Zwide's invasions very nearly obliterated the Ngwane state. Sobhuza was forced to flee far to the north to escape successive Ndwandwe attacks, and was only able to return after Zwide became embroiled in new conflicts in the south. Zwide was eventually defeated by Shaka in 1819, but Sobhuza had no intention of reliving the nightmare with a newly triumphant Zulu power. He chose rather to colonise the north, absorbing a diverse collection of Sotho, Tsonga and Nguni chiefdoms into a restructured Ngwane state. By the end of his reign Swaziland had begun to assume its modern form. This chapter concerns itself with the early vicissitudes of Sobhuza, the construction of a transformed Swazi state and the external policies that Sobhuza initiated to protect his new realm.

Zwide's invasions shook the Ngwane state to its core. Sobhuza is usually pictured in Swazi historiography as executing a tactical withdrawal to a sanctuary in the north, from the security of which he quickly reconstructed the Ngwane state. Nothing could be further from the truth. Rather than resuming a barely interrupted career of expansion, he was obliged to abandon the bulk of his followers to Zwide, and to lead the life of a rootless refugee whom Zwide hounded further and further north.

This study was embarked upon while I was a student at the School of Oriental and African Studies, University of London, and completed after I had taken up a teaching position at the University of the Witwatersrand. It bears the imprint of the varied intellectual concerns of scholars at both institutions. The SOAS African History Seminar, the SOAS and Institute of Commonwealth Studies Societies of Southern Africa Seminar and the University of the Witwatersrand African Studies Seminar each helped influence the direction this enquiry has taken. Thanks are due to all those participants who wittingly or unwittingly guided me on my way, and in particular to Dr Shula Marks, who introduced me to southern African studies and supervised the thesis upon which this study is based. Without her encouragement and support this book would never have been written.

A debt of gratitude is also owing to all those who gave their assistance and encouragement to my researches in Swaziland – in particular to Prince Makhungu Dlamini, who opened doors which would otherwise have remained closed; to Arthur Dlamini for giving me the benefit of his wide knowledge of Swazi history and oral historians, and for his unstinting help in a variety of other ways; to Mr T. Simelane and Mr F. Buckham for their excellent translations of interviews; to Mr J. Masson and to Mike and Fiona Armitage for smoothing my way after my arrival in Swaziland; and to a host of officials from the Department of Local Administration and Community Development who helped with my interviewing in their individual districts.

The Swazi state congealed out of the turmoil and flux of late eighteenth–century northern Nguni society. Expelled from the eastern side of the Lebombo by a rampant Mabudu power bent on the monopolisation of trade, the Swazi were early casualties of the growing integration of the area into European commodity exchange. The dislocation induced by these pressures is suggested by the dynastic schisms of the time, but the Swazi were ultimately able to surmount present adversity and turn it to future gain. Squeezed from the east, they infiltrated the Shiselweni area, lying north of the Pongola and west of the Lebombo, where, through a process of conquest and assimilation, they forged a new and powerful Swazi state.

The area on which they alighted was well suited to their mixed agricultural and pastoral way of life. Like the centres of the Ndwandwe and Mthethwa, two of the other leading northern Nguni states, it boasted particularly desirable configurations of natural resources. Year-round grazing was possible, and the basin of the Ngwavuma River allowed for the easy cultivation of fertile river banks. Cattle and human population multiplied accordingly, particularly in the period of heavier rainfall of the late eighteenth century. From its very inception, therefore, the Swazi state was caught up in the two dynamic forces shaping northern Nguni society: competition for trade both within and between rival chiefdoms, and competition for scarce combinations of natural resources which could underpin a powerful and expansive tributary state.

For the Swazi, as for other northern Nguni societies, these were the real midwives of change.

Mswati died at the beginning of August 1865, leaving behind a notably ambiguous legacy. On the one hand, an infrastructure of state had been firmly established, so that the country now divided more on class than on ethnic or chiefdom lines, and looked for the resolution of its grievances in the machinery of state. On the other, Mswati had died very young – he had yet to reach forty at the time of his death – and left behind no acknowledged heir. Successive efforts to obtain a main wife from Mzilikazi and Mpande had miscarried, and before new approaches were under way Mswati was dead. The resulting struggle provides an interesting commentary on the new-found maturity of the Swazi state, both in so far as all parties were competing to control rather than dismember society, and in the way in which the defeated party reconciled itself to its position once the die was cast. This chapter will examine the implications of that struggle and the regency that followed, together with their repercussions on Swaziland's regional position.

