COURTING DESIRE?: LOVE AND INTIMACY IN LATE 19TH AND EARLY 20TH CENTURY KWAZULU-NATAL [1]

Mark Hunter, Department of Geography, University of California at Berkeley

In 1878, John Colenso, the first Bishop of the British Colony of Natal, published an updated version of the isiZulu dictionary that he had first authored in 1861. To help define the word uthando (love) he used the phrase uHulemente ute izintombi azitshaye ngotando, zingabotshelwa emadodeni, which he translated as "the Government says that girls should choose through love, and not be compelled to husbands." [2] Implied within this definition is that the British settlers brought love — defined here as the ability to choose a marital partner — to young isiZulu speakers, especially women. This view was consistent with the passing in 1869 of a law on African customary marriage which made compulsory the presence of an "official witness" to protect women from "forced marriages." Indeed, many settlers expressed outrage at forced marriages, polygamy, and the apparent trading of women (through ilobolo); a moral position that rather neatly coincided with interventions to push "idle" African men into the labour market.

Of course, the colonial voice was by no means unitary: Bishop Colenso himself rebelled against many elements of mainstream settler thinking, particularly the view that Africans should not be allowed to convert to Christianity unless they withdrew from polygamous marriages. [3] Even so, the impression that African society was loveless gained currency throughout the 20th century including within the discipline of social anthropology. The lens of structural-functionalism in particular could position marriage as a mechanical exchange of women between kinship groups and, according to the most famous anthropologist of this ilk Radcliffe- Brown (1962: 46): "The African does not think of marriage as a union based on romantic love ..." Affection, he argued, "is the product of the marriage itself ... " [4]

Recognition of love was also squeezed out by prominent views on African sexuality. Vaughan (1991: 129) argues that African sexuality tended to be seen as either "'primitive', uncontrolled, and excessive" or as "innocent" but susceptible to "degeneration" through the social and economic changes of colonialism. [5] These tropes maintain a presence, albeit a shifting one, from the 19th century to research on Aids today — no more so than in Zulu society long associated with all-that-is-African. Consistent with the view of African sexuality as uncontrolled, the Caldwells (1989) root the African AIDS pandemic in a "distinct and internally coherent African system of sexuality" which they assert "is vulnerable to attack by all coital-related disorders" (Caldwell et. al, 1989: 187). Sex they argue takes place outside of a moral frame and is seen as a "worldly activity like work or eating and drinking" (1989: 222). But if this standpoint led the Caldwells to suggest that sex exchanges were longstanding, ideas of African sexuality as "degenerating" drove another perspective on the materiality of sex: that money and sex was becoming progressively linked as sexuality degenerated though the forces of industrialization, modernity, and apartheid.

These are relevant themes today. My PhD thesis will argue that the "materiality of everyday sex" — the widespread link between gifts and sex in everyday relationships — is, at its present scale, in fact a late twentieth century phenomenon fuelled by rising (gendered) inequalities, continued agrarian decline, chronic unemployment, and a virtual ending of the project of "building a home", at the centre of which is marriage. But in approaching the "materiality of everyday sex" I argue that we must resist the familiar dichotomy made between gifts and love and see sex-gift exchanges as having being formed through, and not simply in opposition to, shifts in intimacy over time. Even the most material courting relations today are not simply commodified transactions but can extend for long periods and be positioned within a narrative of romantic love. Condoms, the most reliable protection against HIV, can often be used in the least, not the most, commodified relations, ones that are situated within a love culture. [6] Yet, on the whole, the public health industry addresses the Aids pandemic with a conceptualisation of African relationships as being instrumental and loveless.

An examination of the history of intimacy among isiZulu speakers can also be turned towards a further paradox within the contemporary Aids literature: how very poor women can often engage in relationships with multiple partners for gifts but talk about these through the modern language of "rights" and opportunity. [7] This realisation of modern "rights" is invariably stated by informants (and sometimes repeated by researchers) to have stemmed from the advent of democracy in 1994. But there is a deeper history of political and social identity formation that requires exploration and relevant to this is how romantic love itself, a powerful modernist narrative, might have contributed to the emergence of new gendered subjectivities. For several decades now scholars have been noting the importance of romantic love to the production of modern Western subjects and this literature has become less Western centred in recent times. [8] The Aids pandemic in sub-Saharan Africa, however, while greatly increased research on sexual violence, dominant masculinities, and sometimes women's instrumental response to these, has largely bypassed questions of intimacy and love. The assertion made here is that exploring African relationships from the perspective of transformations in intimacy can open a space through which to move beyond common representations of African women as being either simply victims of men or as modern subjects instrumentally exerting new "rights" brought by democracy.

This article directs its attention to intimacy in KwaZulu-Natal during roughly the 50 years up to the 1930s. In the 1930s, the anthropological project of recording African society "as it was" emerged alongside attempts to track moral and sexual degeneration that resulted from "cultural contact"; indeed, Krige's (1936a) structural functionalist account of Zulu society The Social Systems of the Zulu was published in the same year as her pioneering work on social breakdown in a Pretoria township (Krige 1936b). Yet rejecting a framework whereby traditions were either historically static or in linear degeneration, I argue that at the turn of the century courting traditions were being simultaneously created and undermined. This is seen for instance in the way that traditional institutions supposedly in decline — such as the amaqhikiza, elder girls who controlled the sexuality of women — may well have been to some extent recent "inventions" themselves. Very important to this rather uncertain path of change is the ambiguous role of Christianity which could grant women an ability to challenge gendered relations in certain spheres while contributing to attempts to control young women in others; as evidence of the latter, I note how in the first decades of the 20th century women's ability to have multiple-sexual-partners appears to have been eroded. The approach adopted here, therefore, stresses multiple intersecting dynamics above linear processes and the making as well as the erosion of tradition. This framework helps us to think through both continuities and discontinuities in relationship patterns over the late 19th and early 20th century.

The chapter draws from court records, early dictionaries, oral testimonies, ethnographies, and accounts from missionaries and government officials. An important question hanging over this chapter, and one long familiar to historians and anthropologists, is how to engage with sources that purport to represent isiZulu speakers, in the subtitle of Bryant's (1949) famous work, "as they were before the white man came". In this example, Bryant's book was actually completed in 1935 and based on his experiences in Natal and Zululand from 1883 to the 1930s. It is far from clear when and where exactly he collected the numerous insights he used to make claims about pre-colonial Zulu society. Similarly, Monica Wilson chose Pondoland for her classic study Reaction to Conquest partly because it was an area with relatively few missionaries and with weak connections to urban areas, and yet the material was collected in the 1930s. Testimonies in the James Stuart Archive, a collection of 200 oral testimonies by colonial official James Stuart at the beginning of the century, can also be vague as to the period to which they refer. [9] In some instances I have used such sources, even when they were collected at a later date, to support claims about pre-colonial and early colonial society. At the same time, I have tried to point out examples, such as the amaqhikiza discussed in Krige's (1936a) The Social System of the Zulu, where there seems to be evidence that traditions seen as longstanding may well have been relatively new. The chapter begins by discussing courting practices in the 19th century; continues by exploring tensions of change emerging from colonialism, migrant labour, and Christianity; and ends by critiquing narratives of linear "degeneration" taking place either through a "commodification of sex" or the erosion of tradition.

