus subject to the same segregation legislation that had transformed urban communities elsewhere. Yet Mthatha's black middle class was adept at using local politics to protect its land rights. Tensions around black landownership in Mthatha had surfaced as the governing National Party sought to enforce segregation. Local Afrikaners, emboldened by the party's rise, had petitioned the government about what they saw as the deplorable ‘mixing of the races' in Ncambedlana and requested it to act against black ownership in line with the Group Areas Act (Redding, 1992). Mthatha's black middle class responded with their own campaign, lobbying the government to declare Ncambedlana a black area, ‘thrown open’ for African ownership. This was, however, declined by the Minister of Native Affairs, HF Verwoerd (who became Prime Minister in 1958). Verwoerd outlined government policy as follows:

Umtata is at present the centre of all governmental activities in the Transkeian Territories and is also probably the largest and most important commercial centre. As a result it has a large European population, and many of the Europeans have large vested interests involving a considerable amount of capital. For the moment this centre accordingly is regarded as European in character and in general the policy of the government in regard to it must be identical with that pertaining elsewhere throughout the Union. (MAR, 1951)

The campaign by Mthatha's black middle class was a calculated move: they had reasoned that excising Ncambedlana from the municipality to allow for black occupation would have been a highly complicated and controversial undertaking and that the government would have preferred to maintain the status quo. A number of white people owned properties in the area and would have had to be compensated. Most whites in the Transkei reserve, including the Umtata Municipal Council, were aligned to the opposition United Party, which had been defeated by the National Party in the 1948 elections (Southall, 1982:149), and they would probably have resisted such a move. The gamble paid off: black landowners were able to retain their properties, as Verwoerd confirmed the distinctive challenge this presented:

I realise that the position at Ncambedlana is unique in that the area borders on the Native Reserves and that the general authority permitting the acquisition of freehold title to the lots in that area was granted with the full consent of the local authority. I have, therefore, agreed that the position there should remain unchanged, although I would prefer to see Ncambedlana excised from the municipal area. (MAR, 1951)

This decision led to two simultaneous processes: National Party adherents exerted pressure on black residential and trading rights in the central business district in order to monopolise these for whites, while the black middle class bought out almost all of the white-owned properties in Ncambedlana. As a result, the second generation of the black middle class was established before the full implementation of the Bantustan strategy, and was firmly entrenched by the time Transkei was accorded self-governing status in 1963.

While the first and second generations had roughly the same professional backgrounds, the latter attended university (Fort Hare) rather than college. This educational mobility gave them access to a more diversified range of occupations (e.g. teachers, lawyers, doctors and entrepreneurs), better-paid jobs and a more secure income base. It also conferred a status that transcended the local setting, since they attended university with many of the major nationalist leaders. According to Aunt Laura Mpahlwa, who settled in Ncambedlana in the 1960s, the black middle class of the time was a highly status-conscious social group, who would occasionally flaunt their university education to distinguish themselves from others:

The majority of them were teachers, eh, the majority of them had degrees, studied at Fort Hare and Lovedale, and we used to boast about those things. That if you didn't study in Lovedale or Fort Hare then, you were really a nobody. You were a nobody. But in their simple way, they were people who were accommodative; they did not really look down upon those who really were not able to reach the highest educational standards. (L Mpahlwa, interview, Mthatha, 20 July 2010)

Marriage patterns also showed significant consolidation: members of this social group had generally attended the same missionary schools as their spouses and had married within their class. Social and cultural capital in the sense of Bourdieu (1984) circulated within and around this class. Recruitment to jobs, especially at prestigious black schools such as St John's College, utilised the social networks established at Fort Hare. Information about land sales was also circulated in these networks. Ultimately, unlike the first generation, which had lived elsewhere and used the area mainly for farming, a settled middle-class community with strong social bonds developed in Ncambedlana. They built family homes, schools, churches and medical facilities. In the cultural realm, aspirational middle-class behaviour included the organisation of choral eisteddfods,⁶ with competitors coming from as far afield as Kimberley.

