10—
A Nation Divided? The Swazi in Swaziland and the Transvaal, 1865–1986

Hugh Macmillan

Introduction: The Swazi of South Africa

South Africa's intention of transferring to Swaziland the KaNgwane Bantustan and the Ingwavuma district became public knowledge in June 1982 when the KaNgwane Legislative Assembly was suddenly abolished and when the South African government took over the administration of the Ingwavuma district from the KwaZulu Bantustan. Legal action by the leaders of the KaNgwane and KwaZulu Bantustans, Enos Mabuza and Chief Gatsha Buthelezi, respectively, resulted in defeats for the South African government and a return to the status quo ante pending the results of a commission of enquiry which was established under the chairmanship of Mr Justice Rumpff.^[1]

The mass of publicity following the publication of this proposal drew attention to the existence of a Swazi population in South Africa and to their relationship with the Swazi of Swaziland. The resident population of Swaziland itself was estimated in 1976 at 470,270 in an area of over 6000 square miles. According to the latest census figures, in 1980 there were 786,835 people described as 'Swazi' in South Africa, of whom about one third lived in the KaNgwane Bantustan. This consisted of two blocks of territory in the eastern Transvaal, one contiguous with Swaziland and one separated from it, with a combined area of about 1000 square miles. The de facto population of KaNgwane had increased from about 100,000 in 1970 to about 300,000 in 1982 as a consequence of the South African government's policy of forced relocation of black people from nominally 'white' areas. About 80 per cent of the population of the Bantustan was described as 'Swazi' and the remainder as 'Shangaan' (Tsonga), Zulu or Sotho. Some Swazi also lived under their own chiefs in the Lebowa Bantustan but the majority of those outside KaNgwane remained settled in the towns, on farms, and in so-called 'black spots' within 'white' South Africa. Many lived on the Witwatersrand, far removed from the influence of Swazi language and culture. Allowing for population increases, there must today be about 1,500,000 people described as Swazi, of whom about 55 per cent live in South Africa and about 45 per cent in Swaziland.^[2]

My main purpose will be to examine the changing relationship between the Swazi of Swaziland and those of the Transvaal from the time of the demarcation of a border between them in the late nineteenth century until the present day. After a brief examination of the creation of a single Swazi identity by the ruling

group in the context of the expansion and consolidation of political power in the pre-colonial period, I shall try to trace the changing patterns of ethnic awareness, or at least ethnic assertion, in the two halves of the divided nation, and to relate this to changes in their political and material conditions.

Members of the Swazi ruling group in Swaziland had shown a reluctance to assert claims in the Transvaal following their subordination to colonial rule, but from about 1910 they began to demonstrate a greater awareness of the influence which they might achieve in South Africa as a whole through a more assertive approach. Further shifts in their position occurred in the 1930s, when they began to lay the basis of a new cultural nationalism; in the 1960s, when they were finally able to use ethnic mobilization to re-establish a position of real political power within Swaziland; and in the 1970s, when they began to see the possibility of realizing long fermenting irredentist claims in the eastern Transvaal.

Among the South African Swazi a renewal of ethnic awareness by the chiefs and their councillors can be traced from the early 1920s. While it has been argued that the Native Administration Act of 1927 heralded a new policy of 'retribalization', there is little evidence that the Swazi of the eastern Transvaal were much affected by it. Rather, efforts at retribalization were spontaneously generated within the Swazi community and were largely ignored by the South African state until the 1950s, when there was a shift at the centre from simple segregation towards the encouragement of specific ethnic formulations of political awareness. The attempt in the 1980s to 'reunite' the Swazi, who had been so long and so fundamentally separated, was doomed to failure. Political and economic developments in South Africa and Swaziland since World War II had acted to deepen the divide between the two halves of the Swazi nation. In Swaziland an exclusivist cultural nationalism has triumphed since independence in 1968, while in South Africa those who sought to mobilize the Swazi as a political force were confronted by formidable obstacles, in the shape not only of competing ethnicities but also of a broader based South African nationalism.

The Partition of the Swazi People

The pre-colonial Swazi state reached its greatest extent shortly before the death of King Mswati I in 1865. By this time, or soon thereafter, a distinct Swazi ethnic group, composed from Nguni, Sotho and Tsonga elements, had been created. The establishment of a tributary state under royal Dlamini leadership broke down the self-sufficiency of the homestead through the formation of nation-wide regiments in which young men from all the chieftaincies were brought together. The building of national rituals such as the ncwala, a first fruits ceremony and ritual of kingship, helped in the process of integration, as did dynastic marriages and the widespread establishment of royal villages under wives of the king with loyal indvunas . At the same time the composite nature of the kingdom was perpetuated and emphasized by the continued ranking of clans according to whether they came with, preceded, or arrived after the dominant Dlamini clan in the present Swazi area.^[3]

A distinct Swazi language (siSwati) was first recorded in 1846 but regional dialectal differences persist until the present day, with Sotho influence discernible on the northern borders of the Swazi linguistic area in the Nelspruit and Lydenburg districts and Zulu influence strong in the southern part of Swaziland and the Piet Relief and Ermelo districts of the Transvaal. Standard Swati is based on the dialect spoken in the north central area of Swaziland itself, an area which includes the royal capitals of Lobamba and Lozita.^[4] Many of what are regarded as

the most distinctive features of Swazi culture and political institutions—such as preferred cross-cousin marriage, the conciliar form of government, and the status and political role of the Queen Mother—have been attributed to Sotho influence.

The death of Mswati coincided with the first pressure of white settlers on the Swazi state, which had come to control the area between the Crocodile river in the north and the Pongola river in the south, and between the Lebombo mountains in the east and a boundary on the eastern highveld plateau in the west. The Transvaal Republic made various attempts to annex or establish a protectorate over Swaziland, but these were thwarted by the British who saw the Swazi as a useful barrier between the Boers and the sea. The present boundaries of Swaziland were first roughly demarcated in 1866, but they were more specifically defined by the British during their occupation of the Transvaal in 1880 and were included in the Conventions of Pretoria and London, in 1881 and 1884 respectively, which also guaranteed the independence of the kingdom.^[5] In 1881 the Swazi accepted the southern and much of the western boundary line but objected to the line in the north and northwest which cut off from Swaziland three important royal villages: Mbhuleni in the modern Carolina district; Mjindini, near modern Barberton; and Mekemeke, northeast of Barberton.^[6] The Swazi king, Mbandzeni, continued to dispute the border in this area until shortly before his death in 1889, and until that time the Swazi definition of 'Swaziland' extended beyond Barberton to the Crocodile river in the north and along the upper reaches of the Komati river in the west.^[7]

From the mid-1860s onwards the Swazi on the eastern highveld, most of whom in fact lived in the middleveld valleys, such as the Komati, which intersected it, rather than on the cold highveld itself, came under increasing pressure. White settlers, many of whom were associated with the New Scotland scheme, at first sought Swazi labour and paid for it, but then by the mid-1870s began to demand tribute from people who lived on 'their' land. There was some resistance, but the people were gradually transformed into squatters paying rent in labour, kind, or cash. The population was increased by the labour demands of farmers who brought people in from Swaziland or elsewhere.^[8] The attitude of farmers and officials on the highveld, where most farms by the end of the nineteenth century were occupied by the owners, was distinctly hostile to the preservation of chiefly authority and ethnic loyalties. Speaking of the Ermelo district in 1914, General Tobias Smuts told the Beaumont Commission with evident satisfaction: "There are no tribes there.'^[9] The Native Commissioner for Carolina commented at the same time that the widow of Mswati at Mbhuleni lived on a private farm 'just as another squatter'.^[10]

In the predominantly middle and lowveld district of Barberton to the north of Swaziland the pressure of white settlement began to be experienced with the gold rush of the mid-1880s, but the prevalence of malaria deterred owner occupation. The bulk of the land in this district and in the low-lying areas of the Piet Relief district remained until after World War I in the hands of the Crown or of land companies and thus effectively in 'native occupation'. Although there was some competition from highveld farmers who used these areas for winter grazing and as private labour reserves, many chiefs survived relatively undisturbed. They were able to retain influence in their areas through the performance of religious rituals, the settlement of disputes for small fees, and the retention of relatively large herds of cattle which enabled them to continue to make alliances through marriage, aided by the higher rates of lobolo (bridewealth) obtained for their daughters. In some cases they were able to command some tribute in labour or kind and continued to perform the chiefly duty of hospitality. The agent of the New

Scotland Land Company observed in 1917 of some low-lying farms in the Piet Retief district:

I do not know whether you know that all that ground is parcelled off among the chiefs living in that area. A native will not go into the area of a chief unless he gets that chief's permission to do so. The chiefs down there have more power over the natives than the white man does.^[11]

By the early twentieth century a distinction had emerged, at least in the view of the officials of the Transvaal Native Affairs Department, between a predominantly highveld district such as Ermelo, where the population was regarded as 'detribalized' and 'unattached' to chiefs, and a middle and lowveld district such as Barberton, where two-thirds of the population was regarded as 'attached' to chiefs. In the latter district almost half of this population was living in an unsurveyed block of 500 square miles of Crown Land lying south of the Lomati river and adjacent to the borders of Mozambique and Swaziland.^[12] A witness to the Transvaal Labour Commission of 1903 commented that only an experienced observer could tell apart the Swazi of Swaziland and the Swazi of this district, as 'really only an imaginary border' separated them. As late as 1935, Van Warmelo, who had been unable to discover any information in official records on the chiefs of Ermelo district, or to make any reliable estimate of the ethnic composition of the district, described the Swazi of the Barberton district as 'almost an integrating part [sic ] of the Swazi nation'.^[13]

The official distinction between the 'detribalized' highveld and the 'tribal' middle and lowveld areas reflected as much as anything the division of the African labour force between the farms and the mines. Crown and company land was the major source of labour for the mines, while labour on the highveld was immobilized in the interests of the local farming community who continued to complain for many years about a labour shortage. Pressure from the mining companies induced the government in 1921–22 to purchase 20,000 morgen of land in the Piet Retief district which was to be held in communal tenure and was intended to relieve congestion on company farms and keep up the flow of labour to the mines.^[14]

