|
home | franco frescura | architecture | urban issues | lectures | graphic work | postal history | historical archive
STRUCTURE, SETTLEMENT AND SOCIETY - A Southern African Case StudyFranco Frescura ABSTRACTThis paper is written from a post-structuralist, or Derridian, standpoint (Sturrock 1979) which focuses upon the indigenous rural built environment of southern Africa and interprets it as a series of "texts". These are constituted by the background of the observer, by previous research on the subject, by intrusions on the observer's consciousness, by such factors as oral history, and by the "text" as written by its "context". The latter makes specific reference to the concept of intertextuality which will cause the same object or dwelling placed in different contexts to give different textual readings. Contingent to this argument is the acceptance of architecture as a built environment constituted by a series of material and spatial elements, each of whom is endowed with a particular contextual symbology, and whose relationships are governed by a larger set of rules of grammar. As such then architecture may be considered to be a cognitive language (Hall 1973), and its elements, textures, details and spaces may be "read" and interpreted as a series of texts, each having meaning in their own contexts (Alexander 1977). In these terms rural architecture or, in this particular instance, rural settlement, is viewed as a text whose reading yields much information in the way of the economic activities, social mores, material environment and technological culture of its creators. It is analysed not only as a given, but it is also deconstructed in order to take into account underlying gaps and silences. This does not constitute, as it were, a reading of what is between the lines, but rather of what has been left unstated, often unconsciously. In the text which follows, the historical settlement patterns of indigenous southern Africans have been divided into two major classifications. This has been the result of a number of disparate factors covering a wide range of concerns, including inheritance patterns, gender relations and religious beliefs. Although both groups may be seen to share in a number of common features, their contextual reading reveals these to be manifestations of more fundamental structural differences between the two societies. Ultimately, it is believed, that these are representative of wider differences in their respective cosmological value structures. INTRODUCTIONThe western concept of a "traditional" architecture for southern Africa or, indeed, for Africa as a whole, has been beset historically by a preoccupation with the aesthetics of the individual dwelling. Authors such as Walton (1956) and Denyer (1978), as well as Frescura (1981) have all published studies which focus upon the technologies and forms of the dwelling unit but generally fail to emphasise their role in the larger settlement structure. In some instances this work has been misinterpreted by other agencies seeking justification for the creation of "ethnic" stereotypes, thus reinforcing the western idea of house-form as the primary manifestation of architecture as a cultural activity. Sadly this preoccupation with form is misplaced. The individual dwelling seldom stands alone and is usually but one of a number of interlinked residential spaces, both indoor and outdoor. These are not set randomly in the rural countryside, but are organised according to a number of socially predetermined norms. A study of the southern African rural homestead therefore will reveal that its settlement patterns are subject to considerations of social structure, inheritance hierarchy, burial rituals, religious beliefs and economic practices. Thus any questions of architecture as a material manifestation of culture and identity has to be resolved in the wider context of the region's settlement customs and not according to ephemeral and transient values of aesthetics. SOME BRIEF DEFINITIONSThe human perception of such concepts as house, homestead, and village tends to vary as one moves from culture to culture, from people to people and from place to place. Whilst such terms may outwardly appear to find application in a number of disparate backgrounds, generalisations of this nature usually fail to stand up to closer scrutiny as further considerations of kinship structures, spatial organization and social hierarchies are brought into the argument. It is seldom, for example, that the individual dwelling unit may be termed a "house" in the Western sense of the word. It may comprise the total house in some singular instances where it serves to shelter a hermit or a bachelor, both rare in rural society, but generally it is but one unit within a larger domestic complex. The term "settlement" is also ambiguous, and capable of finding a number of different interpretations. In this case it is used as a generic word to describe human habitat in its widest possible context, embracing in its scope the full scale of population numbers, ranging in size from the individual family through to the larger village and town. Within such a broad spectrum therefore the individual dwelling unit may he defined as a single structure, usually one roomed, whose form covers the full range of building types available to the indigenous resident. The "domestic unit" may he said to occur in those groups where each wife in a polygamous family is allotted her own living space comprising