Courting freedom in the late 19th century

In 1887 Gumakwake, a young man from the District of Klip River in Northern Natal, was hauled before an all white jury at the criminal court in Ladysmith for murdering a young woman, Timbaiya. He was sentenced to be hanged after only 20 minutes deliberation by the jury. [10] The records, which totalled 94 pages, included testimonies from 12 witnesses for the prosecution and 2 for the defendant and the case was presided over by Justice Cragg. Timbaiya was, according to the witness, Umhlebi, "a full grown girl, she reached puberty ... She was quite marriageable. She was a good girl, not quarrelsome." Gumakwake, had not "done anything wrong before this, I do not know that he is a quarrelsome man."

The murder occurred near Job's Kop in Northern Natal on a moonlit October night in 1887 in the wake of two social gatherings. The first gathering, a "dancing party" according to Umhlebi, was held at the Kraal of Ndomba. Many people walking home from this gathering stopped off at Umhlebi's kraal to drink beer including the murdered young lady, Timbaiye, and the accused, Gumakwake. After most guests had left, Gumakwake was said by several witnesses to have persuaded Timbaiye to accompany him telling her that she was being called by her former boyfriend, Nkosana. According to the testimony by Timbaiye's friend, Ncwebu, Timbaiye had broken off the relationship with Nkosana some months previously and had "chosen" another lover while Nkosana had also done the same. Having lured Timbaiya away from Ndomba's kraal, Gumakwake is then said to have brutally murdered her, the district surgeon reporting that parts of the face were "smashed to a pulp".

In order to understand the background to the case witnesses were asked to explain prominent courting practices. Ncwebu noted that it was common for women to move quickly between different lovers: "It is usual for a girl to throw up a lover, and to take another lover — when a girl has "chosen" a man, he and she have external sexual intercourse with each other for a short time and then they throw each other and take others: this external sexual intercourse is "ukuhlobonga" and it takes place at night — some men go themselves to call the girl, some send for her." That young men and women engaged in intimate liaisons at dances is supported by missionary and Zulu linguist A T Bryant (1949: 567) who said: "There was, however, a match-making agency among the Zulus much more effective even than their sisters. That was the dance ..."

Further examples of affection and desire are present in this court case as well as in other records from the time. In his testimony to the court Nkosana uses the metaphor of being driven by the "heart" to describe his passion for his girlfriend and this is a telling one — inhliziyo (heart) according to Colenso's 1878 isiZulu dictionary also meant the "seat of emotion". Indeed, the very language of courting also supported the existence of passion and desire. The isiZulu verb ukushela describes a man who proposes love to a woman and was recorded as literally meaning "burning with desire for" and the Zulu verb ukuqoma, which describes a woman's acceptance, also means simply "to choose". [11]

Unlike in many societies of the time, the relative courting freedom appears to have extended to women, at least in the late 19th century. The young lady, Timbaiya, who was said by the District Surgeon to have been about 14, was able to move with her friends from one gathering to another with relative ease and was apparently free to go and spend the night with her former lover Nkosana — a degree of freedom that would have startled Victorian women of this era. It seems that this relative candour also allowed women to secure more than one concurrent sexual partner at certain times. One of James Stuart's informants, Ndukwana, noted in 1900 how "It was a common thing for a woman to have three lovers (sokas) each belonging to a different regiment" so long as she soma'd with only one per month so that pregnancy could be accounted for. [12] In her classic account of Pondoland, an area she chose for its lack of "cultural contact", Monica Wilson (1936: 182) further says that unmarried women could engage in ukumetsha (ukuhlobonga/ukusoma) relations with more than one partner: "The more skulls the better." For sure, courting was enveloped in strict gendered expectations; as Bryant (1949: 564) noted "it was the height of impropriety, indeed positively lewd, for a female to make overtures to a man." At the same time, in many important respects young women had a certain freedom that was arguably denied to them by the 1930s. This provides support for the claim that it was young women's fertility and not sexuality that faced the greatest control, although in practice the distinction is a difficult one to make (see Guy, 1987). [13]

Yet it appears that the courting freedom of young men and women varied considerably over the course of the 19th century. There is some evidence for instance that restrictions on courting were tightened at the height of Shaka's rule over the Zulu Kingdom in the first quarter of the 19th century. [14] At this time, young unmarried men were collected in amabutho (age-set regiments) and not allowed to marry until they completed what was effectively military service at the age of about 40. Although not physically brought together, women were also placed in amabutho from which they would be juba'd (allowed to marry) to a particular male amabutho. But examples of rebellion against the amabutho system suggest that even during this period young people were willing to take momentous risks in order to be with their own lovers. The most well known illustration of this point is the case of the inGcugce age-set of girls who in the 1870s flouted King Cetshwayo's order to marry an older regiment and ran away with their sweethearts, only for many of the young rebels to be captured and brutally killed. [15] Demonstrating the fact that some choice existed despite the centralised state King Cetshwayo himself testified in the late 19th century that forced marriages were not common: "In a few cases the girls are forced, but not as a rule", he said. [16]

But a second basis for courting restrictions was the strength of kinship obligations, a point stressed by Radcliffe-Brown, quoted earlier. As numerous civil court cases from the early 20th century demonstrate, strong social bonds and obligations could stretch within and across kinship groups; for instance through the practice of ethula cattle for ilobolo (bridewealth) could be loaned and then paid back when the first born girl of the union was herself lobola'd. [17] Well-to-do families could be required to furnish daughters to the royal isigodlo in order to cement social alliances, as demonstrated by the case of Paulina Dlamini. [18] Moreover, early missionaries in KwaZulu-Natal recorded that many of the first women who ran to missions did so to escape forced marriages (see Etherington, 1978). All of these examples do support the view that marriage was far from an individual act.

Divergent sexual moralities and understandings of love

The relative courting freedom of young people can be drawn upon to suggest that sex was outside of a moral realm in pre-colonial Zulu society. As Delius and Glaser (2003) note, even in the case of marital relations in southern Africa the existence of extra-marital affairs predates migrant labour, thus suggesting that marriage was not solely based around sexual exclusiveness. Addressing these points, Bryant (1949: 568) argues that sex among isiZulu speakers was not a moral but a social principle, a distinction that was quite prominent in early ethnographies:

But arguably Bryant, a missionary like many early recorders of African tradition, approached morality too readily from the perspective of Christian ideals of morality. If we are to recognize that "social principles" could also be moral mores, Bryant's division must be seen as rather artificial and actually between two different types of "morality". It is clear for instance that in some circumstances a woman's chaste demeanour and sexual conservatism was highly prized; sexual modesty was, after all, central to the powerful custom of ukuhlonipha (literally 'to respect'). Ukuhlonipha was defined by Dohnë in 1857 as "to be bashful, to be shy, to keep at a distance through timidity, to shun approach". These practices also included women avoiding the use of words similar to the names of their husband's relatives. [19] A "loose" woman, one without inhlonipho, was labelled as isifebe, or worse isihobo (deflowered woman), enormous insults that could deem a woman as unmarriageable and place shame on her age-group. [20] These discourses certainly bore down heavily on women even if they did not stimulate the same individualistic Christian notions of guilt which placed women's bodies and consciousnesses in a direct relationship with God, embodied in the phrase "The Body is the Temple of God". As a consequence, in many respects profound and deep-seated morals were attached to sex in African society. [21]