While it would be an oversimplification to equate the survival of chiefly authority with the maintenance of ethnic identity or consciousness, there was probably some correlation. Certainly chiefs were to be vital agents in the mobilization of ethnic power at a later date. That chiefs and 'tribal' institutions were repressed rather than eliminated is demonstrated by the fact that the government was able to rediscover chiefs, when it eventually wished to do so in the 1950s, even in districts such as Ermelo which had for many decades been officially 'detribalized'. While officials in the years immediately following the Anglo-Boer War showed little interest in either chiefs or ethnicity in the Ermelo and Carolina districts, the 1904 census did estimate, on uncertain criteria, the Swazi element as 55 per cent of the total. It found a significant minority of Swazi in the neighbouring districts of Piet Retief and Wakkerstroom, as well as in Standerton and Lydenburg. The largest proportion was found in the Barberton district, to the north of Swaziland, where about 63 per cent of the total population was said to be Swazi.^[15]

The later 1880s brought a rush of concession hunters into Swaziland, and these 'bought' the land and its minerals several times over. As a result of a deal between Britain and the Transvaal, in 1895 the country became, much against the wishes of the Swazi themselves, a protectorate of the Transvaal. The convention of 1894 protected, on certain conditions, the rights of the Swazi king over his subjects, and

also included a non-disturbance clause which protected the rights of Swazi occupiers against the claims of concessionaires, whose titles had generally been recognized by a special court in 1890.^[16] Following the Anglo-Boer War, the country was administered as a part of the Transvaal from 1902 to 1906, when it was saved from complete incorporation into the Transvaal by the impossibility of sorting out the tangle of disputed concessions before the transfer which was provided for in the Schedule to the Union of South Africa Act of 1909 could be effected.^[17]

The discussions which preceded the land partition of the latter year provide classic material on questions of colonial social engineering. Drawing on evidence to the Lagden Commission and on the slightly earlier partition of Zululand, there was lengthy debate on the amount of land required to sustain the average family and on the advantages and disadvantages for government and settlers of a few large or many small reserves. The Zulu Rising of 1906 overshadowed the partition, and while this prompted the High Commissioner, Lord Selborne, to advocate caution, the settler leader and newspaper editor, Allister Miller, demanded the 'denationalization' of the Swazi as the only guarantee of settler security.^[18] While Miller saw reserves as an impediment to progress, the Swazi leaders also opposed reserves because most of the country had been alienated through medium-term grazing leases which safeguarded rights of cultivation. These areas would eventually revert to the nation. The eventual compromise between the settler and official positions resulted in a three-way division of the country between the concessionaires, the Swazi, and the Crown. The Swazi were led to believe that the bulk of Crown Land would be available for their use, but in fact it was almost all sold off in later years to finance recurrent governmental expenditure.^[19]

Almost half of the Swazi within colonial Swaziland were left outside the 32 scheduled reserves and were given until 1914 to decide whether to come to terms with their landlords as labour tenants or to move into the reserves. Care was taken in demarcating the reserves to inconvenience chiefs as little as possible and important graves were included in Swazi Nation areas. The partition was intended to satisfy the conflicting interests of settlers who sought labour tenants and of Witwatersrand mining interests, which owned concessions in Swaziland and sought labour migrants. Lord Selborne did not envisage that the reserves would become the home of all Swaziland's Swazi or that those who moved into them would be able to meet all their needs through agriculture. The partition would have failed in its objects if it had.^[20]

The consequence of the partition was to bring about by 1914 in Swaziland a situation which was broadly similar to that in the Transvaal. Most of the highveld passed into the hands of white owner-occupier, though often seasonally absentee, farmers with labour tenants. Much of the middle and lowveld remained in effective Swazi occupation as reserves, Crown land and unoccupied farms. In so far as the Swazi of Swaziland had some land reserved specifically to them they were better off than their relatives in the Transvaal, but only marginally so. Chiefs in Swaziland whose people came to be divided between reserves and farms began to confront many of the difficulties of jurisdiction which had already affected their counterparts in the Transvaal.^[21]

The Swazi Elite and the Changing Political Economy

After the death of King Mbandzeni in 1889 there appears to have been a long period during the minorities of his son, Bhunu, and his grandson, Sobhuza, when

the Swazi ruling elite took very little public interest in the Swazi of the Transvaal. Although there are occasional references in the speeches of councillors to the former extent of the Swazi kingdom and to the shifting of boundary beacons, there seems to have been an acceptance of the new boundaries as an indisputable fact of life.^[22] In all the voluminous documentation relating to the Swazi deputations to England in 1894 and 1907, for instance, as well as in numerous other petitions, and the records of izindaba, there is hardly a reference to the Swazi of the Transvaal.

As late as 1913 a deputation from Swaziland, under the leadership of Prince Malunge, son of Queen Labotsibeni and uncle of the young Prince Sobhuza, visited Barberton to pay their respects to the High Commissioner, Lord Gladstone. They presented a long list of grievances but made no reference to the Swazi of the Transvaal.^[23] But in the following year the manager of Moodie's Gold Mining and Exploration Company, on whose land the Mjindini royal village near Barberton stood, commented that the Swazi in the district 'still cling strongly to tribal ties and look, I believe, as much to Swaziland for guidance as to the Union Government'.^[24]

There is ample evidence that the two other major royal villages at Mbhuleni and Mekemeke also maintained close contact with the royal house in Swaziland. There were also a number of other Swazi chieftaincies which kept up their links with Swaziland. The Shongwe chief, Matamu, on the northern border had in 1914 an induna and followers in Swaziland.^[25] Mbudula, the chief of the Mahlalela, whose territory had been split three ways between the Transvaal, Swaziland and Mozambique, maintained homesteads in all three countries, though his headquarters was in the Transvaal. His clan acquired a reputation for independence, but he, when giving evidence to the Beaumont Commission in 1914, claimed: 'I am a Swazi.'^[26] Hilda Kuper summarized the position in the mid-1930s as follows: '. . . outside the territory in the Barberton, Carolina and Piet Relief districts are large groups of loyal subjects who inform their king, Sobhuza, of any important events, occasionally send him tribute, receive emissaries from him and have him ratify the appointment of their local chiefs'.^[27]

From about 1910, the Swazi Queen and council, after a lengthy period of introspection in which their entire attention appears to have been concentrated on the battle to avoid Swaziland's incorporation into the Transvaal and on resistance to the land partition, began to adopt a more extrovert position, seeing that advantages could be gained from participation in South Africa-wide political activity. As a result of the land partition and high taxation, the Swazi of Swaziland were becoming much more involved in labour migration to the Witwatersrand where they were lumped together with the South African Swazi as a single category.^[28] Above all, the Schedule to the Union of South Africa Act, providing for the incorporation of the High Commission Territories, and the common knowledge that Swaziland was first on the list for transfer meant that the Swazi had nothing to lose, and possibly something to gain, from adopting a less parochial stance. They now saw threats to the interests of South African 'Natives' as equally a threat to themselves.

Queen Labotsibeni put the position very frankly in 1914 when she was reported as saying:

that as Swaziland would no doubt enter the Union at some future date she was in sympathy with any efforts tending towards the betterment of the conditions under which Union natives live, and for this reason her son Malunge had become a member of the Native Congress and her people had assisted in

contributions for the proposed delegation to England in connection with the Union Land Act.^[29]

Queen Labotsibeni had earlier made contact with Pixley kalsaka Seme, who had recently returned as a lawyer from the United States and Britain. He drafted a petition for the Swazi council in 1912, and in the same year Queen Labotsibeni provided almost all the capital for Seme's paper, Abantu-Batho, which the Swazi royal family continued to support until its collapse in the late 1920s. It was founded in connection with the South African Natives' National Congress which held its first meeting in January 1912.^[30]

The congress was seen by its founders, among whom Pixley Seme was a prime mover, as a response to Union. Its aim was to foster the unity of the 'natives' and to combat 'tribalism'. The South African 'tribes' were not, however, seen in themselves as an impediment to unity but as the building blocks from which unity could be constructed. The Paramount Chiefs of the major southern Africa 'tribes' were appointed as honorary presidents of the organization and were viewed not only as a potential source of financial support but also as a bridge between the growing African intelligentsia and the rural masses. The congress, which was conceived along parliamentary lines, was to have an Upper House of chiefs and a Lower House of commoners. The legitimacy of the chiefs as the 'natural' rulers of their people was assumed. Seme himself, who eventually married princesses of the Swazi and Zulu royal families, was a consistent champion both of chiefly authority and of 'race pride', by which he seems usually to have meant ethnic or 'tribal' loyalty.^[31]

The primary interest of the Swazi Queen and council in the congress was undoubtedly in the campaign against the Land Act. They had embarked on a campaign to buy back concession land from its holders and therefore had a special interest in the restrictions on purchase which it imposed. The extension of the terms of the Act to Swaziland would bring an end to their schemes. Prince Malunge, together with Benjamin Nxumalo, the brother of the young Queen Mother Lomawa, and uncle of Sobhuza, played an active part in the early meetings of the congress, as did Josiah Vilakazi, secretary to Queen Labotsibeni. When Prince Malunge attended the special congress held in 1914 at Kimberley to protest against the Land Act, he was treated as the most distinguished delegate.^[32]

The premature death of Prince Malunge in 1915, which was publicly mourned by congress leaders J.L. Dube, Pixley Seme and Sol Plaatje, did not end the assertion of a wider South African role by the Swazi royal house.^[33] In 1916 Cleopas Kunene, one of the first editors of Abantu-Batho and a member of the Swazi deputation to England in 1894, organized an extravagant reception in Johannesburg for Prince Sobhuza who was on his way to school at Lovedale.^[34] A few years later, in 1921, Sobhuza was to buy six stands and a house in Sophiatown which became a meeting place for the Swazi in Johannesburg.^[35]

The first serious attempt at intervention by the Swazi Queen on behalf of the Transvaal Swazi came in 1918 in connection with the implementation of the section of the Land Act which provided for the creation of additional reserves. Queen Labotsibeni heard in January 1918 that the Eastern Transvaal Land Committee was seeking evidence and asked the Resident Commissioner in Swaziland to inform it that it was 'my wish as well as the wish of all the Swazi living in the Transvaal' that they be given 'a strip of land [from] the Barberton line right down to the Pongola river near Chief Sithambe'.^[36] The High Commissioner, Lord Buxton, indicated that her views would be put to the committee, but if they were, they can have had little impact as there was no question of any highveld area in the

eastern Transvaal being set aside as reserve.^[37]

Evidence was also given to the committee at Ermelo by a delegation of the South African Natives' National Congress, most of whom were Swazi, and described themselves as such, and one of whom, Joseph Hlubi, had close ties with the royal house. They expressed their desire to remain on the highveld, whether in reserves or not, and protested vigorously at the proposal contained in the Native Administration Bill to charge licence fees for squatters and labour tenants in a drive to reduce the highveld population to the status of full-time servants or to force them to move into lowveld reserves.^[38] The committee made an impassioned plea for a gradualist approach. It stressed the need for development in new reserves and pointed out in prophetic language that: 'to transport Natives even from the exiguous conveniences of settled life in non-Native areas suddenly and in large masses to areas which they still have to prepare before they can exist in them would be dangerous in the extreme'.^[39]