of one or more dwelling units set within their own system of courtyards (Illustration 1). The "homestead" may be defined as that area occupied by a nuclear family, either monogamous or polygamous, under the leadership of one person, usually the husband and/or father. Such a homestead consists of a number of dwelling units which may be grouped into separate domestic units but which will usually involve the disposition of the family's constituent parts according to socially predetermined patterns which recognise and formalise the status of individuals within that family (Illustration 2). Although considerations of "homestead" normally also involve elements of locality, these need not be conclusive, as in cases where the proximity of the constituent domestic units of a homestead was either made impossible by development of the settlement form itself, or was made undesirable by social custom. ARCHITECTURE AS A LANGUAGEA criticism which has been levelled against architecture from time to time by linguistic purists has been the former's appropriation of the term "vernacular". This is a word generally associated with language, but more currently it has also been applied to certain genres of domestic and pre-industrial buildings. The reasons for this are simple. Not only do architects themselves use the concept of a "language of architecture" (Viollet-le-Duc 1859), but some close parallels may also be drawn between these two disciplines. Both are possessed of a basic vocabulary of expression; both are subject to certain set rules of grammar; both may be considered to be systems of symbols and codifications; and both may be classified as having "high" and "vernacular" branches. The similarities do not cease there. Language, or more generally "communication", is widely accepted by social scientists as an important, socialising factor in the induction of new members into a community (Piaget 1962). Other unspoken, and even unconscious, value systems exist as part of the socialising process: role playing for example or some social taboos such as incest. In the case of architecture it may be seen that society attaches values to the organisation of living spaces, to the forms and even to the textures they are given and that these tend to vary as one moves from society to society (Levi-Strauss 1968). Thus there would appear to be some justification for the claim that architecture provides an unspoken and generally unconscious language for the classification of man's built environment. This means that the disorientation experienced by new visitors to a country may be ascribed to both their differing perception of spatial systems and of building forms as well as their unfamiliarity with local languages and social customs. It is possible to assume therefore that, over the years, the expression of architecture has undergone a series of changes where current value systems and symbologies are linked to those of previous eras and societies. The idioms of Victorian architecture, for example, are not so far removed from those of current society that a modern person would not be able to assign values to its spaces and forms with a reasonable degree of accuracy. On the other hand, the specialised knowledge of a historian or an archaeologist might he required to decipher the spatial distributions of a Hellenic dwelling or a Roman courtyard house. It has similarly been found that various rural groups in southern Africa also developed their own value systems of spatial organisation. Venda fortified villages, for example, were last used some six or seven generations ago and are believed to have been largely the product of colonial land wars which beset the region during the nineteenth century. Current Venda settlement, on the other hand, has been moulded by a different set of social, economic and environmental pressures. Yet current research has shown that present-day Vendas are able to read the barest signs of settlement by their ancestors and give them spatial significance. As was the case with the Victorian villa, the idioms of nineteenth century Venda settlement are not so far removed from those used currently as to require specialised historical knowledge in order to interpret them. Today visitors approaching a rural domestic unit or homestead will often be informed about the residents within before a single word has been exchanged. Provided they are also familiar with the "language" used, they will be able to determine their cultural background, how many wives are part of the extended family, the status of each within the family hierarchy, whether their husband is still living, the wealth and status of the family, the location of the kitchen, living and sleeping areas, the method of approach visitors should follow and even their etiquette. In some regions of southern Africa it is also possible to ascertain whether the husband is chief or commoner, the location of his mother's dwelling and the spaces reserved for cooking, for entertainment, for work, for meeting and for judgment and mediation (Frescura 1985). The outside spaces of settlements, although not always clearly defined in architectural terms, are subject to a high degree of specialisation. In the first place there exists a measure of interchangeability between inside and outside space with seasonality, weather and individual preference acting as the major determinant factors. Secondly, outside spaces tend to he defined by a highly developed set of visual signals and