If sex could take place within a moral realm, we must question whether it could be easily exchanged in pre-colonial/early colonial African society. There is, at first glance, evidence supporting the tradability of sex from South African ethnographies. Monica Wilson's writings on Mpondoland in the 1930s suggested that a group of women called amadikazi, though not explicitly trading sex, could benefit materially from sexual relations, including with married men. Wilson (1936a: 208) described this group, whom she believes predated colonialism, as the "artists of the community". Such evidence certainly supports the view that there was no absolute morality around the exchange of sex as there was within a Christian framework. But while it may well have been the case that in some circumstances sex and giftswere traded within an accepted "sphere of exchange" we should not take such findings too far. [22] The amadikazi women were divorced, widowed, or single mothers and therefore by definition seen as largely ineligible for marriage as first wives, generally the most prized outcome for a young woman. It was arguably this fact that placed them outside the moral requirements not to exchange sex; it is likely that a young unmarried woman would have been seen as isifebe (a loose woman) were she to be caught trading sex, a view supported by Vilakazi's (1962) later ethnography. At the same time, the presence of the amadikazi in Pondoland reminds us to give due attention to geographical differences. It is noticeable that a group of amadikazi women did not seem to exist among isiZulu speakers. Concubinage and adultery appear to have been heavily frowned on in KwaZulu-Natal in the early colonial period. [23]

The fact that pre-Christian and Christian notions of sex appear to draw from different conceptions of morality prompts us to question whether there was also a single understanding of love. Though in the late 19th century young men and women had relative freedom to choose partners we should not confuse this with the choice enshrined in modernist narratives of "romantic love". Accounts at the beginning of the 20th century suggest that women could use izintando, or love potions, to attract a man and men could also use umuthi (medicine/potion) to hayiza a woman and induce her to love him. [24] These love potions might have had their origins in human desires but they worked through the spiritual world of amadlozi (ancestors). Consequently, unlike the individualistic love of the romantic love culture, they could not be so easily resisted through self control, a point well demonstrated by the involuntary screaming of a woman who was subjected to umhayizo. At the same time, similarities between the powerful consequences of love potions and the spontaneous emotion associated with romantic love should not be overlooked: both could "come from nowhere" and seize the heart of a man or a woman in a powerful manner.

Tensions of change: Migrant labour, Christianization, and the "invention" of women's traditions

The Gumakwake case, as stated, cannot be taken in itself to exemplify a pre-colonial Zulu sexuality. The mention in this case of blankets and candles suggests that certain forms of Western material culture were well embedded among isiZulu speakers in this region. Geographical differences, especially between Natal and the later colonised Zululand, are also important but difficult to explore. This section continues with the two themes already discussed — love and the link between money and sex — and explores the possible consequences of migrant labour, customary law, and Christianity for courting.

Ben Carton's (2000) Blood From Your Children draws attention to the escalation of generational conflict in Natal at the turn of the century as young men embraced migrant labour. Some of these tensions can be seen in the following murder case heard by Justice Wragg in Verulam in 1885. [25] Unkala was a young man engaged to be married to Ndikiza, the ilobolo having been paid in full with the exception of one beast. The murder took place a week after Unkala had returned from a stint of work in Durban. The dispute seems to arise following Unkala's wish to marry the 18 year old Ndikiza quicker than her father was prepared to allow. In a rage, Unkala was said to have jumped onto a path to stab Ndikiza with an assegai in full view of eight witnesses. Although it is of course impossible to know for sure the reasons behind the murder — Unkala said that he was drunk and knew nothing about it although all the witnesses testify to his soberness —- most observers appeared to agree that Unkala's action resulted from his anger at his prospective father in laws' insistence that he follow customary time periods around marriage. Unkala's impatience may have been propelled by his own experience of work, especially his subsequent control over a powerful unit of universal exchange, money, and his familiarity with "industrial time". [26] Undoubtedly, migrant labour threw relationships into a precarious new set of geographical and material linkages.

Yet just as significant as migrant labour to courtship were the powerful changes unleashed by Christianization. The province of Natal in the 19th century was among the most Christianised region in Africa, although pre-conquest Zululand was less fertile recruiting ground; the Norwegian Lutheran churches were the most successful denomination in Zululand but they had still failed to attract more than 300 recruits in the 1870s after 3 decades of efforts (Etherington, 1997). Early converts to Christianity tended to come in "dribs and drabs rather than waves" and were often refugees looked down as outcasts (Etherington, 1997: 97). Nevertheless, the rise of Christianity in the 20th century was remarkable. African people claiming to be Christian rose from 26% in 1911 to 76 % in 1990 (census figures quoted in Elphick, 1997).

One early change ushered in by Christianization, migrant labour, and the erosion of agricultural capacity in rural areas, centred on the value of fertility. There appeared to have been a relatively rapid decline in the practice whereby a "barren" woman could have the ilobolo paid for her returned or her responsibility for "raising seed" transferred to a sister. [27] If this suggests that values around fertility, at the very centre of isiZulu speakers society, were being transformed, one of Christianity's most important interventions was to quite unequivocally associate pre-marital and extra-marital sex with sin. Missionaries looked down at the lewdness of practices such as hlobonga/ukusoma and the singing of extraordinarily explicit songs at puberty ceremonies (as described much later in Krige, 1968). Yet in other important respects Christianity was not so radical a break with gendered roles in African society. Missionaries validated the authority of the male led umuzi even if they tried to introduce an entrepreneurial spirit and an ideal of marriage as based more on companionship. According to Meintjes (1990) in her study of mission station in Edendale in the mid 19th century:

The church, however, was a powerful transformative force not only through its direct interventions into sexual norms. Christianity, as the Comaroffs (1991) have argued, worked as much through a modification of everyday practices from clothing, to cleanliness, to agricultural techniques, to literacy, as through the promotion of new ways of thinking. Hence, the very idea of what constituted modernity and tradition were being reshaped more widely by the penetration of Christianity into African society. The first amakholwa tended to convert to Christianity by moving to a mission station but this pattern was increasingly replaced by Africans attending more locally situated churches, including independent African churches. Moreover, Western practices such as clothing became progressively adopted even among non-Christians and therefore more ambiguous in their meaning. Demonstrating an increased ambiguity around Christianity one meaning of the term amagxagxa was "in-betweens", those who adopted Western styles but did not attend church. [28] At times, however, this fluidity could be crystallised in new binaries constructed between the categories of amakholwa (Christians) and amabhinca (traditionalists). According to Dohnë's 1857 dictionary, the verb ukubhinca meant to "cover obscenities ... to bind any kind of cloth or dress around the hips". But by the 1940s the noun ibhinca had appeared in dictionaries and clearly signalled backwardness, or a "Raw Native, one who still wears native costume". The term amaqaba, according to dictionaries and oral testimonies, had also come to meant an ignorant, uncultured person, as well as a heathen. [29]

The demeanour of women, including the value around chasteness, was also reworked through a blending of Christianity and "tradition". Writing about KwaZulu-Natal in the 1920s, Shula Marks (1989) has drawn attention to how the growing mobility of women fuelled alliances between chiefs, administrators, and Christians centred on women's "purity". White missionaries, Zulu nationalists, African Christians, and the Department of Native Affairs all railed against the disintegration of "tribal discipline" evidenced by the increasing "immorality" of woman in urban and rural areas. According to Marks (89: 225): "It was in the position of African women that the forces of conservatism found a natural focus." An important character embodying this trend, and noted by Marks, was Sibusiwe Makanya, a prominent Christian and teacher who founded the Purity League, an organization that sought to preserve the morality of young Zulu women.

Quite radical shift in courting emerged not only from this neo-traditionalism. Two further possible related dynamics were migrant men's greater bargaining power as they increasingly earned their own ilobolo and the codification of customary law that effectively legislated against women's but not men's infidelity and placed women as "perpetual minors" unable to own property. [30] One dramatic change at the beginning of the 20th century appears to have been the reduction in women's rights to have multiple partners. In contrast with court cases and testimonies from the 19th century, the early isiZulu ethnographies were adamant that only men could have more than one partner. [31] Moreover, the 50 or so male and female informants I spoke on this subject with in KwaZulu-Natal who grew up from the mid-1930s to the mid-century looked on at surprise at the suggestion that women might have been able to have more than one hlobonga/soma partner (see Hunter, 2004). This seems to be graphic evidence for a quite radical shift in the early parts of the 20th century towards the greater control of unmarried women.

