Lorato had struck on the idea for a family party quite spontaneously, not long into the new year. ‘Isn’t the old woman turning 65 this year?’ she had remarked with careful nonchalance as several of us sat in the lelwapa one morning. Nobody was quite sure; Mmapula herself was fuzzy about the year she had been born. ‘Anyway, we should have a party for her,’ Lorato continued, adding, ‘We’ve never had a big party at home.’ Smaller parties had been frequent enough, but Mmapula and her children often voiced their disappointment that nothing larger – specifically, no weddings – had yet been held in the yard.

Modiri, Kelebogile, and Oratile were all sitting nearby. Almost immediately, they began thinking up what they could provide and whom they could invite, assessing potential problems and solutions. They were pragmatic and muted, but undoubtedly keen. Modiri noted that having a party for Mmapula without involving Dipuo would create serious misunderstandings and would worsen existing tensions between them; so the siblings agreed to have an event that would celebrate both of their parents together, as a way of thanking them for having raised their children so well. Modiri was deputed to speak to Dipuo, and Kelebogile was asked to sound out Mmapula, to ensure that both were on board and to seek their advice.

Once the proper motivation and type of party had been established, and the elderly couple had given their approval, a date in December was set and preparations began. They were extensive and drawn-out, moving slowly and stalling frequently at first, picking up urgency as time progressed and the scale of the event grew. What started as a simple idea quickly became ambitious – and costly. We met monthly, and at every meeting it seemed that a new expense had been identified. Didn’t we need a tent? A sound system? What about drinks? More food? Printed invitations? And the house had to be fixed up … Each time the new cost was voiced, everyone would shift uncomfortably and look at their shoes. Kelebogile was often quick to say that she had no money; none of us had much to spare, and the everyday costs of running the household already weighed heavily. And yet there was no question that the expense – whatever it was, whether hiring a tent or printing T-shirts – was necessary; it was simply accepted as such. And so the issue would be left hanging, the oppressive weight of expectation over everyone’s heads.

Addressing these emergent costs was all the more difficult because not all of the siblings attended the meetings regularly, or at all. Moagi was out of town; Kagiso was seldom home, regardless of how often we changed our meeting times to anticipate his schedule. Khumo came once or twice, nearer the end, but everyone was aware of her financial circumstances and expected her to help mostly on the day itself. A flat contribution rate per adult family member was decided among the lead organisers who were present – usually Lorato, Kelebogile, Oratile, and Modiri – but it was virtually impossible to enforce it with those who had not been there to agree to it. Hoping to draw in help from the extended family, a larger meeting was called perhaps two months before the main event, involving representatives from among the siblings’ age-mates, identified with Mmapula’s help. But, on the day, only two people came, and certain key figures – the sister’s daughter Mmapula had raised as her own child, and Mmapula’s malome (her mother’s brother’s son, who had inherited the position from his late father) – were absent and sent no word. Such a discouraging silence puzzled and dismayed the siblings, and Mmapula as well.

In the context of this uncertainty, Mmapula indicated that we should go to make invitations in person. Doing so was a much more formal process than I had anticipated; it involved us going as a small contingent – Mmapula, Oratile, and Lorato, with myself as driver – from yard to yard among the relatives, most of whom lived some distance away, in the next village. We moved in a specific order: first to Dipuo’s relatives (from his father’s brother’s son, to his sister’s daughter, to his brother’s children); then to Mmapula’s brother’s house. Each time we were offered chairs in the lelwapa of our hosts and sat shoulder to shoulder, facing outwards; and each time, after exchanging greetings and ensuring that our hosts knew who each of us were and how we were related, Mmapula conveyed the formal invitation. Oratile and Lorato were occasionally as clueless as I was about the relationships, even where we all knew the house and people of the yard from weddings and funerals we had attended. ‘I’ll never remember all of these relationships!’ sighed Lorato as we drove home. ‘At least if one of my sons was married I would have a daughter-in-law to send,’ rejoined Mmapula with a note of melancholy, gazing out of the window.