The Botha government withdrew the bill, as, faced with the conflicting claims for labour from highveld farmers and the mining companies, it was unable to get a majority for it. The same conflict ensured that the scheduling of reserves was indefinitely delayed.^[40] Meanwhile in 1919 the Swazi Queen and council made one further attempt to intervene on behalf of their 'subjects' in the Transvaal. In a trenchant petition, presumably drafted by Seme and clearly influenced by the tone of Wilson's Fourteen Points, they demanded among other things the recognition of 'the independence of Swaziland with its own sovereign power' and the provision of land for the Swazi in the districts of Barberton, Carolina, Ermelo, Piet Relief, and Wakkerstroom. The petition referred for the first time to what would be a recurring theme: promises allegedly contained in the conventions of 1881 and 1884 that 'locations' would be established for the Swazi in the eastern Transvaal who were said to be living under a 'veiled form of slavery'.^[41] The responses to this petition and to petitions in 1921 and 1922 which sought the recognition of Sobhuza as king of the Swazi on both sides of the border were negative.^[42]

At about this time the continuing influence of Queen Labotsibeni over the Swazi of the eastern Transvaal was vividly demonstrated by the experience of a group of South African Natives' National Congress fund-raisers who visited the area in connection with the sending of a deputation to Europe. Swazi chiefs in the eastern Transvaal told them that nothing could be given without the consent of the Queen Regent. She told them that they could collect in the Transvaal but that all funds raised should be brought to her for allocation as she too was planning to send a deputation. When the President of the Congress, S.M. Makgato, sought to use official channels to claim the resulting £500, the Queen's Secretary informed the Resident Commissioner that the Swazi of the Transvaal had only agreed to contribute for the Swazi deputation.^[43] When Sobhuza had his first ncwala in 1921, it was reported that many Swazi from outside Swaziland attended. Before he left on his long-planned deputation to London in 1922 he paid a visit to the Mjindini royal villages in the Transvaal at Barberton to say farewell and thank you.^[44]

In the years following World War I, the position of the Swazi in the eastern Transvaal deteriorated seriously. Even before the war, pressure on squatters on the highveld had begun to increase as a result of the subdivision of farms and the increase of commercial farming. Chiefs with their retainers and often large herds of stock were especially vulnerable. Early in 1914 the Mbhuleni royal village had set off on the first of a long series of migrations. It was to move five times by 1949.^[45] After the war similar pressures began increasingly to be felt in the Barberton

district as a result of the sale or lease of land to new settlers. These pressures resulted not only in frequent evictions, but also in a deterioration in the terms which squatters could obtain from the landlord. Many squatters were forced to become labour tenants, and many labour tenants had to become full-time servants at a minimal wage.^[46]

These economic pressures were accompanied and intensified by a barrage of legislation intended to reduce the rural population to serfdom. The poll tax was doubled in 1925 and amendments to the pass laws in 1929 and 1930 increased the immobility of farm labour. The Native Service Contract Act which was finally passed in 1932, despite the opposition of the mining lobby, provided for the licensing of squatters and labour tenants, for the tying of whole families to labour service, for whipping in the case of breach of contract, and for summary eviction. The act, which could not be totally enforced, aimed to redistribute labour within the farming sector while the pass laws blocked lines of escape to the towns.^[47]

Whites and Blacks Press for 'Retribalization'

Meanwhile the Native Administration Act of 1927 provided for the establishment of a separate legal system for the African population, for their government by proclamation of the Governor General, which meant the Native Affairs Department, and for the strengthening of the powers of chiefs in reserves over their people. The measure was opposed by the intelligentsia, who lost the possibility of exemption from customary law, and by the chiefs, who saw both that they were being reduced to the status of salaried government officials and that the creation of the Governor General as 'Supreme Chief' was 'a violation of [the] Bantu system of government'.^[48] The act was widely seen as marking a shift in policy towards segregation and 'retribalization', even though it was welcomed by some radicals as a final blow against the legitimacy of chiefs and the 'tribal system', which Herbst, the Secretary of the Native Affairs Department, saw as 'the most efficient machinery of government in definitely native areas'.^[49]

The case of Chief Mhola, of Mjindini, one of the three most important Swazi chiefs in the Transvaal, however, vividly demonstrated that the policy of 'retribalization' applied only to malleable chiefs in reserves, and not to independently minded chiefs who were squatters on white farms. It also demonstrated the reawakening of ethnic consciousness as a defensive strategy against proletarianization on the part of a rural elite and the strong official hostility towards such manifestations. Chief Mhola, who was a contemporary of Sobhuza and had been educated with him at the Swazi National School at Zombodze, was installed as chief in 1923. In the following year he was evicted, together with 37 other Swazi homesteads, from a farm which lay within the 'town lands' of Barberton. He sought the assistance of Sobhuza and his council who in 1925 asked the Swaziland administration to intercede through the High Commissioner on his behalf. They asked if he could be given space in a 'location' or on Crown Land.^[50] The High Commissioner, Lord Athlone, who believed that the South African government had its eyes on the High Commission Territories as dumping grounds for 'surplus' population, took up the case. He felt that South Africa should make proper provision for its population within its own borders and may have feared that the eviction of prominent Swazi would lead to a flood of immigrants into Swaziland where congestion in Swazi Nation areas was already a problem.^[51]

The response of Herbst clearly justified such fears. He indicated his preference that Mhola should remove himself to Swaziland as he had clearly shown in the past that

he considers he is resident in the Transvaal as a representative of the Paramount Chief of the Swazis rather than as a subject of the Government. His influence with the Swazis in the Barberton district has been such as to hamper smooth and efficient administration and his continued residence in this area is considered undesirable.^[52]

Sobhuza and his council were equally insistent that Mhola and his people should not come into Swaziland. Sobhuza stated: 'He is in charge of all the Swazis there and if he does not look after them, they will become vagabonds.'^[53] At a slightly later date Mhola's councillors repudiated with equal vehemence the suggestion that they could 'return' to Swaziland. They acknowledged that they were Swazi but stated that 'we do not regard Swaziland as our home, our home is here where we originated'.^[54]

Herbst did hold out some prospect that reserves might be created in the district if Hertzog's draft Native Bills were passed, but he complained that Mhola had 'no special status' and that 'his affinity with the Swazi Chief confers no claim to consideration upon him . . . .' He pointed out that in terms of the Convention of 1894, 'the interests of the Swazi Chief were to be confined to the territory of Swaziland'.^[55]

No reserves were in fact scheduled in the Barberton district for many years and Chief Mhola and his followers survived as labour tenants on farm land for at least thirty-five years. In remarkably lucid evidence to the Native Economic Commission in 1930 several of his councillors protested vigorously against the terms which they had to accept as labour tenants. Acknowledging that there were differences in wealth within the 'tribe', they demanded the creation of a reserve into which the most 'hard-pressed' could move. They made it clear that these would include the chief and the larger cattle owners who always found it difficult to maintain their herds as labour tenants.^[56]

Living as close as they did to the town of Barberton, Mhola's councillors also displayed an acute awareness of the economic possibilities of the local urban area. They protested that they were treated in the urban area in the same way as 'other natives'.^[57] This was presumably a reference to the influx of 'Shangaans', 'Nyasas' and others who less than twenty years later were said to acknowledge Mhola as chief and pay tax as his followers. They protested at their inability to buy plots or stands, or to engage in business within the urban area, as well as at indiscriminate arrests of people found drinking beer there.^[58] They clearly believed that their historic claim to the area entitled them to privileged access to whatever commercial opportunities existed. There was a remarkable similarity between their urban demands and those of Swaziland's intelligentsia, which were articulated, as will be shown below, in the same year.

An alliance between the intelligentsia and the chiefs, as will be recalled, was one of the basic premises upon which the South African Natives' National Congress had been established. This was an alliance which, by the later 1920s, had been seriously eroded by a number of factors. These included the failure of congress either to reverse or to have fully implemented the terms of the Natives Land Act of 1913, and the increased rate of permanent African urbanization which followed the growth of secondary industry during World War I. Attendance by chiefs at the annual conferences of the ANC declined, as did their financial contributions to it, and special conventions of chiefs had to be called in 1927 and 1928 to obtain authoritative chiefly responses to Hertzog's Native Bills and the Native Administration Act.

At the same time the Industrial and Commercial Workers' Union (ICU) provided a mass-based alternative to the ANC and swept like a millenarian

movement through the eastern Transvaal and parts of Swaziland itself. Many Swazi chiefs, on both sides of the border, joined with thousands of their followers. Sobhuza's uncle, Norman Nxumalo, brother of Benjamin, was actively involved as an organizer in the Transvaal. This may have increased the acceptability of the union among Swazi chiefs who were drawn to it, as they had previously been drawn to the ANC, by the hope of regaining their land. The ICU could do nothing to satisfy the expectations which it had aroused, and by 1928 it was fading out of the eastern Transvaal as rapidly as it had appeared two years before.^[59]

Following the collapse of the ICU, and in the face of the increasingly repressive policies of Hertzog and Oswald Pirow, who fought the 1929 election with the slogan of 'Swart Gevaar ' ('Black Peril'), there was a move within the ANC to restore the alliance with the chiefs. The proven ability of Swazi chiefs in the eastern Transvaal to deliver their followers to the ICU may have inspired Pixley Seme, Sobhuza's close associate and legal adviser, to advocate this strategy, though his motives were undoubtedly conservative. At the annual conference of the ANC, in April 1930, J.T. Gumede, who was sympathetic to the Soviet Union and who wished the ANC to adopt a more militant stand, was, after a dramatic call for the 'Native Republic', replaced by Seme. There is no evidence that Sobhuza actively canvassed on Seme's behalf, but he certainly was pleased with the result. A Swaziland branch of the ANC was established by Benjamin Nxumalo at Sobhuza's house in Sophiatown in the same year, and it is probably not fortuitous that Norman Nxumalo stands behind Seme in the 1930 ANC group photograph.^[60]

A move in a similar direction was made by A.W.G. Champion, the leader of the ICU yase Natal, the foundation of which had been seen by Clements Kadalie as exploiting ethnic divisiveness. Champion, who had in 1929 scorned the government view that 'Natives' should be ruled through chiefs, and who had supported Gumede in the ANC election, was by August 1930 seeking to forge an alliance with Solomon kaDinizulu, the de facto Zulu paramount. The threat of such an alliance led to his banishment from Durban to Johannesburg where he made contact with, amongst others, Sobhuza.^[61]

Seme, who had little in common with Champion, was also committed to the 'full restoration of the paramount chieftainship of the Royal House in order to revitalize all the Zulu native institutions [and so] re-establish the old esprit de corps of the Zulu nation'.^[62] Seme presided over the ANC at a time when it was almost destroyed by internal dissension. He stated his own political philosophy clearly in 1932 when he answered those who accused him of 'culpable inertia'. He wished the ANC to return to its 'original constitution' by reviving the upper House of Chiefs where they could discuss 'the new problems which face them today, the problems of employment for their people, the problems of conserving national pride, customs and traditions'. He also called for the support of the younger intelligentsia while denouncing the 'common agitator who only wants to create strife and class hatred' as well as 'hatred between whites and blacks'. He deplored 'detribalization' and the fact that 'the Chiefs and their uneducated people are despised and forsaken by their educated tribesmen'. He called for a revival of 'Race Pride' and for a restoration in confidence between the educated and uneducated. He made it clear that he looked to ethnic mobilization as a counter to the disintegration caused by education, urbanization and proletarianization.^[63]

Sobhuza II and Ethnic Mobilization

me signally failed to revive the ANC along these lines, but he did assist Sobhuza number of cultural initiatives which he undertook in Swaziland at the time.