This escalation of constraints on women challenges the view that modernity simply brought love to women, if choice over courting is taken as one definition of love. At the same time, it is very clear that women were quick to draw upon Christian notions of modernity to counter customs that denied them the freedom to choose a husband for instance ukungena (the 'entering' of a deceased husband's brother into marital arrangements with the widow). Court records and oral testimonies suggest that women vigorously opposed this "tradition" and that it rapidly declined in the early decades of the 20th century. [32] There is also evidence that husbands' double standards surrounding extra-marital affairs were being challenged by more educated, Christian, or urban women. [33] Men were caught in something of a bind. The rising status of Christianity, propelled by its association with education and work in a declining agrarian setting, meant that some men aspired to be amakholwa (believers/Christians). Yet implicit in the acceptance of God was a commitment to monogamy and more equal forms of companion marriages.

As noted, a variety of studies on courting practices have noted how education, the greater economic independence of young people, and the modern technology of writing, fuelled a more individualistic notions of romantic love. As Lipset (2004) recently pointed out, however, there is no necessary association between modernity and romance and we need to be wary of teleological narratives; this article has already pointed to the existence of love in pre-colonial Zulu society. At the same time, there is strong evidence of new forms of love becoming intertwined with older expressions in quite radical ways at the beginning of the 20th century. During this period, Christianity and schooling were not yet widespread in South Africa but letters still provided an important form of communication across migrancy routes and between lovers. Tellingly, Isaac Shapera's (1940) classic Married Life in an African Tribe written in the 1930s opens with a love letter discussing relationships where greater stress is placed on penetrative sex. "I still think of how we loved each other; I think of how you behaved to me, my wife; I did not lack anything that belonged to you." (39). He saw these letters as an indication of quite sweeping changes in Kgatla society including towards choice in marriage.

Reviewing letters sent from migrant workers in South Africa at the beginning of the 20th century, Keith Breckenridge (2000) has argued that they were "intrinsically indirect and collaborative" (338) and yet "helped to constitute new kinds of individualised subjects" (348). Breckenridge suggests that the development of a private sphere, including through love letters, could foster new political identities, a point worth underlining when accounts today present "rights" as simply ushered in by democratisation in 1994. Yet, as Breckenridge notes, these were far from individual exercises. Based on research in Kenya, Lynn Thomas (forthcoming) draws attention to how love letters were often written in full knowledge that they might be read by others, for instance as evidence in courts, and therefore constituted what she calls a "collective privacy". At the same time, because literacy was a profound symbol of a "modern" persona rather than simply a private act they also constituted a "personal publicity". Such categories not only blur boundaries between the private and the public but also help to ground love letters in the divergent geographies in which they were written and read, for instance schools or migrant men's mining communities.

In the example of the letter below, written in 1917, the records offer very few details about the case from which it was sourced. The only background to the case is a short typed document summarising the "facts found proven". [34] From this document, it appears that the case was brought by the brother of the lady for whom the accused, Matoleni, had paid 4 cattle. The crucial question under consideration by the magistrates court was whether Matoleni had abducted his fiancé. In the letter, one of two used as evidence, Matoleni is making a plea for his pregnant girlfriend to come to Pietermaritzburg to meet with him. At that time, he was working as a Warder in the Pietermaritzburg gaol. A striking feature of the letter are the phrases "Kiss, kiss, kiss, kiss, kiss, kiss" and "My dear" towards the end. Even if this letter may have been a collective exercise shared with friends and ultimately the courts, and even if we have no idea whether Matoleni and his girlfriend might also have used love potions as well as the more modern written forms of persuasion, the letter does raise questions about how new forms of gendered subjectivities and intimate practices may have been stimulated by Christianity and literacy.

Social "degeneration" and the Tradability of Sex

Along with a view that Christianity and modernity brought positive changes to African courting practices came an acknowledgment of the negative effects of modernisation. Central to this was how sex was becoming progressively commodified due to the corrosive effect of industrialization and urbanization. The figure of the African women prostitute in urban areas epitomized this view of social breakdown and had particular resonance among prominent men in both colonial and African society. [35] There is no doubt that the introduction of money and migrant labour propelled the ease with which sex could be exchanged and, indeed, it is often claimed that "prostitution", the impersonal exchange of gifts for sex, was unknown in pre-colonial African society. [36] White men were at the forefront of the sex trade enticing African women through gifts or money in urban and rural areas. [37] The persistent presence of money and gifts in rape cases also suggests that the cash economy gave men, often young men, new powers, including over women. In these cases, men defended themselves by arguing that the people who they had allegedly raped had accepted money in exchange for sex; the fact that this defence could be made suggests that such exchanges could occur. [38]

Nevertheless, it is important to challenge the idea that money simply dissolved social relationships and that sex became easily commodified. This point will be made through an examination of the giving of gifts from a young man to his girlfriend's father or chief. Around the turn of the century colonial officials expressed horror at such payments which went by a number of names, including umnyobo. To some colonial officials, these seemed to suggest that men and chiefs were trading the sexual (soma, thigh sex) services of their daughters — something of an extension of the common view that ilobolo represented simply the sale of women. [39] But how can we understand these gifts? Through demanding such payments, fathers and chiefs were undoubtedly trying to cash in on young men's earnings, especially since the government had restricted ilobolo payments to 10 head of cattle, and since rinderpest had devastated cattle holdings. [40] Yet umnyobo and other payments were also central to maintain courting as a public affair. By bringing fathers into the realm of courting, the umnyobo payments could make pre-marital affairs even more public and hence, increase, and not decrease, control on young people's activities, as suggested by Monica Wilson (1936a). To see such monetary payments as simply buying sexual services, as colonial officials sometimes did, is to strip them of their meaning as signs of public commitment that, in fact, could limit sexual behaviour. [41]

Further evidence questions a narrative of linear "degeneration" in African sexuality. Elderly informants I interviewed often spoke with great pride about how groups of older girls, or amaqhikiza, were the most important institution in guiding and controlling the chaste behaviour of young women; indeed, the amaqhikiza are arguably the most powerful symbol of past control over Zulu women. In The Social System of the Zulus, Eileen Krige (1936a) also discusses these young women as acting in a supervisory role. [42] Moreover, Krige's (1936b) study of relationships in urban Pretoria, noted that one reason for the propensity of pre-marital pregnancies was the breakdown of social controls over women, including from the older girls who supervised women. [43]

But despite their prevalence in oral testimonies and Krige's ethnography it is worth noting that amaqhikiza were not prominent in the 19th century court cases, in the two hundred testimonies in the James Stuart archive, a number of which contained detailed accounts of courting, and in Kohler's (1933) detailed account of courting and marriage in Natal. Dictionaries from the mid 19th century also did not mention the role of amaqhikiza as a guardian of chasteness and defined an iqhikiza as being a girl who has reached an intermediate stage between girlhood and marriage, for instance in Colenso's 1861 dictionary: "Young girl full-grown, not yet wearing the red top-knot (that signified marriageability)". [44] It appears therefore that at the beginning of the 20th century the institutions of amaqhikiza took on heightened, practical and symbolic, role as guardians of women's purity — but their partial failure to do so was interpreted as them being elements of "traditions" in decay. The existence but apparent decline of amaqhikiza, and the moneterization of umnyobo, have both typically been interpreted as evidence of social degeneration in African society. Yet they can both be seen as the opposite: as "traditions" forged through greater attempts to control women. This explanation is consistent with the argument made earlier that greater attention was given to women's "purity" at the beginning of the 20th century — materialized in the fact that women were less likely to be allowed to secure multiple partners.