A leading contender for the kingship was Mbilini, the son of laMakhasiso, Mswati's chief wife at Hhohho. Constitutionally Mbilini was debarred from taking office, because his mother was Mswati's first wife, but there are signs that Mswati had considered him as a possible successor. Mbilini certainly thought so, and claimed that Mswati had announced precisely that as he lay dying in the arms of his brother Maloyi. Mbilini's claim must obviously be treated as a piece of special pleading, but it would be equally unwise to ignore the reasons that must have existed to push Mswati in this direction.

The final years of Mswati's reign had as far-reaching repercussions internally as they did in foreign affairs. Externally, Mswati used his freedom from challenge to restructure relations with neighbouring powers. Within Swaziland, he took advantage of the same lull to consolidate his domestic administration, without having to worry about creating a potential fifth column which might ally itself with enemies abroad. The precise nature of these changes, the circumstances which brought them about, and the way in which they impinged on Swaziland's relations with the Zulu, the Portuguese and the Shangane are the subject of the present chapter.

For Mswati, 1852 was a year of almost unrelieved disaster. Under the impact of invasion and foreign occupation, large numbers of Swazi had fled to neighbouring states, and one can only presume that this was just the visible tip of a much larger submerged group, whose loyalty wavered during the crisis. Once the Zulu armies had departed, Mswati took steps to weed out the waverers and to eradicate the conditions which had brought him so near to collapse. The strategy he adopted fell into two distinct parts, each of which had been tentatively developed even before the attack. Some time after the Zulu invasion of 1847, Mswati had evacuated his capital from Ekufiyeni in central Swaziland to Hhohho, which had meant shifting the main locus of royal power considerably further to the north, and after the occupation of 1852 Mswati greatly accelerated this process, by allocating numerous chiefdoms in the area to his brothers and his wives.

The history of the Ngwane, or at any rate, of their royal line, the Dlamini, stretches back far into the past. According to genealogies collected by James Stuart at the end of the nineteenth century, the Dlamini were able to trace their line of succession back some forty generations. Genealogies of this length are, sadly, notoriously subject to inflation, as many students of oral tradition will confirm. Eponyms, toponyms and patronyms are often included, as are the names of regents and contenders for the succession. Even outright fabrication can sometimes occur, when societies are suddenly faced with the need to create a remote past which sanctions the present. No reliable time depth can therefore be drawn from the king lists of the Dlamini, beyond saying that they seem to have lived in the vicinity of Delagoa Bay for some considerable time.

Ngwane history only dawns in the mid-eighteenth century. (The term ‘Ngwane’ is used here to designate the nucleus of what was later to become the Swazi, who took control of southern Swaziland (Eshiselweni) in the 1760s and 1770s. I have confined its use to the Eshiselweni period prior to the Ngwane's later expansion north. Thereafter I use the term ‘Swazi’ since there is evidence of its having been used from this time.) From this period genealogies become more detailed, and for the first time one finds agreement about the correct sequence of kings. Two reasons probably account for the change. The Dlamini were experiencing a period of crisis, precipitated by the efforts of the neighbouring Tembe to expand their monopoly of a burgeoning maritime trade, and were rent in the process by a series of dynastic feuds.

The year 1874 ushered in a new period of uncertainty for the Swazi. With Ludvonga dead, a struggle for the succession ensued, which concluded with Mbandzeni being installed as a virtual puppet of the queen mother and the regency council. The next eight years, until Mbandzeni disposed of the queen mother and assumed the full perquisites of power, were a period of decentralisation of royal authority as the regiments were partially demobilised and as local leaders reclaimed some of their lost powers. In the wider regional context this was also a period of marking time. At first sight no clear pattern can be discerned. Fresh permutations of the old political order following hard on the heels of one another, in a bewildering kaleidoscope of political change. Renewed pressure from the Zulu in the mid-1870s, and a new Swazi defeat at the hands of the Pedi, were followed by the Republic's collapse in the 1876 Pedi War, and by Britain's annexation of the Transvaal. Yet beneath this surface swirl of events deeper currents were running which were reshaping the regional balance of power. The repercussions of the mineral discoveries of the late 1860s and early 1870s pulsed their way through the whole of the region, while demographic pressure was building up in Zululand, Pediland and the Trans-Vaal. The squeeze on Swaziland increased accordingly, and was applied in a far more systematic fashion, although the quarters from which it emanated constantly changed. The deepening contradictions of the period were at least partly responsible for Britain's annexation of Transvaal, which heralded the birth of a new era, in which the Zulu, the Pedi and the Swazi would all succumb successively.