Sobhuza's failure to have the land partition reversed through the deputation to London in 1923 and through the case against Allister Miller which the Privy Council finally rejected in 1926 prompted him to take a fresh interest in ethnic mobilization. By then Sobhuza was particularly concerned at the effects of labour migrancy on Swazi society, the decline in royal and chiefly authority, the growing division between educated and uneducated and between Christian and 'pagan', and by what he saw as the breakdown of discipline and morality among the youth. The Ballingers, who visited Swaziland in 1931, commented on the relative looseness of royal authority away from the royal capitals, while Sir Brian Marwick recalls the independence of the chiefs on the Lebombo in the mid-1920s and Sobhuza's efforts to 'tame' one of them through a dynastic marriage. From the late 1920s, therefore, Sobhuza and his council embarked upon a deliberate policy of reviving royal authority and central control.^[64]

There was a superficial coincidence of interest between Sobhuza and white advocates of 'indirect rule', though their objectives were fundamentally different. Sobhuza and his council did not scruple to use the colonial authorities to bring to heel recalcitrant chiefs, but they were to fight a long and bitter battle to preserve their own independence and legitimacy. A new Resident Commissioner, T. Ainsworth Dickson, was transferred from Kenya to Swaziland in 1928, and he was expected to prepare the territory for some form of 'indirect rule'.^[65] Among his first moves was to call upon Sobhuza to compile a 'Swazi National Constitution'. With the help of Seme and A.G. Marwick, who had served in Swaziland since 1903, Sobhuza produced a lengthy and elaborate memorandum entitled 'The Original Swazi Political Organization'. This outlined the positions of the King, the Queen Mother, the chiefs, and the councils, the general council, or Libhandla and the executive council, or Liqoqo . It was calculated to present Swazi procedure in the most favourable light, stressing as it did the customary checks on the arbitrary use of royal power.^[66]

At the same time, Dickson prompted the establishment of the Swaziland Progressive Association which was intended to bring together members of the intelligentsia to discuss matters of common concern. He evidently hoped to break down the renowned conservatism of the Swazi National Council, either by introducing into it 'new men' from the rising intelligentsia or by establishing a more representative Native Advisory Council.^[67] Sobhuza did not oppose this move, but he did fear that it could become a rival centre of influence to the Swazi National Council, and that it might fall into the hands of the non-Swazi Africans who predominated in the teaching and clerical professions. His fears were allayed by the election as first president of his uncle, Benjamin Nxumalo, the natural leader of the intelligentsia, a leading layman in the African Methodist Episcopal Church, a long-serving member of the Liqoqo and a former secretary to both Sobhuza and his grandmother. Furthermore, the constitution expressly subordinated the association to the Swazi National Council, and laid down that complaints should be channelled through it.^[68] There was, as already noted, a close similarity between its first demands and those of Chief Mhola's councillors at Barberton. They called for the establishment of locations close to the towns of Mbabane and Bremersdorp (now Manzini) where stands could be bought; for the easing of restrictions on 'Native' access to retail trade; and of the controls on beer-drinking in urban areas. Although Swaziland was soon to be hit by the full force of the Depression, a number of educated members of the 'traditional' elite were attempting to establish themselves in retail trade at this time. Sobhu himself is alleged by one source to have dabbled in trade under the cover of st men. Only a handful of Swazi were able to obtain trading licences at this time,

however, and it was not until after the war that the Swazi National Council, itself, entered the commercial arena through the establishment of a pressure group, 'The Swazi Commercial Amadoda' (Amadoda = men).^[69]

The simultaneous establishment of the Swazi National School at Matsapa was another initiative with which Sobhuza was more closely identified. This was intended to become 'a genuine national undertaking and cater for the cultural, social, and industrial development of the Swazi people'.^[70] There was to be an academic stream leading to the matriculation class at Fort Hare University College, but it also had a strong emphasis on agricultural training. It was intended that the school should 'take advantage of the better elements in the traditional native code and culture', though an early visitor, Sir Alan Pim, saw considerable difficulty in the way of its doing this.^[71]

In 1933 Sobhuza wrote a memorandum on 'Native Education' in which he criticized the education then being provided by the missions. His grounds were that:

(a) It causes the Swazi scholar to despise Swazi institutions and his indigenous culture;

(b) It causes him to become ill-fitted to his environment;

(c) It releases him from the wholesome restraints which the Swazi indigenous method of education inculcated, and does not set up any effective substitutes for them.

To counter the disintegration of rural life which had aroused concern in both Sobhuza and Seme, he proposed an extension of national schools and a revival of the regimental or age-grade system, the Ibutho .^[72] His proposals unleashed a storm of protest from the missions and the settlers, from some of the 'native intelligentsia', and from Mr and Mrs Rheinallt Jones, who were then attempting to introduce the segregated Pathfinder and Wayfarer movements into Swaziland. Sobhuza had the active support of A.G. Marwick who believed that 'a modernized age-grade system more in vital touch with the conservative elements in native life' might protect the youth 'from the objectionable form of hooliganism known as Amalyaita' . Sobhuza was able to recruit the support of a number of anthropologists including Mrs Winifred Hoernlé, Isaac Schapera, and the young Hilda Beemer, later Mrs Kuper, as well as Bronislaw Malinowski, who accompanied her on her first venture into Swaziland in 1934.^[73]

Although the old regimental system, involving not only military training but also tribute labour, was clearly moribund, Sobhuza and his council had begun in the early 1930s to exact fines from young men who married without permission. While some settlers feared that Sobhuza intended to use the regiments as a form of cheap labour, the missionaries were publicly concerned that Christian youth would be contaminated by 'pagan' sexual practices. A.G. Marwick felt that the battle was principally one for control over the educational process.^[74]

The Ibutho was eventually established on a trial basis at the Swazi National School. The curriculum of training included 'Swazi history, custom, lore and law' as well as 'ceremonial'. Although the experiment continued for a number of years, it was ultimately a failure. There was a good deal of resistance among the pupils themselves, many of whom objected to the singing of the Swazi anthem, 'Inqaba kanqofula' and to attendance at the ncwala which had been revived after Sobhuza's installation in 1921.^[75] Many of those who had passed through the regiment in the 1930s, however, later formed part of an educated conservative elite. The first indvuna of the Ibutho was Mfundza Sukati, who was to become the Deputy Prime Minister of Swaziland at independence, and the second indvuna

was J.S.M. Matsebula, who became Secretary to the King in 1967 and was the leading Swazi historian, publishing Izakhiwe ka Ngwane (in Zulu) in 1952, and A History of Swaziland in 1972. He was to play a leading part in the development of siSwati as a written language as the author of a series of primary school texts published in the early 1970s.^[76]

Sobhuza also had some success in reviving the regimental system outside the schools. Reinforcement of the system came with its use for the recruitment of the Swazi contingent during World War II. The recruiting centres, or tinkhundla, were to be developed in the 1950s as the new, but allegedly 'traditional', basis of local government and in the 1970s for parliamentary elections. There was not much place for girls in the Ibutho, but in 1935 Sobhuza revived the umncwasho, a two year pledge of celibacy by adolescent girls with a reciprocal pledge in relation to them by all men.

Another area of initiative at this time lay in the compilation of nationally-oriented history. J.J. Nquku, a graduate of Edendale, said variously to have been of Mpondo or Zulu origin, was appointed in 1929 as the first supervisor of 'Native' schools. Sobhuza encouraged him to write a work of history which was published in Zulu as Amaqhawe ka Ngwane (Heroes of Swaziland) in 1939. Although he was himself an Anglican, Nquku played a leading part in the contemporary movement, in which Sobhuza was also interested, for the establishment of a Swazi National Church pulled together from a variety of Zionist sects. He was also one of the founders in 1934 of Izwi lamaSwazi, Swaziland's first vernacular language—at this date Zulu—newspaper. He continued to play a prominent part in public life, reviving in 1949 the Swaziland Progressive Association, which, in 1960, he was to transform into Swaziland's first political party. Only then were his non-Swazi origins turned against him, and he was effectively excluded from political life.^[77]

It may be surprising that there was no effort made at this time to promote the Swazi language. With the arrival of Zulu-speaking 'Amakholwa' (Christian) evangelists and missionaries in Swaziland from the 1880s, Zulu had become the language of church and school. This was in spite of the marked difference in pronunciation, syntax, and vocabulary between it and Swati. The influence of Zulu was such that Queen Labotsibeni had to insist that Sobhuza was brought up to speak Swati as well as Zulu, which was his mother's tongue.^[78]

Linguistic nationalism seems to have progressed no further in the 1930s than the adoption of Swati forms of surnames and, in some cases, the public use of Swati first names rather than Christian names. The apparent lack of interest in the promotion of Swati may be traced to the fact that Zulu influence within Swaziland's intelligentsia was strong. At a time when there were very few jobs for the educated in Swaziland itself, there were sound practical reasons for learning to read and write in the most widely used of southern African languages, Zulu.