Courting up to the 1930s

Let me be summarise my arguments concerning pre-colonial/early colonial society. A first point is that there is evidence of the presence of passionate choice in early Zulu society; if kin played a dominant role in determining who married, young people also had considerable autonomy. In a society where the very language expresses how women qoma'd (chose) their lovers, we must recognize that the roots of love among isiZulu speakers does not lie in the colonial government, as Colenso's dictionary definition suggests, but in courting practices with a much deeper history in the region. A second is that I see notions of both sex as always having been tradable or as been progressively commodified as problematic. For young unmarried women, sex could not, on the whole, be exchanged without being deemed immoral. A third is that Christianity appears to have placed increased control on women's "purity" but it also gave women an ability to challenge unpopular practices such as ukungena and situate marriage as a bond between two individuals. Related, "traditions" such as amaqhikiza, seen to be eroded by the impact of modernity, may well have been one of its products. A more complex picture of 19th and early 20th century sexuality, crucial to an understanding of the Aids pandemic today, must recognize the ambiguous impact of modernity on society and the simultaneous making and unmaking of sexual traditions.

Mark Hunter is currently completing his PhD in the Department of Geography,University of California at Berkeley. His research focuses on the social roots of the AIDS pandemic in South Africa through an historical-ethnography of one area, Mandeni, situated on the North Coast of KwaZulu-Natal.

References

Atkins, Keletso E. 1993. The moon is dead! Give us our money! : the cultural origins of an African work ethic, Natal, South Africa, 1843-1900. Portsmouth, NH; London: Heinemann; Currey.

Bao, Jiemin. 2005. Marital Acts. Gender, sexuality, and identity among the Chinese Thai Diaspora. Honolulu: University of Hawaii Press.

Barnes, Teresa. 1999. "We Women Worked So Hard" Gender, urbanization and social reproduction in colonial Harare, Zimbabwe, 1930-1956. Cape Town: David Philip.

Bledsoe, Caroline H. 1980. Women and marriage in Kpelle Society. Stanford: Stanford U.P.

Bohannan, Paul. 1959. The Impact of Money on an African Subsistence Economy. The Journal of Economic History 19:491-503.

Bonner, Phillip. 1990. "'Desirable or undesirable Basotho women?' Liquor, prostitution and the migration of Basotho women to the Rand, 1920-1945," in Women and Gender in southern Africa to 1945. Edited by C. Walker, 221-50. Cape Town: David Philip.

Braatvedt, H. 1927. Zulu Marriage Customs and Ceremonies. South African Journal of Science XXIV:553-65.

Breckenridge, Keith. 2000. Love Letters and Amanuenses: Beginning the Cultural History of the Working Class Private Sphere in Southern Africa, 1900-1933. Journal of Southern African Studies 26:337-48.

Bryant, A. T. 1905. A Zulu-English Dictionary with notes on pronunciation. Marianhill: Marianhill Mission Press.

Bryant, A. T. 1949. The Zulu People, as they were before the white man came. Pietermaritzburg: Shuter & Shooter.

Bujra, Janet. 1975. Women "entrepreneurs" of early Nairobi. Canadian Journal of African Studies. IX:213-34.

Caldwell, John, Pat Caldwell, and Pat Quiggin. 1989. The Social Context of AIDS in sub-Saharan Africa. Population and Development Review 15:185-234.

Carton, Benedict. 2000. Blood from your children : the colonial origins of generational conflict in South Africa. Charlottesville: University Press of Virginia.

Chanock, Martin. 1985. Law, custom and social order: the colonial experience in Malawi and Zambia. Cambridge: Cambridge University Press.

Colenso, John. 1878. Zulu-English Dictionary. New ed. Pietermaritzburg: Davis.

Colson, Elizabeth F. 1958. Marriage & the Family among the Plateau Tonga of Northern Rhodesia. Manchester: Published on behalf of the Rhodes-Livingstone Institute by Manchester University Press.

Comaroff, Jean, and John L. Comaroff. 1991. Of revelation and revolution. Chicago: University of Chicago Press.

Davis, William. 1872. A dictionary of the kaffir language: including the Xosa and Zulu dialects. London: Wesleyan Mission House.

Delius, Peter, and Clive Glaser. 2003. The myth of polygamy: A history of extra-marital and multi-partnership sex in South Africa. Paper presented to the sex and secrecy conference, 22-25 June.

Dlamini, Paulina. 1986. Paulina Dlamini: servant of two kings, compiled by H. Filter; edited and translated by S. Bourquin. Pietermaritzburg: University of Natal Press.

Dohne, J. 1857. A Zulu-Kafir dictionary etymologically explained: with copious illustrations and examples. Cape Town: Pike.

Doke, Clement M., and B. W. Vilakazi. 1948. Zulu-English dictionary. Johannesburg: Witwatersrand University Press.

Elphick, Richard. 1997. "Introduction. Christianity in South African History," in Christianity in South Africa. A Political, Social & Cultural History. Edited by R. Elphick and R. Davenport, 1-15. Cape Town: David Philip.

Etherington, Norman. 1997. "Kingdoms of this world and the next: Christian Beginnings amomng Zulu and Swazi," in Christianity in South Africa. Edited by R. Elphick and R. Davenport, 89-106. Cape Town: David Philip.

Etherington, Norman. 1978. Preachers, peasants, and politics in southeast Africa, 1835-1880 African Christian communities in Natal, Pondoland, and Zululand. London: Royal Historical Society.

Fynn, Henry Francis. 1986. The Diary of Henry Francis Fynn. Compiled from the original sources and edited by James Stuart and D. Malcolm. Pietermaritzburg: Shuter & Shooter.

Guy, Jeff. Love and Violence, Sex and Fertility, in the Zulu Kingdom. Workshop of the Project on the History of Sex in Southern Africa (POSH), University of Natal, 4 October.

Guy, Jeff. 1987. Analysing Pre-capitalist Societies in Southern Africa. Journal Of Southern African Studies 14:18-37.

Guy, Jeff. 1983. The heretic : a study of the life of John William Colenso, 1814-1883. Johannesburg; Pietermaritzburg: Ravan Press; University of Natal Press.

Hamilton, Carolyn. 1998. Terrific majesty : the power of Shaka Zulu and the limits of historical invention. Cambridge, Mass. ; London: Harvard University Press.

Hanretta, Sean. 1998. Women, Marginality and the Zulu State: Women's Institutions and Power in the early Nineteenth Century. The Journal of African History 39:389-415.

Hawkins, Sean. 2002. " "The Woman in Question": Marriage and identity in the Colonial Courts of Northern Ghana, 1907 - 1954. " in Women in African Colonial Histories. Edited by J. Allman, S. Geiger and N. Musisi, Bloomington & Indianapolis: Indiana University Press.

Heald, Suzette. 1995. The Power of Sex: Some Reflections on the Caldwells' 'African Sexuality' Thesis. Africa 65:489-505.

Hirsch, Jennifer. 2003. A courtship after marriage. Sexuality and Love in Mexican Transnational Families. Berkeley: University of California Press.

Horrell, M. 1968. The Rights of African Women: some suggested reforms Johannesburg: Institute of Race Relations.

Hunter, Mark. 2004. Masculinities, multiple-sexual-partners, and AIDS: The Making and Unmaking of Isoka in KwaZulu-Natal. Transformation 54:123-53.