As the party approached, we met more frequently, the question of contributions became more urgent, and there were more favours to be asked, things to be bought or collected, and choices to be agreed upon. Money began to materialise from somewhere – motshelo contributions, debts recalled or incurred, partners, savings, it was never quite clear where – and would sometimes be noted in meetings, sometimes not. No one wanted to advertise their wherewithal too liberally. I didn’t even realise the old man was getting a new suit courtesy of his sons until I saw him wearing it. We met for the last time late into the night before the event – it was the first time all of the siblings had met together, in Mmapula and Dipuo’s presence – and we ironed out the last costs, contributions, programme details, and errands to be run. Moagi, in charge of the meeting, thanked everyone for their hard work and invited his parents to offer words of encouragement or advice. ‘Some people are jealous, and they will try to make problems with what you have done,’ noted Dipuo. ‘Work together, show them you are together,’ he added, without apparent irony.

From the moment guests began trickling into the yard the next morning, they were carefully managed. The women – mostly friends and neighbours, plus a few younger relatives – began arriving first, and were directed to long tables set up in a fenced-in space under the trees to help clean and prepare mounds of potatoes, carrots, squash, and cabbage that had been bought. Enormous logs, gathered by the young men all week, were set in radiating circles to create several low fires not far from the tables, where still other women cooked bread for the helpers’ breakfasts. The large pots waited in the wings, deployed later for the cooking of stews and vegetables, samp and sorghum, with a few left to the young men for cooking the tender beef seswaa. Older men and women arrived in the early afternoon, the men sitting with Dipuo in an impromptu semicircle just behind the tent, the women helping with the work that remained to be done as everyone waited for the official programme to begin.

Everyone stayed outside. We had spent hours painting and stuccoing the houses, but they were really just a backdrop to the event: virtually no one went in. I chopped fruit in the indoor kitchen in the morning, as it was the only counter space I could find, but even the children from the yard were reluctant to join me there to help. After we left, it remained empty. Mmapula’s adjoining room out front was used to dress Dipuo and, later, the children. In the secondary house, Mmapula and the women used Kelebogile’s room to change in – we all went from cooking clothes to formal clothes and then to T-shirts as the day proceeded. It was also used as a storeroom for drinks and plates, as well as anything of value; access was regulated by Kelebogile’s key and was restricted mostly to immediate family. My room had been cleared out and now housed everything from meat to cooking dishes, ginger beer to salads. Family, close neighbours and friends or relatives who were helping with the cooking – generally only the women – came and went freely, but efficiently, and did not linger. But perhaps most strikingly, the lelwapa was left clear the entire day. The large tent, where the tables were set out for guests and the speeches given, sat at its front edge; at mealtime, older women sat in chairs on the stoep around its edge to eat, children perched below them. And while it became a thoroughfare for those of us cooking and serving, no one dallied or sat in the lelwapa, which – as we have seen – was where most of the family’s meals were taken and where guests were usually welcomed. While anyone and everyone had been invited into the yard, they were not only differentially restricted from the intimate spaces of the house (the bedrooms); they were also uniformly excluded from the shared living spaces (kitchen, sitting room) and even the distinctly public–private heart of the home – the lelwapa. People were drawn into the yard, but they were kept at a distance befitting the boundaries of the family and their existing relationships to it, which the party served to rearticulate.

Establishing boundaries of this sort was in many ways the business of the day. They were established in space and in movement, in terms of who could contribute what and how, and in terms of which relationships were on display and which were not. When Lorato’s boyfriend turned up unexpectedly on the perimeters of the yard after dark, she served him outside with some annoyance. ‘Two of my uncles saw him,’ she explained later, adding, ‘I don’t need him to be seen by my uncles at a party like this.’ Although Lorato’s failed pregnancy had made her relationship visible, the rockiness of negotiations thereafter made her boyfriend a figure better hidden from both the family and their guests.