There was, in fact, some official consideration in the early 1930s of the feasibility of producing literature in Swati, but it was concluded that because of the relatively small size of the Swazi population, it was improbable that a full range of school texts would ever be produced.^[79] It was not until the publication of D. Ziervogel's A Grammar of Swazi in 1952 that official interest in the development of Swati was revived.^[80] In the following year a committee was established under the chairmanship of A.G. Marwick 'to investigate the question of a Swazi Orthography and the introduction of Swazi as a written language'. This committee produced a draft orthography in 1956 but decided that there should be no change in the schools until written Swati had been further developed.^[81]

It was not until the fresh upsurge of cultural nationalism in the 1960s that the

question of Swati as a medium of instruction in primary schools became a political issue. A national commission which sat from 1967–68 considered the question and proposed the introduction of Swati into the first three years of school from 1969. Ziervogel produced a revised version of his book in 1967 in collaboration with Enos Mabuza, an inspector of schools who was in 1977 to become the chief executive councillor of the KaNgwane Territorial Authority. He had established a Swati Language Committee in 1974 on which representatives from Swaziland sat as the first practical expression of cooperation between the two entities, and he supervised the introduction of Swati into KaNgwane's schools in 1978.^[82]

Sobhuza II and the Swazi of South Africa

While there seems to have been little or no demand for the development of Swati in the 1930s, and while the impact of Sobhuza's other cultural initiatives on the Swazi outside Swaziland may have been limited, he did make a number of moves intended to involve the Swazi of South Africa. He established in 1931, at his house in Sophiatown, the Swazi National Royal Club, which was intended to provide a social centre for all Swazi on the Witwatersrand, regardless of origins, and to promote 'all aspects of Swazi welfare'. This club has had a long and active life and survives at Kwa Thema, Springs, but another scheme for a Swazi Labour Institute which would have been concerned with the recruitment and welfare of Swazi workers apparently came to nothing.^[83]

The major initiative in which both Sobhuza and Seme were involved at this time was 'The Petition of the Swazi Tribes of the Eastern Transvaal to the Union Parliament', signed and dated by ten Swazi chiefs at the Ntfonjeni royal village in Swaziland on 25 March 1932. The petitioners requested reserves for their 'tribes' in the areas in which they currently lived in the districts of Barberton, Carolina and Ermelo. The printed version of the petition was prefaced by splendid photographs of Sobhuza and his mother taken during the visit of the Prince of Wales to Swaziland in 1925. According to Seme, these photographs were included 'only for sentimental reasons and to indicate the nationality of the humble Petitioners'. In a brief sketch of the history of the Swazi he pointed to some of the finer points of the 'Swazi constitution', such as the high respect paid to women, and staked an historical claim to territory as far north as the Sabi river.^[84]

The Native Affairs Department took more than two years to reply to the petition, but the Native Trust and Land Act which was finally passed in 1936 did provide accommodation for four of the Barberton chiefs, all of whom had resided on or close to Crown Land in the lowveld. It was, however, to be many years before they received official recognition. This partial success prompted Seme to draft another petition in July 1936, which may never have been submitted, but which was intended to be signed by the same chiefs, and to make a special plea for land for the chiefs of Mbhuleni, Mjindini, and Mekemeke who remained on farm land.^[85]

At this stage the question of the transfer of Swaziland to the Union came once again to the fore. The provision of at least some land for the Swazi on his northern border seems to have led Sobhuza to take a more positive interest in the possibility of incorporation as a means to the 'unification' of his people. Officials in Hertzog's entourage on a visit to Britain in 1937 planted the rumour that Sobhuza would be prepared to consider a deal which would 'formally bring under his allegiance the large number of Swazi hitherto outside the borders of Swaziland'.^[86] A year or so later, Sobhuza held discussions in Swaziland with Douglas Smit,

Secretary for Native Affairs in the Union, which touched on the position of the Swazi in the Transvaal. He subsequently outlined in great detail the historical case for the creation of further reserves, making once again a special plea for the three royal villages. Smit maintained that Swaziland itself had always been regarded as the Swazi 'location' and stated quite frankly that: 'our difficulty is that the districts of Carolina and Ermelo are purely European and we have no scheduled Native Areas or released area there'.^[87]

On the outbreak of World War II, Jan Smuts replaced Hertzog as Prime Minister of South Africa and decided, much to the annoyance of the British, to press for the immediate transfer of Swaziland. His emissary, Deneys Reitz, indicated that 'Sobhuza could be got round'. As inducements. Smuts was prepared to offer the 'purchase of land for the enlargement of Swazi reserves', as well as reconsideration of railway plans and a reduction in 'Native' taxation.^[88] Apparently in connection with this deal, Sobhuza's lawyers sought the Governor General's consent for the purchase by the Swazi Nation of sixteen farms in the Barberton, Carolina and Ermelo districts, all but two of which were close to the borders of Swaziland. The farms included the land on which the three best known royal villages were established. Nothing came of the deal, but it is possible to see in it the germ of the later KaNgwane 'land deal' proposal.^[89]

As late as 1947 the attitude of the Department of Native Affairs to Swazi chiefs of the royal house had changed very little from the time of the Mhola case over twenty years previously. James Maquba Nkosi, who had been installed as chief at Mbhuleni in 1931, and who maintained close contact with Sobhuza, was, not for the first time, under threat of eviction. The Secretary for Native Affairs complained that, though he was not recognized by the government as a chief, he was apparently related to Sobhuza and 'regarded as the Chief of the Umbhuleni section of the Swazis in the Eastern Transvaal'. As he had for over ten years declined to be resettled in the lowveld, and in view of 'Sobhuza's interest', the department wanted to know if it would be possible for him with his wives, followers, and stock (68 cattle and 72 goats) 'to remove to Swaziland'.^[90] Sobhuza and his council were quite determined that he should remain where he was. He should not be expected to move into the lowveld, which was malarial and belonged to the Shongwe people. His removal from the district would leave 'the remaining natives in the area without a chief'. If the government could not provide a reserve in the Carolina district, it was preferable that he should move to another farm rather than to the lowveld or Swaziland. It was eventually reported that Chief Maquba had 'found asylum' on another farm. Sobhuza took the opportunity to protest that not a single Swazi chief in the Transvaal had been officially recognized.^[91]

By the time that the Afrikaner Nationalist government came into power in 1948 there is, then, little evidence that the Swazi of the eastern Transvaal had been subjected to any policy of 'retribalization'. There is plenty of evidence that at least their chiefs wanted to be 'retribalized', but had managed to make very little headway in that direction. It may not be surprising that the chiefs on the highveld farms had made such little progress, but it is surprising that even the chiefs on Trust land in the Barberton district had failed to achieve recognition. There had, however, been one or two indications that the government was taking more interest in ethnicity. The Native Commissioner at Bushbuckridge had sought, on his own initiative, Sobhuza's advice in 1940 as to who should be regarded (not recognized) as the senior Swazi chief in the newly created Nsikazi reserve in the Nelspruit district.^[92] The ethnological section within the Department of Native Affairs began in 1946 to collect information for a survey of The Tribes of

Barberton District which was published in 1949. A similar work on the Carolina district was not published until 1956, by which time the much-reviled Chief Maquba, though still not officially recognized, had become a' groot-kapitem' with numerous subordinate headmen and merited a fall page photograph.^[93]

Though the Bantu Administration Act of 1951 provided the machinery, in the form of a three-tier system of Tribal, Regional, and Territorial Authorities, which was to be used in the political evolution of 'Bantustans', it was not until the publication of the Tomlinson Commission report in 1955 that a blueprint for the consolidation and development of the reserves along ethnic lines was produced. The commission saw Swaziland as the nucleus of a Bantustan which would include reserves on its borders in the eastern Transvaal and would have access to the sea through the Ingwavuma district. It was, therefore, the source of the later proposals for the consolidation of KaNgwane and for the proposed 'land deal' of 1982.^[94]

Dr Hendrik Verwoerd in 1959 indicated that if Swaziland accepted South African 'guardianship', it would be able to incorporate the Swazi areas of the Transvaal to form a single 'black unit'.^[95] Rumours circulating in Swaziland in 1963 suggested that Sobhuza and the Swazi National Council were preparing to make a deal with South Africa which would involve the cession of some land in the south of the country in exchange for territory to the north. These rumours were taken seriously enough to be strenuously denied.^[96] Verwoerd seems to have reconciled himself somewhat reluctantly to the forthcoming independence of the High Commission Territories, but in 1966 the then Minister of Bantu Affairs repeated that South Africa would be prepared to cede territory to its neighbours to create ethnic units. He pointed to the illogicality of establishing an independent Swazi state within the borders of South Africa if Swaziland itself was about to become independent.^[97]

Political Developments in Swaziland After World War II

The years after World War II saw dramatic changes in Swaziland. The development of forestry and of irrigated sugar and citrus production, as well as mining and the completion of the railway, transformed a stagnant economy. Opportunities for local employment were greatly increased and dependence upon labour migration to South Africa reduced. Labour was, in fact, drawn into Swaziland from Mozambique, Nyasaland and South Africa to meet the demands of the expanding economy. Political development was relatively slow. From 1939 until 1950 Sobhuza had been largely preoccupied with resistance to the government's schemes for indirect rule, realizing that the bureaucratization of the chiefs, including himself, would undermine their authority. He was largely successful in this campaign, ensuring that the appointment and dismissal of chiefs was kept out of government's hands and that they did not become salaried officials.^[98]

When, in 1959, the colonial government, moving with the times, broached the question of constitutional development in Swaziland, Sobhuza and the Swazi National Council's executive committee—the Liqoqo —moved quickly to pre-empt the development of mass-based political parties. They proposed a deal with the conservative settlers under which each group would elect by its own 'traditional' methods an equal number to the new legislative body. Control over Swazi law and custom, as well as land and minerals, would be vested in the king and council.

Sobhuza's initiative prompted J.J. Nquku to form the Swaziland Progressive Party on the basis of the old Progressive Association. Nquku's party soon split and the majority of his following joined the Ngwane National Liberation Congress, which was founded and led by Dr Ambrose Zwane and Prince Dumisa Diamini, a nephew of the king. Both were closely associated with the 'traditional' hierarchy, as were Dr Alien Nxumalo, son of Benjamin Nxumalo, and Simon Sishayi Nxumalo, who emerged as the leaders of the Swaziland Democratic Party.^[99] The Ngwane National Liberation Congress sought to acquire a mass base by involving itself with the widespread labour unrest which led to a crisis in 1963 and the calling in of British troops. The 'traditional' methods of labour representation through works indvunas, which had been developed in the Witwatersrand's mines from the 1920s and applied in Swaziland during the 1940s, had failed, and the Resident Commissioner, Brian Marwick, blamed both the Swazi National Council and the employers for their failure to heed warnings. The Swazi National Council blamed foreign labour and foreign agitators for the situation.^[100]

Sobhuza was under strong pressure from the political parties and the colonial authorities to remain above politics and prepare himself for the position of constitutional monarch. He was, on the other hand, pressed by the Liqoqo and the settlers to venture into politics. Finally, after the rejection by the Libhandla, or general council, of his power-sharing scheme, he launched in 1964 the Imbokodvo National Movement as the political arm of the Swazi National Council.^[101] Sobhuza's major tactic against the political parties was to label them as 'foreign', divisive and hostile to Swazi 'tradition', of which he was not only guardian but the unchallengeable interpreter. The fact that, with the exception of Nquku and the South African Swazi, MacDonald Maseko, who was prominent in the NNLC, all the leaders had impeccable Swazi credentials does not seem to have worked against this tactic. Sobhuza's major assets were the prestige he had acquired over 40 years of low-key resistance to colonial rule, the authority of the chiefs, and the foundations of cultural nationalism which he had himself begun to lay in the 1930s and continued to consolidate thereafter. He also enjoyed the support of substantial capitalist interests which saw in him and Swazi 'tradition' a bulwark against more radical forms of African nationalism.