Hunter, Mark. 2002. The Materiality of Everyday Sex: Thinking Beyond 'Prostitution'. African Studies 61:99-120.

Jeffreys, M. 1951. Lobolo is Child-price. African Studies 10:145-83.

Jochelson, Karen. 2001. The Colour of Disease. Syphilis and Racism in South Africa, 1880-1950. Houndmills: Palgrave.

Kohler, Mas. 1933. Marriage Customs in Southern Natal. Pretoria: Government Printers.

Krige, Eileen. 1936a. The social system of the Zulus. Pietermaritzburg, S Africa: Shuter & Shooter.

Krige, Eileen. 1936b. Changing Conditions in Marital Relations and Parental Duties Among Urbanized Natives. Africa 1:1-23.

Leclerc-Madlala, Suzaane. 2003. Transactional Sex and the Pursuit of Modernity Social Dynamics 29:213-33.

Lipset, David. 2004. Modernity without romance? American Ethnologist 31:205-24.

Longmore, Laura. 1959. The dispossessed : a study of the sex-life of Bantu women in urban areas in and around Johannesburg. —. London: J. Cape.

MacGaffey, Janet. 1988. "Evading Male Control: Women in the Second Economy in Zaire," in Patriarchy and Class: African Women in the Home and Workforce. Edited by S. Stichter and J. Parpart, 161-76. Westview Press.

Mahoney, Mike, and Julie Parle. 2003. An Ambiguous Sexual Revolution? Sexual Change and Intra-generational conflict in Colonial Natal. Paper Presented to African Studies/History Seminar Series, 6th April.

Marks, Shula. 1989. "Patriotism, Patriarchy and Purity: Natal and the Politics of Zulu Ethnic Consciousness," in The Creation of Tribalism in Southern Africa Edited by L. Vail, Berkeley: University of California Press.

Mayer, Philip. 1961. Townsmen or tribesmen; conservatism and the process of urbanization in a South African city. Cape Town: Published on behalf of the Institute of Social and Economic Research, Rhodes University by Oxford University Press.

Mbatha, M. 1960. Migrant Labour and Its Effects on Tribal and Family Life Among the Nyuswa of Botha's Hill. Masters Dissertation, Department of Bantu Studies, University of Natal.

McClendon, Thomas V. 2002. Genders and generations apart : labor tenants and customary law in segregation-era South Africa, 1920s to 1940s. Portsmouth, N.H.; Oxford: Heinemann; James Currey.

Meintjes, Sheila. 1990. "Family and gender in the Christian community of Edendale, Natal, in colonial times," in Women and gender in Southern Africa to 1945. Edited by C. Walker, 125-45. London: David Philip.

Moore, Henrietta, and Megan Vaughan. 1994. Cutting down trees : gender, nutrition, and agricultural change in the Northern Province of Zambia, 1890-1990. Portsmouth, NH; London: Heinemann; Currey.

Obbo, Christine. 1980. African women : Their struggle for economic independence. London: Zed Press.

Parle, Julie, and Fiona Scorgie. 2001. Bewitching Zulu Women: Umhayizo, Gender, and Witchcraft in Natal. Paper presented to African Studies/history seminar, University of Natal, 24th October.

Parpart, Jane. 1988. "Sexuality and Power on the Zambian Copperbelt," in Patriarchy and Class: African Women in the Home and Workforce. Edited by S. Stichter and J. Parpart, Westview Press.

Parpart, Jane, and Kathleen A. Staudt. 1988. Women and the state in Africa. Boulder, Colo.: L. Rienner Publishers.

Powdermaker, Hortense. 1962. Copper Town: changing Africa. The human situation on the Rhodesian copperbelt. New York & Evanston: Harper & Row.

Preston-Whyte, Eleanor, Christine Varga, Herman Oosthuizen, Rachel Roberts, and Frederick Blose. 2000. "Survival Sex and HIV/AIDS in an African City," in Richard G. Parker, Regina Maria Barbosa, and Peter Aggleton, eds. Framing the sexual subject, 165-90. Berkeley: University of California Press.

Radcliffe-Brown, A. R., and C. Daryll Forde. 1962. African systems of kinship and marriage. London, New York: Published for the International African Institute by the Oxford University Press.

Rebhun, Linda-Anne. 1999. The heart is unknown country : love in the changing economy of northeast Brazil. Stanford, CA: Stanford University Press.

Schapera, Isaac. 1933. Premarital Pregnancy and Native Opinion: A Note on Social Change. Africa 6:59-89.

Schapera, Isaac. 1940. Married life in an African tribe. Faber & Faber.

Schmidt, Elizabeth. 1992. Peasants, traders, and wives. Shona Women in the History of Zimbabwe, 1870-1939. Portsmouth: Heinemann.

Schuster, Ilsa M. Glazer. 1979. New women of Lusaka. Palo Alto, Calif.: Mayfield Publishing Co.

Selikow, Terry-Ann, Bheki Zulu, and Eugene Cedras. 2002. The ingagara, the regte and the cherry. HIV/AIDS and youth culture in contemporary urban townships. Agenda 53:22-32.

Shorter, Edward. 1975. The Making of the Modern family. New York: Basic Books.

Simons, Harold Jack. 1968. African women. Their legal status in South Africa. London; printed in the Netherlands: C. Hurst & Co.

Smith, Daniel. 2003. Imagining HIV/AIDS: Morality and Perceptions of Personal Risk in Nigeria. Medical Anthropology 22:343-72.

Smith, Daniel. 2004. Premarital Sex, Procreation and HIV Risk in Nigeria. Studies in Family Planning 35:223-35.

Standing, Hilary. 1992. AIDS: Conceptual and Methodological Issues in Researching Sexual Behaviour in Sub-Saharan Africa. Social Science and Medicine 34:475-83.

Stone, Lawrence. 1979. The family, sex and marriage in England 1500-1800. Harmondsworth: Penguin Books.

Swidler, Ann. 2001. Talk of love: how culture matters. Chicago: University of Chicago Press.

Thomas, Lynn M. 2003. Politics of the womb : women, reproduction, and the state in Kenya. Berkeley, Calif. ; London: University of California Press.

Thomas, Lynn M. Forthcoming. "Schoolgirl Pregnancies, Letter-Writing, and 'Modern' Persons in Late Colonial East Africa" in K. Barber, ed., Social Histories of Reading and Writing in Africa. Bloomington: Indiana University Press.

Thompson, E. P. 1967. Time, work-discipline, and industrial capitalism. Past and Present, 28.

Van Onselen, Charles. 1982. Studies in the social and economic history of the Witwatersrand, 1886-1914. Johannesburg: Ravan Press.

Vaughan, Megan. 1991. Curing their ills : colonial power and African illness. Polity Press.

Vilakazi, Absolom. 1962. Zulu transformations. A study of the dynamics of social change. Pietermaritzburg: University of Natal Press.

Webb. Colin, and John Wright. 1986, 2001. The James Stuart Archive of Recorded Evidence Relating to the History of the Zulu and Neighbouring Peoples. Volumes 1-5.

Welsh, David John. 1971. The roots of segregation: native policy in colonial Natal, 1845-1910. Cape Town ; New York: Oxford University Press

White, Luise. 1990. The comforts of home: prostitution in Colonial Nairobi. Chicago: University of Chicago Press.

Wilson [Hunter], Monica. 1933. The effects of contact with Europeans on the status of Pondo women. Africa 6:259-76

Wilson [Hunter], Monica. 1936. Reaction to conquest : effects of contact with Europeans on the Pondo of South Africa. Oxford: Oxford University Press.