But the boundaries were not always clear. After most of the guests had gone home in the evening, with just a few close friends and neighbours remaining behind to barbecue the leftover meat, the siblings gathered together in Kelebogile’s room to debrief. They invited their parents to join them. The gifts Dipuo and Mmapula had received were all laid out neatly on the floor: large cooking pots, oversized enamel serving dishes, tea sets, other household goods, and money. They had come from friends, neighbours, and family who considered the old couple to be elders or malome (mother’s brother) and mma malome (lit. female uncle/mother’s brother; usually the wife of malome). Dipuo made a special example of the beautiful new cooking pots one of his sisters’ sons had provided. ‘You see what beautiful things my relatives have given us,’ he said. ‘I have been a malome to them,’ he added, before exhaustively listing each marriage negotiation with which he had assisted, weddings and funerals attended, help given for children and houses. His children listened, nonplussed by the implicit, critical comparison: none of them were married, and none of them had completed a house of his or her own. ‘I’m going to take these presents that were given by my family, because they were given to appreciate me and my help,’ he concluded.

Everyone kept their faces studiously blank. After asking Mmapula whether she had any words for them – she had none, except to thank them for the day – they let their parents go so that they could evaluate the party in more depth among themselves. It was only at this point that they voiced their shock and hurt. ‘Did you hear what that old man was saying?’ asked Kelebogile incredulously. ‘Always his sister’s sons, his sister’s sons [bo setlogolo]. Why should he take those things? They’re also for his wife!’

The debrief meeting, held among the wreckage of the day’s event – rumpled piles of clothes and half-finished bottles of soft drink, the jumbled presents and a couple of sleeping children – was in many ways a tallying of the day’s ignominies, many of them generated by Dipuo. ‘Hei,’ began Moagi, ‘this old man was refusing even to get dressed this morning.’ He recounted Dipuo’s complaints about his new trousers being ill-fitting, disliking his tie, and completely refusing to wear his shoes as one might recount the misbehaviour of a stubborn child. Dipuo had a serviceable pair of shoes Kagiso had bought for him, but a couple of days previously he had refused to wear them to the party. Kagiso had dashed to town the day before the party to buy a new pair; these, too, the old man had rejected, just that morning, complaining that he didn’t like their style. Instead, he had chosen a battered pair provided some time back by one of his sisters’ sons. ‘He takes his sister’s sons as if they are his children – as if he doesn’t have children,’ reflected Kelebogile. Modiri and Moagi echoed her last statement word for word, and the others hummed in dismayed agreement. Given everything the siblings had spent on and put into the party, and combined with reports that had filtered back to them on the old man’s earlier speech, it was a particularly bitter pill to swallow.

Someone knocked at the door as these tales and grievances were being recounted. ‘We’re talking!’ answered Modiri, ensuring that the door was shut tight. Despite frequent knocks, no one was let in for the duration of the meeting – with the exception of a neighbour’s child who was sent to ask for a drink. Everyone fell carefully silent while she was given one and sent out.

The siblings reassured one another on having provided more than enough food, noting that they had overheard people commenting with satisfaction on how well they had eaten and how amply even latecomers were served. Grumbling about the lack of food after a party was a common means of signifying the event’s failure and casting doubt on the hosts. ‘Nobody can say they went home not eating,’ noted Modiri with a combination of approbation, relief, and latent concern. They were equally pleased with having kept the programme on schedule, and with the number and variety of guests who had come (aside from one or two notable absences). ‘I heard some people saying it’s like we were marrying our parents!’ noted Lorato with a laugh and visible satisfaction. But it was small consolation. ‘We need to call this old man and talk to him,’ asserted Moagi finally, to general agreement. ‘We have to tell him it’s not okay for him to treat us like nothing in front of everyone,’ agreed Kelebogile.

