|
home | franco frescura | architecture | urban issues | lectures | graphic work | postal history | historical archive
THE ZULUS OF THE HIGHVELD - The Role of Architectural Form in the Establishment of a Competing "Zulu" Political Identity among the Matabele 1822-1897Franco Frescura INTRODUCTIONDuring the early years of the nineteenth century the Khumalo were a Nguni-speaking clan inhabiting the northern reaches of present-day Zululand. At that stage they fell under the overlordship of the Ndwandwe who were then competing with the Zulu to inherit control over Dingiswayo's former Mtwetwa confederacy. Tradition has it that shortly before a crucial military engagement between the two groups, the young Khumalo chieftain, Mzilikazi, broke with the Ndwandwe and fled with his people to Shaka. The Zulu King rewarded them by elevating Mzilikazi to a leadership role, giving them status as allies, and permitting them to fight together in his army as one regiment. Following the Zulu's victorious war against the Ndwandwe, the Khumalo were permitted to return to their ancestral lands. In about 1821 the Khumalo were entrusted by Shaka with the task of raiding cattle from a neighbouring Sotho chieftain. Conventional wisdom has it that, after the raid, Mzilikazi refused to part with the King's share of the booty, and thus treacherously broke with Shaka, who was both his friend and overlord. In order to escape the King's anger Mzilikazi, together with some 300 warriors migrated northwards into the Transvaal. The Khumalo slowly moved across the Transvaal highveld settling at first on the upper reaches of the Oliphants River, in 1823, then near modern-day Pretoria, in 1825, and finally on the Marico River in the western Transvaal, in 1832. On the way they came into conflict with numerous highveld groups, defeating and incorporating them under a larger Ndebele (Nguni) identity. Ultimately at its height this kingdom encompassed an area from the Limpopo to the Vaal, and from the Crocodile to the Marico, although their military expeditionary forces are known to have ranged much further afield. At the end of 1836 the Ndebele were defeated by a well-armed Dutch commando led by Potgieter and Uys, in a running battle which lasted nine days. At its end Mzilikazi and his followers, both Nguni and Sotho/Tswana, migrated northwards to present-day western Zimbabwe where they carved out a new kingdom for themselves, this time at the expense of the Shona. Although outwardly plausible, the popularly held tale of the schism between Shaka and Mzilikazi suffers from some notable inconsistencies. Why, for example, did an intelligent and capable leader like Mzilikazi jeopardise his people's lives and ancestral lands by alienating a powerful and provenly ruthless foe over a mere herd of cattle? Why did the Zulu king fail to exact retribution early on, when the Khumalo were still residing in northern Zululand and numbered no more than a few hundred warriors? And why did the Zulu kingdom send numerous expeditionary forces after the Khumalo, almost on a yearly basis, even after Shaka's death and long after Mzilikazi had long removed his followers beyond the Limpopo? THE DEVELOPMENT OF A MATABELE "ZULU" IDENTITYAlthough the Khumalo clan was initially small, they were also armed and trained in the new Zulu manner. They were thus able to easily defeat and absorb a number of smaller clans who barred their march northwards. They quickly earned for themselves a reputation as courageous and skilled fighters, as well as the name Ndebele, or Matabele, as they were to call themselves later on. This was probably drawn from the Tswana term tebele, which translates as "plunderer", although some informants have also given "stranger" and "wanderer" as possible alternative meanings. By the time they had settled in Matabeleland, the western region of present-day Zimbabwe, in 1837, they numbered an estimated 5000 warriors. During this brief fifteen year period they moved their capital four times, migrating from a timber-rich Lowveld region to the more arid western Transvaal and finally to the Mopane veld north of the Limpopo. Throughout this time and right up to 1897, seventy five years later, they continued to build their dwellings in the form of a hemispherical dome, using the same technology, domestic structures and settlement patterns as they had done previously in Zululand. When they were visited by Smith in 1835 in the Marico valley, thirteen years after their migration from Zululand, he noted that their dwellings were "exactly in keeping with that (plan) adopted by the Amazoolas eastward of Port Natal". (1975) The descriptions and sketches he gives of Matabele dwellings and their construction differs little from contemporary accounts of Natal Nguni architecture. These patterns were continued after the Matabele moved to western Zimbabwe. In about September 1872, Selous visited Lobengula, at Gubulawayo, where he noted that:
It was only once their political ascendancy over the Shona was broken in two successive wars with the British in 1893 and 1896, that they appear to have abandoned their attachment to the hemispherical form in favour of the cone on cylinder dwelling indigenous to their new home region. This historical process is significant for a number of reasons. Firstly, by the time the Matabele had reached the Marico River valley of the western Transvaal in 1832, they had ceased to be predominantly Nguni. It is true that they had been joined from time to time by other Nguni refugees from Zululand, but these groups were generally small and the bulk of their initial numerical expansion was owed to the various Sotho and Tswana tribes they encountered and subjugated in the Transvaal. The position of these recruits was varied. Some enjoyed the status of full allies; others were conquered and their women taken as concubines whilst their young men were eventually incorporated into Matabele regiments as full warriors. Thus a large proportion of the Matabele population in the Transvaal was not of Nguni origin, but rather, came from backgrounds whose domestic architecture did include the knowledge and occasional use of hemispherical domestic structures but was predominantly orientated towards the cone on cylinder. It is surprising therefore that this dwelling form should have persisted among the Matabele for almost four generations after they had abandoned their original homes in northern Zululand. It could, of course, be argued that the hemispherical grass dwelling form was not unknown to the Sotho and that some dome-building Sotho groups are known to have existed. Thus it would have been a simple matter for these newly-incorporated "Matabele" to have made a small adjustment to Nguni building technology. This assumption, however, ignores the practical factors of building process, environment and availability of materials. Secondly, as previously stated, Matabele society underwent four major relocations in the brief space of fifteen years. This is in itself remarkable, for upon each occasion they must have been under considerable stress as the result of pressures brought about by Zulu, Sotho/Tswana and Dutch belligerence towards them. Yet, throughout this time, they tenaciously managed to retain not only their original Nguni identity, but also the knowledge and technology of their old Nguni dwelling forms. This is made even more remarkable when we consider the findings of modern researchers such as Desmond, who have revealed the extent of the trauma suffered under Apartheid by rural communities as the result of a single such move in their lifetimes. Thirdly, the Matabele's fifteen-year migration took them through a number of distinct environmental and climatic zones. Originating in 1822 from a grass-rich area of northern Zululand, in 1823 they halted in the Bushveld of the upper Oliphants River, bordering upon the timber-rich Lowveld; 1825 saw them in a mixed bushveld/grassveld region near present-day Pretoria; in 1832 they settled in the semi-arid bushveld of the Marico valley; and in 1837 they finally made their homes in the mopane veld of western Zimbabwe. It is obvious therefore that the survival of their grass-based technology, particularly from 1832 onwards, must have relied greatly upon the continued importation of building resources from other regions further south and east. In 1831 Collins described the construction of their towns as "a large kraal - or rather city ... a collection of closely built houses two miles in length". The construction of such a settlement must have involved vast amounts of materials, some of which are not available either on the Transvaal highveld, or in the mopane veld north of the Limpopo. Thus a large proportion of the thatching grasses needed would, of necessity, have been ported over some considerable distances. Such an achievement must have required a large amount of determination on the part of the Matabele, as well as an inordinate attachment to their architectural tradition. It would also have necessitated the deployment of vast amounts of human resources. Fortunately for the Matabele this does not appear to have been an insurmountable obstacle, and it seems probable that much of this work was made feasible by the use of local indentured labour. Arbousset and Daumas (1846) related how the Matabele used Pedi captives to conduct major works:
It becomes clear therefore that the Matabele's retention of an architectural style, which lasted for nearly seventy-five years, runs counter to all the dictates of practicality and common sense and cannot be explained in simple environmental and social terms. The solution, if one is indeed possible, will have to be found elsewhere. The clue to one possible answer lies in the fact that the Matabele initially referred to themselves as the abakwaZulu, "the people from Zululand", or the amaKumalo, "the people of Kumalo". Early visitors to the court of Mzilikazi invariably referred to them as being "Zulus" and it was only some time later, certainly after 1835, that they formally adopted the term "Ndebele" in reference to themselves. This was distorted to the Shona usage of "Matabele" much later on, once they had settled north of the Limpopo. The Zulu had only come to the fore in their own region from about 1818 onwards and Shaka's kingdom can only be considered to have emerged in its fullest sense after 1819. The "people of Kumalo" therefore could not have been part of a larger Zulu identity for longer than three or four years, hardly enough time, one might think, for them to identify with the political ideal it entailed, let alone to want to usurp it later on. We do know, however, that the original schism in the Zulu state occurred as the