territorial statements such as low walls, kerbs, shelving, decorative textures and visual lines of connection (Illustration 3). Finally the question of family privacy, which does not really arise in the case of interior spaces, is established by custom and regulated by architectural forms and visual messages. DEFINING THE GRAMMAR OF RURAL SETTLEMENTA number of major themes run like threads through the fabric of southern Africa's rural settlements. These involve the relationships between the individual elements of settlement, and between them and the settlement as a whole. Rural society appears to be highly ordered in this respect and many interpersonal relationships which are internalised in the traditions and culture of a people are often mirrored in the organisation of their habitat to the extent that the individual's status within their society or peer group will predetermine their location within the overall settlement structure. In general terms it may be said that the interpretation of space in southern African settlement is subject to four major considerations:
SOCIAL HIERARCHIES AS GENERATORS OF SETTLEMENT FORMAlthough theoretically it should be possible to claim that each different region in southern Africa is possessed of a settlement form which is uniquely its own, in reality it will be seen that generally only two themes predominate: those where social hierarchies are resolved in terms of "left" and "right", and those where these are resolved in terms of "front" and "back" distributions. THE CONCEPT OF "LEFT" AND "RIGHT"The concept of "left" and "right" has been recorded as existing in both Sotho/Tswana and Nguni societies and is based upon the hierarchical relationship perceived to exist between the "Head" and subsequent wives of a polygamous marriage. The interpretation of which hand is assumed to be ascendant varies from group to group and is the subject of numerous historical anecdotes or myths. Generally speaking it may be said that the belief that "right" is superior to "left" is grounded in the rural metaphor of a warrior wielding his spear with his right hand and his shield with his left (Illustration 4). The right is therefore assumed to have an ascendance over the left being "active" and "aggressive", whilst the left is "passive" and "defensive". On the other hand where the converse is held to be true and "left" is held to be ascendant over "right", this is explained by a person placing their right hand over their head and stating that the right hand serves or protects the "head". In such cases it was also found that wives were not placed in "Right Hand" and "Left Hand" Houses but were either in the "First House" or the "Right Hand House" and at no time was the right hand described as being inferior to the left. This social hierarchy finds reflection in the form of the settlement in that the "First" or "Head" wife will be located at the head of the settlement, in a central position and on axis with the entry to the homestead. Second and subsequent wives are then located alternatively on either side of the first wife in descending order of importance. Such settlements are normally circular and axial in nature which not only emphasizes the importance of the first wife, but also stresses the fact that her dwelling, the Great Hut, is also the formal abode of her husband, the head of the homestead. The other wives will each have their own dwellings as well as, in some instances, their own separate cooking facilities or even their own separate households. In such cases conjugal rights are meant to be enjoyed in strict rotation with the husband visiting each wife in turn. This stands in sharp contrast to Venda social custom where the husband enjoys his own dwelling quite independent of any wives he might have. A Venda wife will take up abode in her husband's quarters until such a time as she falls pregnant. At approximately the sixth month of term she is allowed to return to her own dwelling. Venda male informants showed some reluctance at the prospect of co-habiting with more than one wife simultaneously, something that their Nguni and Sotho, Tswana counterparts thought little of, and at least one expressed the fear that, should two of his wives fall pregnant at the same time, he would be summoned to appear before his chief, be accused of behaving "like a dog" and fined a cow for his pains. Strictly speaking therefore, in Nguni, Sotho and Tswana society, the husband does not have a dwelling in his own right, and although his domicile is held to be that of the first wife, in reality even that belongs to, and is the responsibility of, the woman concerned. This becomes an important factor in our later comparison with those settlements where hierarchy is expressed in terms of "front" and "back". Although considerations of "left" and "right" are basic to both Sotho/Tswana and Nguni settlement the form of the settlement itself can however vary greatly while still permitting individual elements to find their socially predetermined positions within the whole. The Sotho/Tswana of the northern Cape and the western and the northern Transvaal are known to have built their settlement in the shape of an open fan with an open space, the khotla, being located in its centre. The first wife of the senior man was sited at the head of the homestead on the central axis of the settlement directly opposite the main entry to the central space, with subsequent junior wives being located alternatively to her left and