Dipuo was never called. The next day everyone was busy cleaning the yard and house after the party, returning things rented and borrowed. The day after, children were being prepared to visit their other parents’ extended families before Christmas, or to go to help at the lands. Moagi was getting ready for the long drive back to his base. I asked quietly once or twice whether they were still planning to call their father; I was met with shrugs, sighs, and indications that Moagi would be leaving and it wouldn’t be right to have the meeting without everyone concerned present. And so the issue was left to lie – like so many others.

As we saw in Part III, making intimate relationships recognisable is a key means of making them kin relations. The same might be said of large-scale family celebrations: parties involve a public performance of kinship and an explicit display and narration of who is related to whom and how, and of the historical trajectories and qualities of those relationships. The family genealogy was recounted, identifying which villages (and merafe) each ancestor had come from; within that context, Moagi introduced each member of the family by order of age, describing who was whose child and their specific contributions and importance to the family. Similar genealogical accounts characterise Tswana wedding feasts. Just as a pregnancy makes a previously hidden intimate relationship visible and knowable, a party throws the entire network of kin relations into public relief; and, as the frequency of parties suggests, this performance is a key means of expressing and sustaining kinship.

Parties, however, are carefully organised to make certain dimensions of the family publicly recognisable and to obscure or downplay others. Celebratory events are meant to demonstrate the achievement of key family ideals: harmonious cooperation, or tirisanyo mmogo, self-sufficiency, and the ability to provide for others. A beautifully built house, the ability to mobilise contributions of things and labour, comfortable surroundings, ample food, music, and entertainment, and the seamless coordination of everything from invitations to yard preparations, cooking to the official programme – all are key indicators of the achievement of these ideals. In this sense, parties draw together and provide an opportunity to publicly perform all the ideals of Tswana kinship we have explored so far.

Of course, taking on such a task runs a significant risk of failing to live up to those ideals. The entire planning process had been fraught with refusals, absences, regrets, and the risk of failure – dikgang, now extended across a wide field of relations. The family’s images of itself had been challenged; its relative success in achieving kin ideals – of marrying sons to acquire daughters-in-law, for example, or of retaining the support of children raised on behalf of others – had been thrown into question. Just as parties draw together all of the kinship ideals we have previously discussed, they echo the linked sources of dikgang: the organisation and management of space, and movement to and through it; contributions of material resources and of work; silence and speech, visibility and knowability. And the danger of dikgang is exacerbated in the public display that the event involves. Inviting so many people to participate led to heightened scrutiny and substantial potential for disappointment, criticism, and bad feeling. Celebrations risked putting a kin network’s functionalities and dysfunctionalities, successes and shortfalls on display; and these ambiguities were not simply exposed to the family itself, but to friends, neighbours, and even strangers.

But these dangers were anticipated throughout the planning process, too. Holding parties like the one described here deliberately invites risk and danger into the yard, and into the very heart of the family. Celebrations at home perform familial success by setting it a sort of test. The cohesiveness and strength of the family are implicitly proven in its ability to absorb and withstand the dikgang presented by their invitees – incorporating the full range of their extended families as well as friends, neighbours, and colleagues. And the family is given a unique opportunity to identify and deal explicitly with the problems that emerge in the process (such as Dipuo’s intransigence).

Unlike pregnancies and marriage arrangements, however, parties do not involve any explicit, public negotiation or collective reflection on these risks. Dikgang, and the means of their resolution, are obscured, concealed, and restricted to specific members of the hosting family. It is in this containment of problems and their resolution that parties work to define the limits of family. The management of dikgang does not simply extend the possibilities of kinship ad infinitum; it draws its boundaries, too. In spite of the kin-like contributions and behaviours expected of guests – in readying the yard, in making contributions, in the preparation, cooking, and serving of food, in the eating and cleaning up, or in giving gifts, all of which we have seen featured in kin-making – the party is decidedly not a means of extending kinship. Instead, it restricts kinship, performs these restrictions publicly, and defines a public by virtue of its exclusion. At the same time, a certain hierarchy between the family and that public is established. The family demonstrates its capacity to reach distant relatives, friends, neighbours, and other community members – in mobilising resources and labour, in providing food, in accommodating, in calling and sending, and so forth – while containing the dangers that arise from that extension, revealing a power that goes beyond self-sufficiency and draws others into relationships of care and obligation. This process of defining kin and community against one another, and of establishing the priority of the former in generating the politics of the latter, is one way in which we might better understand the proverb motse o lwapeng – the village is in the home.