His party won all the contested seats in the 1964 elections after which many of the political leaders, such as the two Nxumalos and Dumisa Diamini, threw in their lot with it rather than face prolonged exclusion from public life. They had in any case obtained only a derisory share of the vote, with the chiefs throwing their full weight, including alleged threats of banishment and dispossession, behind the Imbokodvo. The party was also helped by the constituency boundaries which favoured the rural areas and by the disenfranchisement of black, but not white, South Africans living in Swaziland.^[102]

The independence of Swaziland was delayed until 1968 mainly because of the stubborn refusal of Sobhuza and the Swazi National Council to allow control over land and minerals to pass to the cabinet and parliament. They insisted that these should be vested in the king and council. This made possible the establishment of the Tibiyo takaNgwane Fund' to which royalties and dividends on joint ventures were to be paid. Under the control of members of the Swazi National Council, this fund was, after independence, to be a major backer of prestige projects, such as the national airline, but also provided capital through salaries, loans and other means for the embourgeoisement of a section of the traditional elite whose financial resources had always been slender. The fund was also used for substantial purchases of land, the market in which was brought under 'traditional' control by the Land Speculation Act of 1971. This measure greatly incensed

Sobhuza's capitalist backers, but they were satisfied with the hostility of his government towards trade unions and the attempt which was made to modernize the 'traditional' system of labour control through the king's representatives or Ndabazabantu .^[103]

The Cultural Element in Swazi Politics

The rise of the Imbokodvo was accompanied by a fresh upsurge of cultural chauvinism. The Swati language was given official recognition in the independence constitution and was introduced into the schools in the following year.^[104] There was a noticeable increase in popular participation in Swazi ceremonies such as the Ncwala . The churches discreetly dropped their opposition, and leading Swazi traditionalists, who were also members of mission churches, such as the first prime minister, Prince Makhosini, attended for the first time. Sobhuza's revived regiments played an overtly political role on at least one occasion, though he did decline to allow their use as auxiliary police in Mbabane in 1963.^[105]

The most vivid expression of this cultural nationalism was a new ethnic exclusivism which was directed not so much against European settlers as against Mozambicans and black South Africans, who constituted a significant proportion of the labour force. An attempt by the colonial government to tighten controls on African immigration in 1959–60 had failed owing to the lack of cooperation from the Swazi National Council which resented interference with what it considered its own prerogatives.^[106] After the post-Sharpeville influx of refugees and the labour unrest of 1963 there were frequent demands for action to be taken against foreign labour, though there was a fundamental difference of opinion between the Swazi National Council and the government as to who was a Swazi and who was not. The government believed that long residence in Swaziland qualified an 'African' to be regarded as Swazi, while the Council maintained that a person could only become a Swazi through ukukhonta —allegiance to a chief. They were determined to keep control over citizenship in their own hands and were able, after independence, to achieve this. Control over the process of ukukhonta gave the 'traditional' authorities a potent economic and political weapon, as well as a useful source of funds.^[107]

The question of the status in Swaziland of the South African Swazi was a complex one. Sobhuza had as early as 1960, in the context of a colonial deportation ordinance, expressed concern that this could be used against Swazi people from South Africa. But as Swaziland approached independence there was popular resentment against immigrants from South Africa, sometimes known as 'paper Swazi', who were thought to be cashing in on claimed Swazi roots to the detriment of the Swazi of Swaziland. There is evidence that Sobhuza himself, and some of his ministers, such as Prince Bhekimpi, who was to become prime minister in 1983 and who, as chief at Nkaba royal village, had followers in the Transvaal, saw the South African Swazi as a special case, but the law made no exception for them.^[108]

The case of Thomas Bhekindlela Ngwenya, however, which became a cause célèbre in 1972–3, demonstrated that at least some of the ruling group were prepared to put political expediency before any notion of an inclusive Swazi ethnicity. Ngwenya, as parliamentary candidate of the Ngwane National Liberation Congress, defeated the Imbokodvo candidate. Prince Mfanasibili, a nephew of the king, and was deported to South Africa. While it was never denied that Ngwenya was Swazi, he was alleged to have been born on the wrong side of the then unmarked border. After he had satisfied a court that he had indeed been

born within Swaziland, the hearing of citizenship cases was transferred from the courts to a hastily created tribunal, and he was again deported. After the tribunal was itself declared illegal, the independence constitution was suspended and parliament handed all power over to the king and council.^[109] Among the laws promulgated after this was an exceptionally exclusive citizenship law which further entrenched ukukhonta as practically the only was of acquiring Swazi citizenship.^[110] The case was probably the excuse for, rather than the cause of, the suspension of the constitution which the king and council had accepted under protest, but it did have the effect of hardening the distinction between the Swazi of Swaziland and the Swazi of South Africa, and increased the insecurity of people of marginal status. It was hardly surprising that Ngwenya was reported in 1982 to be campaigning in KaNgwane against the land deal.^[111]

Ethnic Mobilization Falters Amongst South African Swazi

The development in Swaziland of an exclusive Swazi nationalism had no immediate counterpart in the eastern Transvaal, where ethnic mobilization in the context of Bantustan development had hardly started before 1968. In some ways economic developments in the eastern Transvaal had run parallel with those in Swaziland although the large-scale production of citrus, fruit and vegetables had been developed before World War II in the lowveld around Nelspruit and Barberton. After the war there was a rapid development of agriculture and forestry together with related industries. Developments on the highveld were not so spectacular, but by the 1960s mechanization and afforestation, together with natural population increase, were ending the area's chronic labour shortage and producing a 'surplus'. A major concern of the Tomlinson Commission had been the accommodation of people who were now seen as redundant on 'white' farms but of potential use to plantations and industry as migrant labourers. The Bantustan strategy was devised to provide both 'homes' and social control for such people while simultaneously splitting African nationalism into ethnic fragments.

The eastern highveld farms were in the 1950s an important area of rural support for the ANC under the leadership of the Ermelo-bom 'Lion of the East', Gert Sibande.^[112] The banning of the ANC and Pan Africanist Congress in 1960 did not lead to an immediate channelling of political feeling into ethnically based organizations. The chiefs of the two reserves designated as Swazi were organized in 1959 and 1962 into the Nkomazi and Nsikazi Regional Authorities, but it was not until 1968 that the Swazi National Council of South Africa was established as a pressure group to mobilize the Swazi in the urban areas, 'black spots', and 'white' farms in support of 'separate development'. Its organizer was David Lukhele, a former Seventh Day Adventist evangelist and life insurance salesman, who was also the 'Swazi editor' of Africa South, a paper produced by Lloyd Ndaba, a former South African Information Department official, who promoted a number of ethnically based political parties in the 1960s.^[113]

It was hardly a coincidence that this 'political party' was founded at a time when the systematic resettlement of people from neighbouring white areas into the Swazi reserves in the Barberton and Nelspruit districts was just beginning. 'Commuter' towns were established in the Nsikazi reserve for workers removed from the municipal townships of Nelspruit and White River. The removal of 'surplus' population from farms and 'black spots' was also beginning. There was

never any possibility of these areas being able to support a fraction of their population through agriculture. With a minimum of local employment, the choice for most people was between commuting, labour migration, or starvation. The proclaimed towns were soon surrounded by large squatter settlements and health conditions rapidly deteriorated. The Nsikazi reserve was to be the epicentre of the serious cholera outbreak which hit the eastern Transvaal in 1980.^[114]

A number of problems delayed the development of a Swazi Bantustan. One was the fact that the majority of Swazi chiefs, including the most important Dlamini ones, were still living on 'white' farms or in 'black spots'. So long as they remained there they could not be officially recognized and a Swazi Territorial Authority could not be established. The government was faced with the problem of providing them with a reserve while adhering to its commitment not to add a single hectare of land to the provisions of the 1936 Land Act. A further problem was that the government itself appeared uncertain as to whether a Swazi Bantustan was desirable or whether the South African Swazi could be unloaded on Swaziland without one. It was soon found that the Swazi leadership in South Africa was divided on the issue of their future relations with Swaziland.

As late as November 1970 the South African government evidently hoped that it could avoid the necessity of finding additional reserve land in the eastern Transvaal, and the inevitable conflict with white farming interests which that would entail, by purchasing freehold land for the resettlement of its Swazi within the borders of Swaziland. It is not clear whether this proposal had the support of the Swaziland government, but it was in any case rejected by the South African Swazi National Council. In 1971 a Swazi Interim Committee composed of representatives of the Swazi National Council and two Swazi Regional Authorities was established. This was intended to pressure the government to complete the consolidation of the Bantustan and to pave the way for the establishment of a Territorial Authority.^[115]

There was never, of course, any question of the Swazi Bantustan being consolidated in such a way as to match the actual distribution of Swazi people on the ground. The 1972 proposals envisaged the concentration of the Swazi in a single block of territory along the northern and northwestern borders of Swaziland. This would clearly facilitate a deal with Swaziland in the future, but it involved the excision of the Nsikazi reserve, with the exception of the planned towns which would remain as dormitories for Nelspruit and White River. The Pongola reserve, to the south of Swaziland, which was regarded as Swazi but which had not been constituted as a Regional Authority, was also be to excised.^[116]

The proposals naturally incensed the Nsikazi chiefs and their people who were threatened with removal after 50 years or more of residence. Even the government-sponsored BENSO organization, which was concerned with the economic development of the Bantustans, was alarmed at proposals which involved the removal from the Swazi Bantustan of the only area in which there was any infrastructure at all. For the government the proposals had the merit of providing for the first time land on the highveld into which the 'surplus' population could be removed. They must also have been made in the knowledge that King Sobhuza had little interest in the outlying Nsikazi reserve where the population, although predominantly Swati-speaking, included many people of Sotho, Tsonga and Zulu origin.^[117]

These proposals had the effect of exacerbating the tensions which already existed within the Swazi National Council and the Interim Committee between those who advocated Bantustan development and opposed a land deal with Swaziland, on the one hand, and the supporters of such a deal on the other hand.