The sense of community among the second generation was based on a commitment to middle-class values and a strong communal identity. Nomonde Bam, whose father-in-law (Ngubethole Bam) was a member of the second generation, describes these close bonds of solidarity and support in times of joy and sorrow: ‘whether it's a wedding or funeral or whatever’, people were there for each other. All were involved in planning and logistical support, whether through ‘prayers' or more materially through food, for example, ‘because there would be visitors throughout the week … one didn't wait to be asked for support’ (N Bam, interview, Mthatha, 19 July 2010).

As with the first generation, the second generation continued to combine professional employment and agriculture. However, gender discrimination meant that married women in this generation found their career prospects considerably diminished. In the teaching profession especially, only single women and men were considered for permanent positions (Kotecha, 1994; also Van den Heever, 1975). As a result, married women soon found themselves forced into the domestic sphere. These women, many of whom had equivalent qualifications to their husbands, were not only central to domestic production but were also responsible for the management and control of agricultural production, the market sale of crops, and the contracting and remuneration of labour. This included the organisation and management of hired labour and the amalima (cooperative labour groups, isiXhosa; McAllister, 2004), the two main forms of agricultural labour used.

Helen Bradford's (2000) critique of Bundy's work on the ‘peasantry’ focuses on the silencing of the women's role. According to her, Bundy's ‘peasants' had displaced women from their traditional domain – agriculture – once an agricultural market had developed. This line of argument, however, obscures the underlying social relations of power and control. Lewis (1984) highlighted the gendered nature of the precolonial mode of production, which was divided into agriculture as the women's sphere and cattle rearing as a male domain. This provided cultural justification for women's dependence on men, since cattle outranked cultivation in the food cycle. While the central role of these black middle-class women in agriculture might suggest a gendered recovery of this sphere, livestock remained a male preserve among the second generation of Mthatha's middle class.

In addition to these roles, the women also played a key role in developing and reproducing the attributes of their middle-class status. They opened informal crèches, where local children were schooled before enrolling in primary schools, and instilled and supervised standards of behaviour and propriety (also Wilson & Mafeje, 1963). While the men went to work, they remained in the community, where they could discourage neighbourhood children from anti-social behaviour or give horticultural ‘advice’ to those whose houses appeared unkempt. They formed or joined typical middle-class associations, such as the Young Women's Christian Association and the Zenzele (do it yourself, isiXhosa). Their everyday practices and activities were thus essential to the middle-class ‘habitus' (Bourdieu, 1984). Middle-classness, in other words, is not just a question of ownership, occupation or income, but also of values and policed behaviour. The role of the women challenges the dichotomy that is often assumed by dividing social reproduction into a dynamic-productive-male realm and a passive-domestic-female one. While being central to agricultural production, these women were equally central to the reproduction of social relations. The socialisation of children along middle-class values occurred inter-generationally, and these social relations changed and were reproduced intra-generationally as the children grew up. This is clearly demonstrated by the influence of the women on the educational success of their children. As explained by Aunt Laura Mpahlwa:

Abantwana [the children, isiXhosa] got a very good background of English, what then used to be called Royal Readers. The people who were teachers were Royal Readers so, the children of Ncambedlana knew English like anything and they spoke good English. They wrote good English and were getting very good marks. (L Mpahlwa, interview, Mthatha, 20 July 2010)

Just as with the first generation, the next generation continued with the mixed model of agriculture and reinvested their profits in their children's education. As one of their descendants, Loyiso Mpumlwana, puts it:

 … there was a lot of this subsistence farming and commercial farming. People would consume and also sell … people managed to put their kids to school. We got our education on the basis of commercial farming. We had a donkey cart at home and we sold. We would load it and sell here in town. (L Mpumlwana, interview, Mthatha, 26 July 2010; emphasis added)

One reason for the farming continuities within this class was that black incomes from professional occupations, especially in the civil service, remained notoriously low. As in the first generation, the combination of subsistence and commercial agriculture allowed black middle-class families to thrive despite poor salaries. Thus, the ‘rise’ continued: not only did the peasantry not fall, but the middle class that emerged out of the peasantry also continued to flourish after 1913 and even during apartheid.