There are, of course, concomitant processes or attempts at realigning the relationships internal to the family, too – although they were more experimental, and, in this case, rather less successful. The party’s consistent echo of wedding celebrations – in a context where none of the siblings were married and the family’s attempts to negotiate marriage had been frustrated – marked a certain innovative attempt at self-making on the siblings’ behalf. Dipuo’s public reproach of their filial failures, in this reading, comes across more as a rejection of that particular claim than a wilful exposure of his family to public censure (although it also had to be handled as the latter). Notably, Dipuo did not dissuade his children from throwing the party in the first place, nor did he attempt to divide them or turn them against one another, as he had in other situations; he encouraged their display of togetherness and of harmonious cooperation, both explicitly and by providing them with a common cause to rail against. What he seemed to reject were the claims the siblings were making: the claim that the process of raising them was complete, and that they were therefore fully fledged adults; or the claim that they were self-sufficient enough to remarry their parents, thereby implicitly divesting themselves of further responsibilities to the pair, and celebrating themselves and their ascension to a new social role. Whether by pointedly wearing the shoes and claiming the gifts given to him by his married, established sister’s sons, or by rejecting the presents given by his own children and shaming their behaviour as children (much less adults), Dipuo repeatedly refused to acknowledge these new claims on personhood as being equivalent to those acquired through marriage, building, and other more traditional routes. And his refusal – combined with his wife’s frustrations in not having a daughter-in-law she could send to make invitations, or in being disappointed both by a child she had raised and by her malome – suggests a further implication: that Dipuo’s and Mmapula’s self-making projects had also been inhibited by their and their children’s failures to secure obligations among kin, marriages, and so on, failures that were put on display over the course of the party’s organisation.

Recognising how parents’ and children’s aspirations to self-making are bound up in each other leads us to another way in which boundaries within the family were being renegotiated during the party: in terms of intergenerational relationships. The impression that the siblings were marrying their parents, noted by guests at the event, irked Dipuo in particular not simply for its untoward claim of adulthood on the part of his children, but for the inversion of generational order it suggested. Of course, this particular inversion has precedent in Tswana custom: historically, sons could pay bogadi for their mothers in the absence or after the death of their fathers, partly to ensure their own legitimacy (Schapera Reference Schapera1933). But this customary practice suggests the need to replace a father, where bogadi debts have been unpaid and marriage negotiations unsuccessful. Given that Dipuo was not only present but had successfully managed his own marriage as well as securing the marriages of others, his resistance to that interpretation of the party becomes clear. His refusal to wear clothes provided for him, as might be provided by a parent to a child, and his emphasis on how many quality gifts he had been able to acquire through the superior filial bonds he had crafted with his sister’s sons, indicated resistance to his children’s apparent attempt to undermine and claim his authority.Footnote ¹