While the Swazi National Council claimed branches all over the central and eastern Transvaal and involved all the Swazi chiefs in white areas, it is impossible to judge to what extent it represented Swazi opinion. Many of those who were nominally Swazi no doubt rejected ethnic politics entirely, but it is evident that within these organizations majority opinion was opposed to any deal with Swaziland. The lesson of the Ngwenya affair was not lost on the South African Swazi. David Lukhele was at this stage particularly critical of what he saw as the domination of the Dlamini clan in Swaziland and of the banning of all political parties which accompanied the suspension of the constitution in 1973. Lukhele did not totally rule out the possibility of amalgamation with Swaziland but maintained that this should happen, if at all, after direct negotiations between a self-governing Bantustan and Swaziland, and not as a result of a deal concluded by the South African government. Swazi National Council literature was also critical of the 'interference' by Swaziland in the affairs of the South African Swazi.^[118]

An example of such interference came with the appointment in late 1972 of Dr Lancelot Gama as King Sobhuza's personal representative, or Indvuna General, in South Africa. Dr Gama, a medical practitioner in Springs, was a leader of the Swazi National Royal Club which was apparently at loggerheads with the Swazi National Council in South Africa. By the end of 1973 Dr Gama had established a rival 'Swazi Nation of the Republic', which had the support of a number of Swazi National Council branches and of three chiefs, including Chief Johannes Mkolishi Dlamini, the son and heir of James Maquba Nkosi, of Mbhuleni, who was regarded as the senior Swazi chief in South Africa.^[119] Chief Mkolishi was still technically a squatter on a farm near Badplaas. It was only in 1975 that he received the official recognition which had eluded his father for so long. He then became chief of the Eerstehoek resettlement camp in the new Regional authority of Mswati, which had been established for the three senior Dlamini chiefs in the newly acquired lands along the Swaziland border.^[120]

Politics and History in a Bantustan Under Apartheid

The establishment of the Mswati and Mlondozi Regional Authorities in the area provided some land for most of the Swazi chiefs in white areas and made possible the inauguration of the Swazi Territorial Authority in 1976. Chief Mkolishi was elected chairman of the executive committee, despite the fact that he was clearly in the minority on the question of relations with Swaziland. David Lukhele was not elected to the executive committee, but Enos Mabuza became Councillor for Education and Culture.

Within little more than a year a split had occurred over the question of whether KaNgwane, as it now came to be known, should progress to the next stage of Bantustan development, which was self-government with a Legislative Assembly. Sobhuza was known to be opposed to self-government for the homeland as he correctly anticipated that it would create or encourage vested interests which would be hostile to fusion with Swaziland. In resisting this development, Chief Mkolishi was therefore acting in Sobhuza's interests, though his reluctance to sign documents relating to land consolidations may have reflected his own dissatisfaction with arrangements that left his royal village of Mbhuleni outside of KaNgwane. The majority opinion, however, felt that the acquisition of self-governing status would strengthen the bargaining position of KaNgwane with

the South African government. The Territorial Authority voted to remove Chief Mkolishi and replace him with Enos Mabuza. The Legislative Assembly was rapidly constituted and, after litigation with Chief Mkolishi, Mabuza was able to consolidate his position. David Lukhele joined Mabuza's 'cabinet' as member for Justice and Community Affairs.^[121]

Both Mabuza and Chief Mkolishi then proceeded to form political parties. Chief Mkolishi founded the 'Inyatsi ya Mswati' which sought to unite all Swazi so that 'we will . . . be able to press the South African Government for a fair deal in so far as the allocation of land for Swazis is concerned'.^[122] He and a minority of the officially recognized Swazi chiefs campaigned for an ethnically 'pure' KaNgwane and protested at the continued presence in the territory of 'Shangaan' and other minority groups.^[123] It was in 1978 that Chief Mkolishi with ten other chiefs formally petitioned King Sobhuza to begin negotiations on their behalf for their incorporation into Swaziland.^[124]

Chief Mkolishi's failure to obtain the support of more than a minority of the Swazi chiefs was an indication that the appeal to a kind of Swazi chauvinism, which had worked so successfully in Swaziland itself, was out of touch with 'traditional' opinion in South Africa. It also reflected the fears of some chiefs of Dlamini dominance, or more precisely, royal autocracy, as Chief Mkolishi was not able to muster the support of all the Dlamini chiefs. Above all, it reflected doubts about the desirability of union with Swaziland. Chief Mkolishi did his best to exploit his royal status, holding an annual sibhimbi (dance) for several days at Mbhuleni, which in the early 1980s was attended by up to 5000 people. He could not, of course, hold an ncwala ceremony as to do so would be tantamount to rebellion against the Swazi king. On his periodic visits to the Provincial Supreme Court in Pretoria, he was accompanied by busloads of supporters in 'traditional' dress.^[125]

It is probable that Chief Mkolishi's limited power base lay with the rural farm population, such as his own Mbhuleni people, who had been long resident close to the Swaziland border and who identified with Swaziland. Government policy was from the mid-1970s, however, acting to transform sparsely populated highveld sheep farms into vast semi-urban squatter settlements. These were the classic 'dumping grounds' where people were forced to live in appalling conditions with no prospect of local employment. Chief Mkolishi's father had claimed suzerainty over all the Swazi of the Carolina and Middleburg districts and beyond, but the resettled population of the new areas, though nominally Swazi, were often Zulurather than Swati-speaking people who had been cleared from urban 'black spots' such as Doornkop, near Middleburg, and Kromkrans, near Carolina, as well as from farms all over the eastern highveld.^[126]

The paradoxes inherent in Chief Mkolishi's position became evident from 1982 when he emerged as not only the leading South African Swazi advocate of the land deal and chief at Eerstehoek of about 50,000 relocated people, but also as a prominent symbol of resistance to forced relocation. He consistently refused to move himself and three hundred other families, his royal village of Mbhuleni, from land which 'belonged to the Swazi people by history and blood'. He was given three months' notice to move into KaNgwane in October 1983, but he had not done so by August 1984. There is no reason to doubt that he was a sincere advocate of the land deal, but he may also have calculated that such a deal presented the best chance of getting a readjustment to the land dispensation which would preserve his ancestral home.^[127]

The position of Enos Mabuza, who formed the Inyanda National Movement in 1978, was also full of paradox, though he was a more sophisticated exponent of

ethnic politics. He regarded Chief Gatsha Buthelezi as his political mentor, modelled his movement on 'Inkatha', and proclaimed that it welcomed members from all ethnic and racial groups and that it was opposed to discrimination on grounds of 'tribe or clan'. He joined Buthelezi's Black Alliance and rejected both 'independence' for KaNgwane and any deal with Swaziland.^[128] At a later date he stated that he saw KaNgwane as 'a region of South Africa, and not as an independent political entity or a vassal state to be of Swaziland'.^[129] If KaNgwane is 'a region of South Africa', it is clearly an ethnically defined region, and Mabuza's frequent protests that he does 'not believe in ethnicity at all' cannot be taken at face value. In a typical interview he stated: 'I was born in an area where there was a Swazi [royal] kraal, and I speak siSwati, but my existence in South Africa is not different from that of my Zulu, Xhosa or Tswana brothers'.^[130] Mabuza is, however, in his own right a latter-day 'culture broker' who has played a leading part, as a linguist, in the promotion of Swati as a written language and in its introduction into the schools. He has also lent his name to efforts to create for KaNgwane a distinct symbolic identity.

These efforts, which were doubtless masterminded in a South African government department specializing in such work, had a distinctly synthetic air. A 'Mother Culture' was employed to advise on matters of custom and ceremonial. She was somewhat improbably said in South African promotional literature to be consulted on occasion by King Sobhuza himself. The Speaker's mace of the Legislative Assembly was an extraordinary confection of Swazi symbols surmounted by a bronze bowl with gold flame 'symbolizing the strength of Swazi culture'. The problem for Mabuza was that as long as King Sobhuza lived, he had a virtual monopoly of authentic Swazi symbols and could also delegate their use to his man, Chief Mkolishi. Mabuza himself was not immune to the king's charisma, expressing on occasion his respect for him and speaking of maintaining good relations with Swaziland on the 'cultural' level.^[131]

Sobhuza himself had decades of experience in the manipulation of cultural symbols and clearly had no doubt that it would be possible to create an artificial South African Swazi or KaNgwane nationality. It was for this reason that he opposed every stage of Bantustan development. In 1981 he advised a deputation from KaNgwane, including Mabuza, not to press its demand for the granting of the second stage of self-government, as this would 'lead to the Swazis in South Africa becoming independent from their king'. He anticipated that the South African Swazi would not only have their own flag and anthem, but might ultimately appoint their own king and hold their own Ncwala ceremonies.^[132]

There is, of course, no evidence that Mabuza had any such intentions. He does not appear, in the lifetime of King Sobhuza at any rate, to have pressed the claim that the South African Swazi constituted a distinct ethnicity, and his efforts at 'nation-building' seem to have been somewhat half-hearted. This may have reflected his own ambiguity on the issue, but may equally have reflected the uncertainty of the South African government as to whether it should promote KaNgwane on the road to 'independence' or seek a deal with Swaziland. Such a deal, envisaged by the Tomlinson Commission in 1955, and provided for in the consolidation proposals of 1972, was a long time in maturing. The independence of Mozambique in 1975, and the subsequent upsurge of activity by, and support for, the ANC undoubtedly increased the interest of some elements within the South African government for such a deal, as did the failure of the 'independent' Transkei, and other Bantustans, to achieve recognition internationally.

There was, however, some tension within the government and the Nationalist party on this issue. Successive ministers responsible for Bantustan affairs, M.C.