Wilson, Monica Hunter, and Archie Mafeje. 1963. Langa: a study of social groups in an African township. Cape Town, New York: Oxford University Press.

NOTES

1. This article is based on a chapter of my forthcoming PhD "Building a Home: Unemployment, Intimacy and Aids in South Africa", Department of Geography, University of California at Berkeley. At the heart of the PhD is an ethnography based in one area of KwaZulu Natal, Mandeni, where I lived for a year and a half in total. The dissertation draws upon oral histories and contemporary ethnography to explore changes in relationship patterns over the last generation and to make claims about the importance of chronic unemployment to the Aids pandemic. The chapter upon which this article is based draws mainly from archival sources and takes the dissertation back to an earlier period. I would like to thank both Chandra Bhimull and Lynn Thomas for very helpful comments on a draft of this article, many of which I have been unable to adequately address at this stage. Some of the ideas in this article were first put forward at the History and African Studies seminar series in the department of History at the University of KwaZulu-Natal and I benefited enormously from giving papers at this forum, from attending its lively seminars, and from conversations in KwaZulu-Natal with historians that include Keith Breckenridge, Catherine Burns, Ben Carton, Bill Freund, Jeff Guy, Vukile Khumalo, Rob Morrell and Julie Parle. Thanks also to Vukile Khumalo and Thembeka Mngomezulu for help with translating and interpreting the love letter presented towards the end of the paper. Many of the themes in this article are exploratory and critical comments would be gratefully received, E-mail: hunterm@berkeley.edu.

2. Colenso, J. (1878) Zulu-English dictionary. New ed. Pietermaritzburg: Davis.

3. In this he was in agreement with the more flexible approach to African culture put forward by Natal's Secretary for Native Affairs, Theophilus Shepstone. Colenso famously broke with his long time ally, Shepstone, following atrocities committed by the colonial government to put down a rebellion by Chief Langalibalele and his followers in 1873. On the extraordinary career of Colenso see Guy (1983) and on colonial conflicts over marriage and ilobolo see Welsh (1971).

4. Anthropology did not hold by any means a united view on love. The classic structural-functionalist account of "the Zulu" Eileen Krige's (1936) Social System of the Zulu noted elaborate courting rituals. Moreover, Monica Wilson's (1936) Reaction To Conquest and Isaac Shapera's (1940) Married Life in An African Tribe, neither of which can be seen simply as structural-functionalist embracing the study of social change, are among the best sources on love in the early colonial period. These ethnographies are discussed further below.

5. Recent Africanist work that takes as a starting point a critique of the trope of "degeneration" and constructs alternative understandings from the perspective of colonial subjects include Moore & Vaughan (1994), Ferguson (1999) and Thomas (2003).

6. See, for instance, Smith (2004) for Nigeria and Preston-Whyte et. al. (2000) for South Africa.

7. Seen for instance in the way that women can position multiple partners as the "minister of finance, minister of travel, and minister of entertainment". See Selikow et. al. (2002), Hunter (2002) and LeClerc-Madlala (2003).

8. In the West, key texts that defined the field were Shorter (1975), Stone (1979). Important texts, not solely concerned with the West, are Smith (2003, 2004) on Nigeria, Ahearn (2001) on Nepal, Hirsch (2003) on Mexican workers in the US, Rebhun (1999) on Brazil, and Bao (2005) on the Chinese Thai diaspora in the US. For an excellent study of "love talk" in the US see Swidler (2001).

9. On Bryant as an historical source see Wright (1991). On James Stuart see Hamilton (1998).

10. Rex v Gumakwake (Durban Archives, RSC II/1/42, 85/87).

11. See, for instance, Bryant (1905).

12. The one month rule is itself suggestive of the interwoven nature of fertility and sexuality. Testimony of Ndukwana in Stuart Archive,Vol.4., 300; 353. The quote is from page 300. For an earlier period see Fynn's diary (1986: 295) recorded in the first half of the 19th century, which describes how men who visited a kraal were allowed to hlobonga with available girls "The plan is repeated as often as strangers make their appearance, so that one girl may have 100 sweethearts, as also a man the same".

13. Relevant to the importance given to fertility, it has long been noted that ilobolo payments, rather than marking the simple sale of women,represented in part the transfer of women's reproductive capabilities from one kin group to another; or, as Jeffreys (1951) put it, they represented "child price" and not "bride price". So powerful was the association between fertility and ilobolo that a wife "without issue" could be replaced for childbearing purposes by her sister or have her ilobolo returned. For return of cattle because a woman was "without issue" see Vanganye v Makanyezi (Durban Archives, 1/ESH 2/1/1/2/1, 1907). For a detailed examination of the politics of sexuality and fertility during the colonial period in Kenya see Thomas (2003).

14. Ndukwana's testimony suggests that under Shaka and Dingane young men were not allowed to hlobonga but that this was relaxed under Mpande, James Stuart Archive, Vol. 4, p. 372. See also Bryant (1949: 533). Hanretta (1998) also makes the important argument that gender relations were far from static in the period of a centralized Zulu state (that followed Shaka's taking of power in 1816 but ended with the defeat of Cetshwayo in 1879). She also illustrates the prominent political roles that women played within the Zulu royal house.

15. Jeff Guy (2003) recently reviewed court cases relating to the inGcugce girls. The episode has also been considered within the context of generational and gendered struggles by Ben Carton (2000). See also Hanretta (1998) for the claim that we must be cautious in interpreting this episode since Cetshwayo may have had political reasons for wanting to destroy the inGcugce age set. Another example of early sacrifice for love comes from Mpondoland where Monica Wilson (1936:189) describes a cliff called iNtombi nenDoda (the girl and the man) named after a the tragedy of a girl who, being forced to marry a man she didn't love, bound herself to her true lover and flung herself over the edge.

16. Cetshwayo's evidence to the 1883 Cape Government Commission on Native Laws and Customs, published in Webb and Wright (eds.) (1987).

17. See, for example Mxokozeli kaBobe v Msincomkwenyana (Durban archives, 1/Esh 2/1/1/2/1, 15/1905); Mficezeli Mtembu v Mcondo Mtembu (Durban Archives, 1/MTU 2/1/1/1, 80/1936).

18. Paulina Dlamini, then 14 or 15, recounts how in the mid-1850s a member of the prominent Buthelezi clan to which her family belonged arrived to request that she serve in Prince Cetshwayo's isigodlo, a section of the royal umuzi containing unmarried women who served the royal house. As well as the importance of tribute to society, the accounts suggests the existence of considerable affection between father and daughter. The life story of Paulina Dlamini was compiled by H. Filter, translated and edited by S. Bourquin, and published in 1986.

19. According to Dohnë's 1857 dictionary, "This word describes a custom between the nearest relations, and is exclusively applied to the female sex, who, when married, are not allowed to call the names of relatives of their husbands nor of their father-in-law. They must keep at a distance from the latter. Hence, they have the habit of inventing new names for the members of the family, which is always resorted to when those names happen to be either derived from, or equivalent to some word of the common language, as, for instance, if the father or brother-in law is called umehlo, which is derived from amehlo, eyes — the isifazi will no longer use amehlo but substitute amakangelo (looking)".

20. Ndukwana told James Stuart in 1902 that "An isifebe is one who, during the same month, hlobonga's with more than one man. An isirobo is a woman who has been deflowered (mekezisa'd) or raped." James Stuart Archive, Vol. 4, p. 353.

21. For an excellent criticism of Eurocentric understandings of morality in Africa, including by the Caldwell's quoted earlier, see Heald (1995).