As we have seen, Dipuo’s authority tended to be most visible not in his provision of goods or support for his family, but in his role as a negotiator of dikgang. In underlining his achievements as a malome to his sister’s sons, his success in negotiating their marriages, and his ongoing responsibility for conflict management in their relationships, he was asserting the validity of his claim to authority primarily in those terms (ignoring the failures that may have affected his own children, which implicitly became their responsibility). But Dipuo also seemed to have been asserting his unique authority by creating problems that his children could not address – and about which they could not call upon anyone else to assist them. Whether in his slyly damnatory speech at the party, his self-aggrandisement among his children as they debriefed, or in his past indiscretions and the upheavals they caused, Dipuo’s greatest power lay in his ability to provoke dikgang among his kin. In his work among a neighbouring morafe (polity), Rijk van Dijk observes that ‘[p]laying havoc is reserved for the elderly, particularly for adult men’ (2012a: 152); but more than a right of mischief earned through age, this causing havoc both demonstrates and reproduces power. As we have seen throughout this book, dikgang shape gendered persons, relationships, and hierarchies; and the role that older men, especially bomalome, play in navigating them is key to reinforcing patriarchy as the fundamental moral order (Werbner Reference Werbner2016: 85). I suggest that the ability to make potentially serious trouble that cannot be addressed or ameliorated by anyone else is more than a matter of skills in negotiation, mediation, influence, and consensus building (pace Wylie Reference Wylie1991); it stands at the core of the power enjoyed by older men – and at the core of the gerontocratic patriarchy that characterises Tswana sociality.

Ultimately, Dipuo’s children seemed to recognise and accept their failure, despite the success of the party itself. They did not call the old man to account, as one might do with a wayward dependant or someone over whom one had established some authority, and they did not pursue the matter with anyone else. While the party held out the possibility of self-making for the siblings and their parents, as well as different intergenerational relationships, it also reinforced the limits of those possibilities. It generated the means to engage and negotiate tensions between the preservation of family unity and the promotion of individual members’ self-making projects, but also between ensuring the progressive intergenerational transmission of authority and retaining intergenerational hierarchies and the claims of obligation and support they enable. These tensions, and the ways in which they could be negotiated, became most apparent in the kgang of Dipuo’s intransigence and the ways in which his children handled the situation.

As in previous chapters, attempts to assert and enable self-making while retaining responsibility to the family, or to enable the progression of generations while preserving hierarchies, are a source of dikgang; and dikgang in turn enable a tenuous balance to be struck between those otherwise contradictory imperatives. What the example of the party underlines is the importance of an explicitly public audience or context in this process. Building, cars and metshelo, pregnancy, marriage, and the emergence of intimate relationships, and the care of others’ children – these all derive both their riskiness and their salience to self-making not simply from recognition among kin but from their apprehension by a wider, specifically non-kin audience as well.

Of course, it is not only families, or family-run celebrations, who set the limits and terms of engagement between kin and community. A few months before the party, Dithaba had been preoccupied with the homecoming of the first age regiment – or mophato – to be initiated among the Balete in over a generation. In Chapter 14, I turn to this homecoming celebration to examine how the revival of a lapsed tradition sought to reorder relationships between selves, families, and the tribal polity – and thereby regenerate a collective ethics.

It was a warm afternoon in early September, and hundreds of people from the surrounding villages had gathered at the main kgotla. Anyone who could get away from work and make the trip to the district’s main village, Maropeng – the administrative locus of the morafe (tribal polity) – had come to welcome back the first mophato, or age regiment, to be initiated in nearly 40 years.

People had been milling around the stone walls of the kgotla since late morning, exchanging greetings and speculating on when the initiates would arrive and how the afternoon would unfold. The mophato’s return had been greatly anticipated since they had left a month previously, and the initiation had been the subject of frequent conversations both at home and around the district in the interim. Mmapula and Dipuo had both been initiated, as had many other elders in Dithaba, but anyone younger had learned what little they knew about bogwera and bojale – men’s and women’s initiations – from stories and schoolbooks, and many were acutely curious. For perhaps the first time in my fieldwork, almost everyone was as confused as I was about what would happen next and what it all meant. Our collective bewilderment gave the day an air of festive camaraderie.