Botha and Dr Piet Koomhof, were unenthusiastic and continued to press for Bantustan development as late as 1980, although discussions on border adjustments had evidently started in the mid-1970s. In the end, it was the Minister for Foreign Affairs, Dr Pik Botha, and strategic planners, who pressed for the deal, not only as a step towards a 'constellation of states' but also as a way of 'incorporating' Swaziland and reducing the threat of infiltration by the ANC through the country. Swaziland is, from a strategic point of view, a part of the Transvaal and has a border with South Africa which is as long as that of Mozambique.^[133]

It became clear in 1984 that the land deal had been offered to Swaziland as an inducement to sign a secret non-aggression pact, under the terms of which Swaziland would clamp down on ANC activities. The agreement, signed in February 1982, was a precursor of the Nkomati Accord with Mozambique of March 1984, which, if it held, would reduce the value of the deal with Swaziland and allow South Africa to renege on its commitment without serious strategic loss. It is too early to assess the significance of the dissolution of the Rumpff Commission in June 1984 and Dr Koornhof's announcement that South Africa did not intend to proceed with the land deal in the immediate future, but it is safe to say that South Africa's ultimate actions will be dictated by its own strategic self-interest and considerations of realpolitik .^[134] No government has had more experience in the cynical manipulation of ethnicity, and it was no doubt aware that a plausible ethnic case could be made both for and against the land deal, though its validity in terms of international law was open to serious question.

Once the land deal became a practical possibility, if not a probability, in 1982, Enos Mabuza was forced to argue against it in ethnic terms. It was only then, and following the death of King Sobhuza in August 1982, that he appears to have maintained explicitly that the South African Swazi had become a distinct ethnicity and that there had been since the mid-nineteenth century a 'divergence between the two groups such that each has developed an independent socio-economic and political character of its own'. He went on to argue that many of the Dlamini chiefs in the eastern Transvaal had come there as refugees or rebels rather than as an integral part of the Swazi nation, and that the history of the Ngomane, Mkhatjwa and Mahlalela clans was 'completely independent of the Dlamini dynasty and domain'. The latter arguments were, to say the very least, controversial, and he had to concede that the 'military outposts' of Mbhuleni, Mjindini and Mekemeke had 'a direct relationship with the Swazi monarchy'.^[135]

He maintained that with the establishment of the KaNgwane Legislative Assembly in 1977 the South African Swazi had been recognized as an autonomous political unit with 'their own country, KaNgwane'. He rejected the South African government's claim that the land deal would 'bring together those who belonged together' and asked whether this principle had been applied in granting separate independence to the Transkei and Ciskei. He also asked whether the South African Swazi were the only 'national group' which overlapped into an adjacent independent state outside South Africa and cited the cases of the Xhosa, the Southern Sotho, the Tswana, and the 'Shangaan-Tsonga'.^[136]

These arguments were countered by David Lukhele, who had parted company with Mabuza in 1981 and joined forces with Chief Mkolishi as an advocate of the land deal in 1982. Lukhele neatly inverted Mabuza's historical argument by maintaining that most of those who opposed the land deal were not Swazi, but 'Shangaan' or Sotho. The opposition of undeniably Swazi chiefs, such as Chief Tikhontele Dlamini of Mekemeke, was put down to their feeling that the land deal had not gone far enough. He dismissed the socio-economic argument as fallacious

and maintained that the Swazi were being divided in the interests of the personal ambitions of politicians who were the creation of the South African government. There could be only one Swazi nation and the Swazi had no desire to be divided like the Xhosa and the Tswana.^[137]

Both Mabuza and Lukhele were guilty of special pleading, but Lukhele was clearly unable to explain, except through the self-defeating argument that the population was not Swazi, why it was that a majority of the chiefs and others who had engaged in Bantustan politics were opposed to the deal. The suggestion that they were government puppets was difficult to sustain as they were opposing its policy. Nor did Lukhele produce a convincing refutation of the socio-economic argument, which Mabuza himself did not elaborate. It is a truism that Bantustan politics are elite politics. Support for Mabuza may have been broadened by his opposition to the land deal, but his original power base lay in the small business and professional class which emerged in the planned towns of the Nsikazi reserve, which was finally reprieved from excision in 1977.^[138] Such people had established vested interests in the Bantustan which could only be threatened by a land deal with Swaziland where salaries were lower and business opportunities were tightly controlled by an established elite. Mabuza also had the support of prominent Swazi businessmen operating in Pretoria and on the Witwatersrand, such as A.J. Sibanyoni and Ephraim Tshabalala (not the Soweto 'millionaire' of the same name), who were the first directors of the KaNgwane Economic Development Corporation, established in 1979.^[139] Witwatersrand-based businessmen may have seen participation in Bantustan affairs as providing salaries, status, possible business opportunities, including the chance to acquire freehold land, as well as protection, through collaboration with the system, for their insecurely based urban ventures. They are most unlikely to have supported a deal which would almost inevitably undermine their position in the towns.

For the mass of the population of KaNgwane who were totally dependent either on commuting or labour migration to the towns and farms of 'white' South Africa there could be no possible advantage in the deal. Even less could it benefit the majority of the South African Swazi who remained outside KaNgwane in the central and eastern Transvaal. To such people, ethnic politics were, if not an irrelevance, certainly a luxury. Mabuza's stated commitment to the maintenance of a unitary South Africa, even if it was to be like Seme's vision of 1912, a South Africa composed of ethnic building blocks, reflected harsh socio-economic realities. The conclusion of the land deal is now, following the installation of the new king of Swaziland, Mswati II, in April 1986, and the disgrace of Prince Mfanasibili and pro-deal members of the Liqoqo, most unlikely. But if it were concluded, it would in all likelihood result in ethnic tensions between South African Swazi and the Swazi of Swaziland, for it would result not only in 'denationalization' and the loss of citizenship, which had in any case been compromised by the Bantu Homelands Citizenship Act of 1970, but also loss of livelihood.

If the majority of the South African Swazi can be presumed to have been opposed to the land deal, it is equally difficult to see what the Swazi of Swaziland stood to gain from it. There is no doubt that for King Sobhuza 'border adjustments' meant the reclamation of the land of his ancestor, Mswati, and the reunification of his people. It is, however, by no means certain that he was satisfied with the terms of the deal as they emerged in June 1982, leaving in South Africa, as they did, the royal villages of Mbhuleni, Mjindini and Mekemeke.^[140] He made no reference to the deal in his last public speech on 22 July 1982, just a few weeks before his death. Some of Swaziland's leaders may have seen little to lose in

closer identification with South Africa, the bastion of capitalism in the region. Some civil servants may have been attracted by higher South African salaries and a larger field for their activities, but many of Swaziland's leaders, including King Sobhuza's last prime minister, Prince Mabandla, had doubts if only on grounds of security and Swaziland's probable loss of international status. They did not all relish the prospect of becoming South Africa's policemen in the rounding up of members of the ANC, towards whom Sobhuza had always been sympathetic, conscious as he was of his family's leading part in its founding. It is impossible, at this stage, to unravel the Byzantine intricacies of the succession crisis which followed Sobhuza's death, and which saw the gazetting of the Liqoqo as the 'Supreme Council of State' and the removal first of Prince Mabandla and then of the Queen Mother, Dzeliwe, but it is clear that disagreement over the land deal was one factor in the imbroglio.^[141]

If the land deal were to go ahead, Swaziland would acquire, apart from the dubious advantage of access to the sea at Kosi Bay, a relatively small area of undeveloped and disease-ridden territory with a relatively dense de facto population and a de jure population of close to one million people. Several hundred thousand people, who had been forcibly relocated on Swaziland's borders in nominally rural, but practically urban slums, would become part of its population. The probability was that it would become a dumping ground for more of South Africa's 'surplus' population. It would lose its much-prized ethnic homogeneity with the addition of significant ethnic minorities in the Nsikazi and Ingwavuma areas. The degree of compatibility between the Swazi of South Africa and those of Swaziland was in itself clearly in doubt.

Conclusion

The history of the two halves of the Swazi nation since World War II tends to confirm Mabuza's contention that there has been a 'socio-economic and political divergence'. While KaNgwane was created as a labour reservoir, Swaziland's export of labour declined relative to its population. Swaziland experienced dramatic changes in the period after World War II. The development of forestry and of irrigated sugar and citrus production, as well as mining and the completion of the railway, transformed a stagnant economy. Opportunities for local employment thus greatly increased. In 1981 as many as 57,000 labour migrants and 40,000 commuters from KaNgwane worked in white South Africa. Swaziland, with a far larger population, supplied only 13,000 labour migrants and few, if any, commuters.^[142] While very few of KaNgwane's population had any real access to land, the majority of Swaziland's population, even if in agricultural or industrial employment, retained some stake in the soil. As a crude indicator it could be shown that the ratio of cattle to people in Swaziland, at over 1:1, was four or five times that in KaNgwane.^[143] Swaziland also remained one of the least urbanized countries of Africa. While it is true that the volume of South African investment in Swaziland was large and growing, and that Swaziland was closely tied to South Africa through the Customs Union and the Rand Monetary Area, it did have a viable domestic economy.

On the political level it was clear that KaNgwane's petty bourgeoisie had developed vested interests in the Bantustans and was bitterly opposed to transfer. Ironically, if the South African Swazi had developed a distinct ethnicity, this was at least in part because of the exclusive cultural nationalism which had developed—or been developed—in Swaziland itself. Some of those who, at the time of the Ngwenya affair, espoused a view of Swazi ethnicity which confined it to the

people of Swaziland alone are now among the leading exponents of a wider view. Prince Mfanasibili, for example, featured prominently in both camps. This may be taken as demonstrating that Swazi ethnicity is a highly volatile concept and that ambiguity in relation to it is not the sole prerogative of Enos Mabuza. In Swaziland there has been, among the leaders anyway, some tension between an exclusive and an inclusive view of Swazi ethnicity. Among the South African Swazi the tension has been between the few, like Chief Mkolishi, who give primacy to their Swazi identity, and the many who are either indifferent to it, or, like Enos Mabuza, value it, but in the final analysis put their identity as South Africans first. The impact of the latest state of emergency on the lowveld towns of KaNgwane, which has been reflected in more strikes and stayaways, as well as in the visit of Mabuza and his entire 'cabinet' to the ANC in Lusaka in March 1986, and the assassination of David Lukhele at his home in Pretoria in June of that year, tend to confirm that ethnic politics should increasingly be seen as no more than a sub-theme in the broader struggle for South African national liberation.

This study has shown that KaNgwane would make for Swaziland not only an unpalatable, but almost certainly an indigestible, meal. If the tensions resulting from the conclusion of the deal, which may now be unlikely, were to express themselves in ethnic terms, as between South African Swazi, and the Swazi of Swaziland, this would not be a case of irrational prejudice, but the product of many years of different historical experience. Radical differences have emerged during these years in social structure, in relations with the dominant South African economy, and in political development as well as in the realm of ideology. Chief Albert Luthuli showed prescience when, after a meeting with King Sobhuza almost thirty years ago, he sensed that 'people are clinging to a dream of a return of the old Swaziland—a dream incapable of fulfilment now . . . .'^[144]