result of a personal clash between Mzilikazi and Shaka, and that the military struggle between Zulu and Matabele continued on an almost yearly basis after Shaka's death in 1828 and beyond the Matabele's final northward migration in 1837. The enmity between the two groups during the lifetime of Shaka can be understood, Mzilikazi had, after all, been one of Shaka's generals. He had defied the authority of his King and must therefore be punished for it. However for the feud to have continued well into Dingane's reign and after the Matabele had removed themselves beyond the immediate Zulu sphere of influence, seems to suggest something more than an internecine war to the death. It might be hypothesised therefore, that the quarrel between the two leaders had been part of a carefully orchestrated plan by Mzilikazi. The refusal to hand over the captured cattle was purely symbolic, and the Khumalo migration out of Zululand was carefully planned beforehand, with the full co-operation and approval of the clan's members. Once on the highveld, between 1822 and 1836, the Khumalo deliberately sought to create for themselves an identity based upon the Zulu model. Their warriors dressed and fought like the Zulu, Zulu was their predominant language, and their style of architecture, both in settlement form and dwelling type, was identifiably Zulu. Dress, of course, could be copied; military tactics learnt and emulated; Zulu is a dialect of Nguni, a language common to the entire eastern littoral; but architecture, particularly settlement pattern, is recognized by rural people to be a reflection of their entire cosmology, their economic system, social structure, inheritance hierarchy and religious practices and is thus their most manifest banner of group identity. It could be for this reason, therefore, that the hemispherical dwelling form was retained so tenaciously for nearly four generations, seemingly against powerful odds. Even when the dome was replaced by the cone on cylinder dwelling, the principles of domestic settlement remained in common practice right up to the Chimurenga of the 1970s. The reasons why a Zulu identity should have been so carefully cultivated by the Khumalo are a little more difficult to determine. At a primary level it could be read to have been no more than a simple psychological device which exploited the fearsome reputation which the Zulu regiments had gained for themselves under the leadership of Shaka. The expectation therefore might have been that their opponents in battle would become paralysed with fear, the morale of their own troops would be boosted, and the people themselves, coming as they did from a number of different backgrounds, would have a central, powerful and well established identity to rally round. While these may be reasonable assumptions to make, they also fail to explain one vital historical factor: why was the Zulu leadership seemingly so concerned with the presence of a second "Zulu" state on the highveld? The answer here must lie in the concept of a political identity. It will be remembered that the original Zulu clan was a small group whose number in 1816 can hardly have exceeded two thousand persons. By 1819 a wider Zulu identity encompassed most Nguni along the coastal belt from the Pongola in the north to the Tukela in the south. Under these conditions the term "Zulu" ceased to have connotations of clan and became identified with the political ideal of a larger Nguni state. Thus when Mzilikazi fell out with Shaka, ostensibly over the question of how to divide the spoils of a military victory, what was at issue in reality was not the future of a herd of cattle but the political leadership of this state. The establishment of an alternative "Zulu" presence in the southern African interior was then not simply the work of a petty chieftain who had emigrated and was making good elsewhere, but it was perceived to stand as a challenge to the very identify of a "Zulu" corporate state. It is not known whether the differences between Shaka and Mzilikazi were based on a simple leadership struggle or whether they were of a more fundamental and ideological nature. We do know however that Shaka's Zulu kingdom was predominantly pan-Nguni and hence of a limited and regional nature. Mzilikazi's domain on the other hand encompassed in its time Nguni, Sotho/Tswana and Shona alike. It is possible to postulate therefore that, of the two men, it was Mzilikazi who was endowed with the wider political vision. It is not known whether this encompassed the establishment of a Black southern African state, but his own power base was established on a broader regional front than Shaka's and must therefore be perceived as a forerunner to such an ideal. If this interpretation is indeed correct, then it might also explain why the hemispherical dwelling form, so symbolic of their own group identity, was abandoned after 1896 when White settlers, armed with Maxim machine guns, crushed Matabele political ascendancy in their region. POSTSCRIPTThis paper was originally presented at a conference on Ethnicity, Society and Conflict in Natal, hosted by the Department of Historical Studies, University of Natal, Pietermaritzburg, 14-16 September 1992, under the title of “The Zulus of the Highveld”. To the best of my knowledge it has never been published.
Copyright @ francofrescura.co.za
|