5. The third generation, 1976–2010

The proclamation of the Transkei as an ‘independent homeland’ broadened the occupational base of the middle class through the opportunities offered by the Bantustan state structure. Such new opportunities opened up in administration and management (and even business) as the Africanisation policies of the Transkei regime sought to replace white officials with black ones. It is no surprise that Mthatha's black middle class – drawing on generations of education – was well placed to exploit the new environment. (The same could be said about black advancement in the democratic era.)

In this process, the third generation of the Mthatha middle class was transformed along occupational lines. While the bulk of the first and second generations could be classified in the lower to middle income or education categories (following Crankshaw, 1997; Seekings & Nattrass, 2006), the next generation was more professionalised. The number of university graduates increased, and there was a growing diversity of occupation, including medicine, law, management and business, for example. This allowed people to improve and broaden their income base even further. As one member of the third generation, Sembie Danana, puts it: ‘very few went into teaching’ (S Danana, interview, Mthatha, 18 July 2010). Thus, over time, the middle class had been transformed from one that combined employment and farming in the first and second generations to one that that was educationally distinguished, working mainly in prestigious professional positions. (Members of the middle class went on to find well-paid employment across the country, and after democratisation many rose to even higher positions, supported by the legislative framework of employment equity and black economic empowerment.)

These new opportunities continued the non-discriminatory practices of the ‘self-government’ and ‘independence’ eras, which had also allowed women such as Aunt Laura Mpahlwa and Mrs MMM Raziya to come to prominence in the business and public spheres. For example, Mrs Raziya served for many years as the only female councillor in the Umtata City Council (Umtata City Council, 1982). In politics too, women carved out larger roles; for example, Stella Sigcau (later the Minister of Public Enterprise in the Mandela administration) was appointed Prime Minister of the Transkei in 1987.

These socio-political changes affected the agricultural activities of the middle class. As they had aged, the second generation had become increasingly unable to continue the demanding work of agriculture. For the third generation, however, the situation was different. The curriculum of the missionary schools, taught to their parents and grandparents, focused on unity of mind, body and spirit. Ploughing the land was regarded as important for the students' sustenance, as well as their health and mental well-being. They were trained in agricultural skills, so that ‘they might be afterwards able to instruct their countrymen in the art of cultivating their own soil’ (Shepherd, 1940:90). In contrast, for the third generation, training in industrial and agricultural skills had been removed from the academic syllabus following the centralisation of education under the Department of Education and Training. Such skills were now taught at special vocational schools (Christie & Collins, 1984). Thus the third generation lacked the enthusiasm and skill of their forebears. In addition, agriculture had gradually been ‘stigmatised’ as the last resort of those who could not obtain a university education. In the Transkei, in particular, the education system was geared mainly towards equipping people for urban white-collar professions (Ntsebeza, 2006).

Apart from the school syllabus, other factors leading to the decline of agriculture include the agricultural policies of the Bantustan regime, which favoured large-scale commercial agriculture over household agricultural production or the family farms of the Mthatha middle class. To encourage rapid urbanisation, housing development was undertaken on the commonage that had served as grazing lands for the first and second generations. The suburbs of Hillcrest, Northcrest and Hillcrest Extension were developed in the early 1970s, 1981 and 1987 respectively (Siyongwana, 1990). This effectively pushed cattle, which were needed for ploughing, out of the urban environment. Ecological factors, such as the 1980s drought, also contributed to the decline of agriculture. A final factor was women's growing professional mobility, which meant their withdrawal from household agricultural production as well as from community building. The opportunities that had opened for men and women of the third generation thus affected both agricultural production and the character of the neighbourhood. Out-migration to larger urban centres also negatively affected community cohesion and lowered the social capital of those who stayed behind (see, for example, Beatty et al., 2009). However, land remained important for the identity and well-being of the middle class.