her right. The homesteads of his brothers or other members of his retinue would then also he located alternatively to the left and right of his abode according to their descending order of status in relation to him. They, in their turn, would also follow a left-right hierarchy in the distribution of their wives within their own individual residences (Illustration 5). Today the circular fan pattern has fallen largely into disuse, although individual homesteads tend to follow a linear form along lines of land contour, the same considerations of left and right have prevailed and are still being maintained by succeeding generations. Sotho-like fan shaped settlements have also been recorded among the South Ndebele of the Transvaal where the direct influence of their Sotho/Tswana neighbours has been clearly evident. Today few of these survive and the South Ndebele, like their Sotho neighbours, now appear to build mostly linear settlements which follow the lines of land contour. Like them, they also have retained their "left-right" preoccupation with hierarchy. An interesting elaboration of the South Ndebele distribution of "left" and "right" was recorded in the 1950s by Meiring, at the Msiza homestead on the farm Hartbeesfontein, north of Pretoria, a site more popularly known in its time as "Speelman's Kraal" (Meiring 1955) (Illustration 6). Here, once their sons had achieved adulthood and married, they took up positions in close proximity to their mother's domestic unit and in doing so, also established hierarchical positions among themselves based upon the traditional left-right relationship. Current research among other South Ndebele homesteads (Frescura 1981) has shown that, whilst such a practice is still being largely observed, the economics of land management and the requirements of space are today beginning to have an adverse effect upon their architectural patterns. Although many settlements are still being planned out along historical lines, they are also beginning to make small but significant departures from the layouts of their parent homesteads. Research conducted nearly 30 years later among the self-same Msiza group showed that in 1952, once they had relocated to their new home at KwaMatabeleng, their initial settlement had maintained the old left-right distribution. However, once male children had begun to settle down with families of their own, then this pattern was disrupted by the location of the planting lands and the economic need to keep these intact from human settlement (Illustration 7). Among many Nguni groups the traditional circular pattern of settlement may still be found to a large extent, even in some cases of smaller family groups. During the course of current research, the concept of "left" and "right" was also found to apply in the context of the individual dwelling unit, where differentiation was made between the man's and the woman's side of the internal habitat. The rules governing such a distribution were however found to vary from group to group, and in fact did not necessarily coincide with their respective application of "left" and "right" hierarchies in the larger context of the settlement. The notable exception to this rule was found to be the Venda among whom no such distinction existed within the individual dwelling unit or, for that matter, in the homestead (Illustration 8). THE CONCEPT OF "FRONT" AND "BACK"While it is generally true that the concept of "left" and "right" as a determinant of hierarchy within a settlement was found to be fairly consistent in its application among southern Africa's indigenous groups, the same thing cannot be claimed of the concept of "front" and "back". Although an awareness of the latter was perceived to exist in the settlements of Nguni, Sotho/Tswana and Venda alike, their individual interpretations differed so radically as to make it impossible to establish any generalised rules of application. In the case of the Nguni, both northern and southern groups practice a social structure which inhibits marriage within their immediate kinship circle. Unlike the Sotho/Tswana where the preferred marriage is between first cousins on the mother's brother's side, the Nguni generally marry outside their kinship circle. Thus we find that the traditional Nguni settlement has a built-in schism line between the older generation (at the back) who will continue their residence in the parent homestead and the young (in the front) who will eventually hive off elsewhere and start separate homesteads in their own right (Illustration 9). In the case of the Sotho/Tswana fan-shaped settlement a hierarchical differentiation may be said to exist between those homesteads which are closest to the central common, that is to say the residences of the parents, and those homesteads located on the external perimeter being the residences of the children (Illustration 5). Although strictly speaking, such a differentiation can best be described as one existing between the "centre" and the "perimeter", this is only valid for as long as the settlement form remains circular. Once the settlement breaks up into individual homesteads which tend to follow a linear pattern along the contour lines, a trend in Sotho/Tswana settlement which has accelerated in more recent times, then it will be seen that the traditional centre-perimeter apposition will become translated into one based upon the concept of front and back. In the context of the individual traditional Sotho/Tswana