Two themes dominate the 1880s in Swaziland, and one ceases to have the same central significance. With Cetshwayo's defeat by the British in 1879, the Swazi were able, for the first time in living memory, to enjoy the luxury of disengagement from Zulu affairs. Although rumours circulated in the first half of the 1880s about Swazi support for one or other of the contending factions in the fractured Zulu state, and although the civil war there occasionally spilled over into the Pongola valley or the southern Lebombo, the Swazi kept themselves largely aloof from the conflict, and Zululand faded from the forefront of Swazi leaders' minds. However, if Zululand lost its central significance, two other problems came much more to the fore. The first was Mbandzeni's quest for personal authority; the second, the ‘paper conquest’ of his country through concessions, and the diplomatic complications to which this gave rise. It is with these that this chapter will be concerned.

For much of this period the two questions were closely interwoven, but as the Transvaalers wrested their independence from the British in the summer of 1881, it was the first which occupied the centre of the stage. Mbandzeni, it will be recalled, began his reign from a position of exceptional weakness. He had been chosen as king less for his exceptional qualities than for his exceptional lack of them, his lack of mother included, and the best that could be said of him was that he was 'a quiet puff-adder', itself a notably ambiguous metaphor.

As the sense of crisis gradually lifted on Mswati's southern border, Swazi attention reverted to the north. Here Somcuba remained a nagging irritation, and continued to sour Mswati's relations with the Lydenburg Boers. The sudden dwindling of diplomatic intercourse between the two communities after the Commission of 1851 gives some idea of the level to which relations had sunk. Reference is not even found in surviving documents to the unprecedentedly disruptive invasion of Swaziland by Zulu forces in the last six months of 1852, and even ordinary trading enterprises slowly ground to a halt after Mswati neglected to pay for the goods he had previously received.

The problem of Somcuba was not simply one of his presence in the Republic: his removal of the Ludlambedlu cattle, and his performance of the iNcwala ceremony meant that he was directly usurping Mswati's political and ritual power. Nor, indeed, had he been content to live peaceably under Lydenburg's protection, but had subjected Mswati's people to a variety of harassments, including the murder of Swazi messengers sent to parley with the Boers. Somehow or other his depredations had to be stopped. To begin with, Mswati made at least one attempt in 1853 to tackle the problem by negotiation, but not surprisingly these efforts quickly broke down. In the final analysis Mswati could only be satisfied with Somcuba's death or his delivery into Swazi hands, and this was something to which Lydenburg could plainly not agree.

Despite a relatively long tradition of historical scholarship, the historiography of nineteenth-century South Africa is uneven in quality and in places disappointingly sparse. Within the realm of specifically African history, large areas await serious academic attention, and while a new upsurge of interest has been evident for much of the last decade, only a small proportion of the resulting researches has yet appeared in print. Much of the published African material is therefore still framed within settler, Afrikaner or liberal traditions, and is disfigured in many instances by a strong albocentric slant. The dominant assumptions have been of the backwardness and stasis of African societies, to which is often added their incapacity to shape history for themselves. Liberal writing has been no more immune from this spirit than studies cast in the settler–Afrikaner mould. Settler historians like Theal or Cory may have inveighed against the barbarity of blacks, and framed their accounts in terms of African aggression and deceit, but they did at least devote considerable space to African activities from which the outlines of a history emerge. Liberal historians by contrast have contented themselves with affirming the dignity and validity of African societies, and denouncing the violence and rapacity of whites, but almost invariably with the assumption that Africans were passive objects of history meriting little attention in themselves. Only with the publication of The Oxford History of South Africa in 1969–71 was a more Africanist dimension injected into liberal writing, yet this, while an important milestone in southern African studies, has not been backed up by the range of monographs that it might have been expected to inspire.