22. The concept of "spheres of exchange" was popularized through the work of Bohannan (1955).

23. Mayer's (1971) account of amankazana among isiXhosa speakers bears close resemblance to Wilson's description of amadikazi. Colson (1958: 293) suggests that a Tonga widow in Zambia was "free with her favours". However, among isiZulu speakers Bryant (1949: 577; 580) suggests that there was a heavy penalty for adultery and that concubinage was rare. Most oral testimonies collected also support this view. Further, Thembeka Mngomezulu, a nurse and Masters student at UND in her early 50s who grew up close to Pondoland and then married into a Zululand umuzi,contrasts sharply the freedom of the amadikazi of Pondoland, whom she remembers, to the expected chaste behaviour of widows in KwaZulu-Natal (personal communication, 2003).

24. Kohler (1933: 28-29) describes umhayizo as a kind of hysteria brought on by a man who bewitches a girl with whom he is in love. See Parle & Scorgie (2001) for accounts of umhayizo in the early 20th century as well as its continued presence today.

25. Rex v Unkala alias Nkokana (Durban Archives, RSC II/1/35, 72/1885).

26. See Atkins (1993) on the contested move from "peasant time" to "industrial time" among Zulu speakers. The concepts are of course E. P. Thompson's (1967).

27. Most of my elderly informants, who came of age from the 1930s to the 1950s, said that they were not even aware that this practice had taken place.

28. This was the meaning given in most oral testimonies, see also Vilakazi (1962) and Doke & Vilakazi (1948).

29. Mid-century definitions taken from Doke & Vilakazi (1948). Origins of the term appear to be in ukuqaba, to paint, whereby certain non-Christian Africans would cover their bodies with red clay (see Davis, 1872). The "reds" are a prominent category employed by Mayer (1961).

30. As well as sanctioning polygamy, native courts could be used to claim damages from any male who committed adultery with another's wife although unemancipated women, as "perpetual minors", had no such claim. Natal was the only province in South Africa where native law was "codified" — written into a single legal document — by settlers, this taking place in the late 19th century. For vigorous critiques of the effects the Natal Code of Native Law had on women in Natal see Simons (1968) and Horrel (1968). Chanock (1985), discussing Malawi and Zambiam, argued that customary law was at first seen by administrators as a liberating force for women but that it increasingly became seen as a tool to control wayward women (see also Schmidt, 1992 and Hawkins, 2002).

31. Seen in Kohler (1933) and Krige (1936a) although interestingly not Monica Wilson's account of Mpondoland, an areas she chose for its relative isolation from Western contact.

32. Thomas McClendon (2002), reviewing civil court cases from the 1920s to the 1940s, explored tensions within ukungena rooted in Christianity. It is very notable from court cases that in the second half of the 20th century the practice of ukungena had declined greatly and, from testimonies of elderly informants, one gets a sense of the deep unpopular of the practice with most women. The introduction of an "official witness" at customary marriages also provided a legal basis for women to challenge undesirable marriages.

33. Showing how Christianity fostered great gendered conflicts in rural Mpondoland, Monica Wilson (1933: 274) notes how "In the relations between husband and wife the greatest change lies in the introduction of the ideal of a single standard of morality for men and women ..." But she also says, "There is a double standard of sexual morality, and most of the quarrels between husband and wife turn on this" (266). On contestations over men's use of the "idea" of polygamy in urban areas to secure concubines see Longmore (1959) and Wilson and Mafeje (1963). Men's objections to Christianity on the grounds that it would impose too heavy a duty on faithfulness are described well in Mbatha (1960) writing about the Botha Hill area in rural KwaZulu-Natal.

34. Mgwenya v Donki (Durban archives, 1/Esh 2/1/1/2/1, 40/21).

35. One gets a sense in testimonies recorded by James Stuart of the shared displeasure by himself and his usually well-to-do (male) informants at the existence of "town women". See, for instance, James Stuart Achive, Vol. 1, testimony by Kumalo, 1900, p. 225: "Parents have practically lost control over their girls and women". Carolyn Hamilton (1998) describes the close relationship between Stuart and the amakholwa (African Christian) class.

36. See for instance Colson (1958) for Zambia and Testimony of Ndukwana, James Stuart Archive, Vol. 4. for KwaZulu-Natal.

37. The enticement of African women by white men near a shop in Umzinto and by a white post-cart driver are described in the testimony of Qalizwe, James Stuart Archive Vol. 5, 233-234. See also complaints made to the 1906-7 Natal Native Commission about white men luring African women (Natal archives, Pietermaritzburg, 1/NCP/8/3/76). On women and prostitution in South Africa see Van Onselen (1982); Bonner (1990); Jochelson (2001). In other parts of Africa, see Powdermaker (1962), Bujra (1975); Schuster (1979), Bledsoe (1980), Obbo (1980), Parpart (1988), MacGaffey (1988), and White (1990). See Barnes (1999) on how access to housing could shape sex exchanges in colonial Zimbabwe. Standing (1992) provides an excellent anthropological review of how narrow notions of "prostitution" can be inappropriately applied to Africa.

38. Rape case where woman claims that a man frequently asked to have connection and offered a cotton blanket for her consent: Rex v Umvenyana (Durban Archives, RSC II/1/78 104/1898). Rape case where man is alleged to have offering one pound for the woman to stay quiet: Rex v Umunyu (Durban Archives, RSC II/1/64, 28/1895). These themes persist well into the 20th century: Rape cases where man claims that money was agreed: Rex v Joyini Sibiya (Duban Archives, RSC 1/483 19/1959). Rape case where man is alleged to have offering five pounds for women to stay quiet: Rex v Sidlile Cele (Durban archive, RSC 1/495 133/1958).

39. I first learnt about this colonial sources on courting gifts from Mike Mahoney and followed up the references given in Mahoney and Parle (2003). This piece tends to stress similarities between these payments and contemporary "transactional sex" while I emphasise differences. I have relied on the following records in the Natal Archives, Pietermaritzburg: SNA 1/4/8 Confidential papers 1-117, 1900; SNA 1/4/11 Confidential papers. No. 59, 1902; SNA 1/1/385, Minute Paper 3553-3679, 1907; SNA 1/1/399, minute papers, 1908.

40. On fathers' attempts to circumvent limitations to ilobolo by introducing fees such as imvulamlomo (literally "open the mouth", paid to instigate ilobolo negotiations) see Braatvedt (1927). On umnyobo as one form of "excess ilobolo", see SNA 1/4/8 Confidential Papers, No. 59, 1902. Asst. Magistrate, Pinetown, 22nd August 1911.

41. Moore & Vaughan (1994) provide an excellent critique of money as simply breaking open "tribal" society and undermining "tradition". They describe how the Bemba custom, whereby newly married men worked for their fathers-in-law, waned as young men became employed in the colonial economy. Nonetheless, the money with which they returned replaced tribute labour relatively smoothly as a sign of commitment and for its material value.

42. See Krige (1936a) vii and 104-106.

43. Krige doesn't specifically mention the amaqhikiza but her account of "tribal sanctions" that she says were being eroded is so similar to that described in The Social System of the Zulu that it is clear that these older girls are the amaqhikiza. See also Schapera (1933) on the rise in pre-marital pregnancy in a Christianised group in Bechuanaland, due, he says, to men's greater migration and the reduction in control through initiation ceremonies and practices such as the singing of songs of mockery to humiliate women who fell pregnant.

44. See also dictionary definitions in Dohnë (1857); Davis (1872); Bryant (1905); Doke & Vilakazi (1948).

passages | http://quod.lib.umich.edu/p/passages/