domestic unit, the opposition has always been one between "front" and "back", where the dwelling of the parents is located in the fore court or lobe of the homestead whilst the areas of privacy, of cooking and of children's residence, have been located to the rear. This is a pattern of settlement which archaeologists have termed as "bilobial" (Maggs 1972) (Illustration 10). In both Nguni and Sotho/Tswana cases however, it may be successfully argued that what front-back appositions that do arise are in reality no more than form variants resulting from the circularity of the larger settlement. In both cases also, front-back appositions only arise between parents and siblings and in no example was this found to be the case between husband and wife, or wives. The instance of the Venda can be said to be unique in the context of southern African settlement for whilst all other groups use a system of hierarchical differentiation based upon their concept and individual interpretation of "left" and "right", in their case all distinction is made in terms of "front" and "back" in establishing the relative status of husband against wife, father against son, chief against people. Such a differentiation becomes evident at the level of a humble domestic unit where the dwellings and activity centers of the wife, or wives, will be located before and preferably downhill from those of the father and family head. Similarly the homesteads of a man's sons will be placed in front of and downhill from the parent homestead whilst in themselves repeating the pattern set by the father's. The settlement of a Venda chief or headman will tend to follow the same basic pattern as that applicable to his people, excepting that the process of ascent to his quarters will be complicated by the introduction of defensive ramparts and narrow access passages, all designed to protect his person in the case of attack. An old Shona proverb, also found to be in use in Venda, states that "to climb a mountain one must go zig-zag" (Plangger 1974), an obvious allegorical reference to their chiefs and, indirectly, to the pyramidal structure of their settlement patterns. This is equally applicable to commoners at the level of a small nuclear family. Thus, in many ways the humblest of Venda domestic units acts as a microcosm of their society as a whole (Illustration 11). Venda settlement is also differentiated from that of their neighbours in southern Africa by the fact that, in a polygamous marriage, the Venda father retains the use of a dwelling unit in his own right, with its own courtyards and, in some cases, its own granary. This contrasts sharply with the practice of most other southern African groups where, under similar circumstances, the dwelling of the first wife is considered to be that of the father as well although, strictly speaking, he has no dwelling of his own and is expected to rotate his residency equally among those of his wives. On the other hand, the head of a Venda homestead will be visited by his wives at his dwelling. This also means that, with the exception of the Head wife, no hierarchy of settlement is followed in determining the position of the domestic units of second and subsequent wives, their location being a matter of personal preference and negotiation with the head of the homestead. The Head or First Wife will usually be identified by the fact that the main path of access to the homestead leads directly to her domestic unit. OF DIFFERENCES ... AND SIMILARITIESMuch has been said over the years about the alleged "ethnic" composition of the indigenous population of southern Africa. In this respect, white historians and anthropologists have laid particular stress upon the many supposed differences, material, cultural and linguistic, which they perceived to exist among the various regional groups. Such differences have, over the years and possibly with the abettance of academics, been internalised into many of the misconceptions which white society tends to labour under to the present day. Yet if greater recognition were to be given to the similarities which link cultures (Levi-Strauss 1968), then it is probable that what divergences do exist will be perceived to be no more than regional variants, much as those existing between Cornishman and Yorkshireman, Englishmen both. Certainly were architecture and language of settlement the only criteria available, then it would be seen that if "cultural" divisions do indeed exist, then they are not along the traditionally supported lines between the Nguni and the Sotho/Tswana, but rather along those between Venda society on the one hand and that of the Nguni, the Sotho and the Tswana on the other (Illustrations 9 and 12). To a certain extent this viewpoint has been supported by other researchers in the field (Kuper 1980), some of whom have gone so far as to suggest that the Venda represent a southern and ancient extrusion of the Shona people. The tabulation of such differences in settlement and spatial perception certainly makes a strong case for the existence of one greater and homogeneous body of indigenous South Africans. SOME BASIC DIFFERENCES OF SETTLEMENTThe differences between Venda and Nguni/Sotho/Tswana settlement or, conversely, the similarities linking Nguni, Sotho and Tswana, would appear to hinge upon the following major issues:
SOME SIMILARITIES OF SETTLEMENTIn spite of these differences, Nguni/Sotho/Tswana and Venda groupings share in some strong similarities of settlement.