In the democratic era, some of these factors continued to influence land-use patterns, as did labour shortages resulting from new opportunities for those who historically had provided agricultural labour. In addition, urban population pressures contributed to the conversion of farmland for rental accommodation. As in the rest of urban South Africa, Mthatha's population has increased significantly because people from the surrounding villages moved into town in search of employment opportunities, mainly in white-collar professions and social services (Makgetla, 2010). The population of Mthatha is reported to have increased from 72 000 in 1991 to 91 000 in 2001 (Siyongwana, 2005), while later estimates put the figure at around 150 000 (Harrison, 2010). This, coupled with the inadequate provision of low-cost housing during the ‘independence’ era (Siyongwana, 2005), has increased the demand for housing, which has in turn enhanced the position of landowners. There is differentiation among the landowners in this regard: those with more financial resources undertake large-scale suburban developments, while those with fewer resources invest in smaller projects. Intermediate housing developments involve the building of several rental units, and those with limited resources let individual rooms on their property. Real-estate entrepreneurship has become a way of life for the third generation, just as agriculture had been for the previous generations.

6. Conclusion

This article refutes conventional wisdom by demonstrating that the emergence of the black middle class predates democratisation. The early beginnings of this class were traced to Bundy's ‘peasantry’, which developed in the Cape during the nineteenth century. The article has pointed to important historical continuities in considering the class question. It has shown the relationship between land and education, which has been at the centre of social change within African society, and its continued role in defining family ‘trajectories' and class ‘identities'. This relationship allowed the development of what we call Generation Zero, Bundy's ‘peasantry’, and later that of Generation One, the early black middle class of Elisha Mda and others. His example, in particular, illustrates dispossession and ownership in the lifespan of one individual: land lost, land obtained, and land lost again. His education, the one thing that could not be taken from him, gave him a new chance for new land in Mthatha.

The symbiotic relationship between land and education continued to play a pivotal role in reproducing two more generations of the black middle class, the second and third generations. This article has highlighted the central role of women in wealth creation and cultural transmission, at a time when workplace discrimination (especially in the teaching profession) excluded many married women from the workforce. It has shown the ongoing importance of land to the middle class in a period of agricultural decline, as evidenced by the present-day real-estate entrepreneurship of the third generation.

The historical continuities, albeit nuanced, in accumulation, education, farming and agricultural practices demonstrate the strong links between the nineteenth-century ‘peasantry’ and the middle class that emerged in the Cape in the late nineteenth and early twentieth centuries. While the case of the Mthatha middle class cannot be generalised, this development may not be unique to the liberal conditions of the old Cape Colony, which initially allowed black landownership. Similar developments may well have been replicated elsewhere, especially in the context of the Bantustans, where Africans could own land after 1913.

Brandel-Syrier's (1971) study of the black middle class of Reeftown also shows that the success of this class in the urban environment depends on both education (and, hence, occupation) and the persistence of rural landownership. Methodologically, most current approaches to the black middle class focus on income and quantitative analysis, leading them to locate the emergence of this class firmly within the democratic era. They overlook Bourdieu's (1984) concept of ‘habitus', which mediates the intra-generational reproduction of social relations. Thus, the question of land and wealth transmission has remained obscured by contemporary sociology's overwhelmingly urban focus and its fixation on occupations and incomes.

Notes

²Area between the Kei and Mbashe Rivers.

³Encompassing the present-day districts of East London and King William's Town.

⁴This was a form of local government based on district councils of elected black members under the chairmanship of white magistrates.

⁵ Umcebesi Womlimi Nomfuyi/Agricultural and Pastoral Guide, May 1950.

⁶A Welsh festival of literature, music and performance dating back to the twelfth century.