APPROACH AND ORIENTATION OF ASPECTAlthough much of the discussion has, thus far, centred upon the distinctions inherent in the appositions between "left and right" and "front and back", none of these terms should be seen as representing absolute values. In fact both are as subject to regional interpretation as are potentially more subjective terms as "house" or "homestead". The Venda, for example, will read "front and back" as terms of hierarchy whereas the Sotho/Tswana will understand it as a transition of spaces from the public common to the private courtyard. The question of "left" and "right" is an even more complex one, especially when one considers the number of variants to which it may be subject. In the first place there is no general agreement among the various southern African groups as to which side is dominant, although the metaphor upon which it is based appears to be more or less universally understood. Secondly, the relative positions of "left" and "right" are also subjective ones and depend largely upon whether the viewer is looking into a settlement or out of it. The question is further complicated by the fact that some groups, such as the Pondo, would appear to apply one set of criteria in their settlement patterns and another diametrically opposite set in the case of their individual dwelling units. If we were to carry the analysis further and, as in the case of Adam Kuper (1980), introduce other considerations such as the orientation of the cattle byre, the orientation of aspect, the resident's stated perception of left and right, and the actual relationships existing between the households of the different wives, then the number of permutations achievable becomes such as to make any comparative study almost impossible. Generally speaking however, it might seem that many of the variations which do exist, do not necessarily reflect fundamental differences in social attitudes towards the individual elements of a homestead or settlement, so much as in the actual relationships existing between the elements themselves. As such, differences may well have come about as a result of regional environmental conditions, changing social practices or even deliberate political statements. The case of the Zulu is an apt illustration of this point. The concept of a "Zulu" identity did not exist until the early part of the nineteenth century when Shaka, son of Senzangakona, took up the work of Nguni political consolidation begun by Dingiswayo. In about 1816 he was given the leadership of his father's people, a small Nguni clan which went under the grandiose name of "the children of Heaven", the amaZulu. Using military conquest as a means of sustaining political growth (Morris 1966), he rapidly engulfed the region, incorporating its numerous clans under his chieftainship using the name "Zulu" in the sense of a wider Nguni regional identity. Recent archaeological research has revealed that the region's pre-Shakan Nguni settlements tended to orientate the entrance to their central cattle byre uphill and facing onto the Great Hut. This is a feature common to most other Nguni groups and should be seen to have strong pragmatic and practical roots. One informant in the Transkei more recently explained that "that is (the direction of) the hut of the father and he has to know what is going on". This has remained the general practice of the Southern Nguni of the Transkei/Ciskei region, the one recorded exception amongst them being the Bomvana, an immigrant group from Natal, who describe themselves as "building like the Zulus". However it has been found that post-Shakan settlements have reversed this order, a condition which has been maintained to the present day. The reasons for this are self evident. Nguni culture views the byre as an area of cattle enclosure as well as male gathering and family ceremonial. When however the Zulu king took the basic Zulu settlement form and translated it directly into the larger royal village, its function changed, becoming a military barrack. The function of the byre also changed, becoming an area of military parade and state ceremonial with only one small portion being retained, perhaps symbolically, to house the king's prize cattle. Under such circumstances, the decision to re-orientate the byre by 180ø becomes understandable for both practical and symbolic reasons. The new byre gate, having now been faced downhill and away from the Great Hut, or in this case, the Royal quarters, facilitated the access of troops into the parade ground, gave the king's residence greater privacy and accentuated the central axis of the settlement thereby increasing the dominant siting of the Royal enclosure, an important factor in the ceremonial aspects of occasions of state (Illustration 13). It is not known whether this new settlement order spread to the homesteads of commoners because the people intuitively wished to emulate their leadership, because demobilised Zulu soldiers deliberately chose to implement a settlement form they had come to associate with a Zulu political identity, or perhaps because Shaka himself codified it. In both instances however, the question must arise of just how far one had to travel from the core of "Zulu" culture or authority before this kind of influence began to be cancelled out by other similar factors. There certainly seems to be a strong element of group identity involved, with people being able to identify this element of their architectural culture as being "Zulu-like", as in the instance of the Bomvana of the Transkei and the Tsonga of Mocambique. If this was indeed the case, then we should also ask to what extent did the conquered Nguni clans revert back to their pre-Shakan settlement culture once the Zulu kingdom's authority was reduced by the British in the war in 1879? Current research seems to indicate that, once established, Zulu builders have opted to retain their settlement form and post-Shakan spatial values to a great degree in spite of some notable changes in their social and economic patterns of life. The importance of the Zulu's re-orientation of their byre entrance should also be seen in the context of a centralisation of political power in the region. Nguni settlement, by the nature of its society and its marriage norms, tended to be of a scattered nature (De Jager 1964). The orientation of their byre entrance towards the Great Hut may therefore be seen as a function of controlling cattle and hence wealth. However the rise of larger Zulu settlements, acting as Royal residences and military garrisons, created symbols of a centralised administration and perhaps even the capitals of a nation. This means that the pastoral function fell by the wayside and was replaced by a ceremonial one. Parallel developments may be seen to have taken place at an earlier date among Pedi and Tswana societies whose settlements, perhaps for slightly different reasons, reached a high degree of centralisation during the eighteenth century and probably earlier. Both these groups orientate the entry point to their central activity space, only part of which functions as a byre, downhill towards the entrance and away from the homestead of the settlement head. CONVENTIONS OF PERCEPTION AND ASPECTThe twin questions of orientation and aspect are fundamental to settlement form for they deal directly with the resident's own perception of dwelling and homestead space and are usually governed by wider social conventions. Western society, for example, reads its maps with north turned to the top of the page; similarly architects usually place plans of their buildings along an accepted north-south orientation, although when matters of approach or facade are discussed the norm may then be deviated from for the sake of effect or expediency. An interesting insight in this respect was gained during the course of field work among the Ndundza Ndebele group at KwaMatabeleng. The village elder, Maselwane Msiza, was shown Meiring's plans of the original Msiza settlement at Hartbeesfontein which had been vacated by them nearly thirty years previously. Being an architect, Meiring had drawn his plans with the most dramatic view in mind, showing the courtyard walls facing towards the reader (Illustration 6). In the process he had not only inverted the normal north- south axis but, more importantly in this case, the viewer was looking into the settlement from the cattle byre outside. Both Maselwane Msiza and the other men gathered about failed to recognise the plans until one of them, with a flash of inspiration, suddenly turned them "upside down" so that he was now viewing the settlement from the inside looking out. Thereafter the group was not only able to give an account of the buildings but also a detailed breakdown of their original residents. Similar events were experienced in the course of field research in both Venda and Botswana where informants were able to interpret settlements previously recorded by archaeologists and known to be part of the informant's own historical background (Illustration 14). CONCLUSIONSThis paper has not concerned itself so much with the technical or aesthetic manifestations of southern African rural architecture as with establishing the primary principles, or rules of language, which govern its existence. As such it has concentrated upon the reading of the spatial texts and subtexts, and how these have been assigned differing values in the context of various southern African regions and societies. As a result, I have come to the conclusion that the division perceived by previous researchers to exist between Nguni and Sotho/Tswana does not exist in architectural terms and that both groups can be said to share in the same settlement culture. In fact, if any clear divisions can be said to have emerged in the settlement patterns of the region, these have not been along the preconceived lines between Nguni and Sotho/Tswana, but rather between the Nguni, Sotho and Tswana on the one hand and the Venda on the other. POSTSCRIPTThis paper has undergone a number of working drafts, and was eventually presented at the IASTA Annual Conference on First World-Third World: Duality and Coincidence in Traditional Dwellings and Settlements, Second International IASTE Conference, Berkeley, California, 4-7 October 1990, under the title Structure, Settlement and Society. It subsequently appeared under the same title in the Traditional Dwellings and Settlements Working Paper Series, IASTA, University of California, Berkeley, Vol 19, 1991. 53-84. BIBLIOGRAPHYALEXANDER, Christopher. 1977. A Pattern Language. New York: Oxford University Press. LIST OF ILLUSTRATIONS
Copyright @ francofrescura.co.za
|