No one seemed sure about why the initiations had been discontinued in the first place. Official rationales, provided to local media outlets by the paramount chief’s office, focused on recurrent drought and South Africa’s political instability in the 1980s, which had a habit of spilling over the nearby border (Midweek Sun 2012). Anti-apartheid activists frequently sought safe haven in Botswana’s border towns, of which the district’s main village was one, or in the empty stretches of bush between them, where initiations were held. The unrest, of course, had died down long ago, but the initiations had not been revived in the interim.

Most other merafe in Botswana stopped running initiations in the colonial era – generally under pressure from the missions, which imagined them as lascivious and violent at worst, or as ‘tedious ceremonies’ that created ‘prodigious barriers to the gospel’ at best (Moffat 1842: 66; see also Schapera Reference Schapera1955 [1938]: 105–6; Werbner Reference Werbner2009: 453). Some merafe, however, continued to initiate age regiments of men and women intermittently throughout the colonial era – most notably the Bakgatla (Schapera Reference Schapera1955 [1938]: 106; Setlhabi Reference Setlhabi2014b: 461–3), a group of whom had settled in Dithaba during that time. But even among the Bakgatla, the practice was abandoned, revived, and abandoned again (Setlhabi Reference Setlhabi2014b). Girls’ puberty rituals – which share several symbolic aspects with the initiation of age sets but focus more on individual initiates – had been sustained further north in Tswapong; latterly, however, at around the same time that initiation revivals in the south gained steam, they began suffering a drop-off in participation, as girls sought more ‘progressive’ and ‘modern’ ways of being and becoming women (Werbner Reference Werbner2009; Reference Werbner2014b).

This tension between the ‘traditional’ and the ‘modern’, the political claims and subjectivities each category marked, was one likely source of the lapses – and revivals – in bogwera (the men’s initiation). When Lentswe II, the paramount chief or kgosikgolo of the Bakgatla, reintroduced bogwera in 1975 after a long absence, it was explicitly aimed at ‘the restoration of our cultural values and civilisation’ (quoted in Grant Reference Grant1984: 15). In response, Sir Seretse Khama – the first president of the Republic and the paramount chief of the Bangwato – denounced bogwera as primitive, divisive, and tribalistic, an ‘impediment to progress’ (ibid.) for a nation seeking to assert a unified, homogeneous national culture and identity in place of the ethnic fragmentation of Britain’s colonial government (Werbner Reference Werbner and Werbner2008: 38–40). The echo of the missions’ disparagements was not accidental.

As Keletso Gaone Setlhabi argues for the bojale (women’s initiation) among the Bakgatla, initiations are often bound up with the political needs of the paramount chief to assert authority, an ‘indication of his power and prestige’ (Setlhabi Reference Setlhabi2014b: 463) that harks back to a time when mephato were military and labour regiments at the service of the kgosikgolo. As she observes, the initiations of Kgatla mephato map directly onto the years when new Kgatla chiefs were installed, with additional initiations held when needed for their wives or heirs. The revival initiations run by Lentswe II in 1975, and later by Kgafela II in 2009,Footnote ¹ also marked a ‘more ambitious project to reclaim the power of the Tswana chiefs, abolished by the Botswana constitution and a succession of laws, by enhancing … chiefly autonomy and independence’ (Werbner Reference Werbner2014b: 375–6).Footnote ² But while initiations could be highly responsive to a kgosikgolo’s political needs, they could equally well lapse in his (or her) absence (Setlhabi Reference Setlhabi2014b: 462); and, as became clear among the Balete in the apartheid era, they were also highly sensitive to political upheaval.

Political imperatives were less obvious at the Balete initiation in 2012. The paramount chief, Mosadi Seboko, had been installed nearly a decade previously – the first woman to hold the role. Unlike the contentious kgosikgolo of the Bakgatla, she had no political axe to grind with the Botswana government. But kgosikgolo Seboko shared other concerns with her Bakgatla peers. In Maropeng, the revival of initiations was justified explicitly in terms of repairing the moral fibre of tribal and family life – which, in the alarmist terms of one local newspaper, was beset by corruption, degradation, and the inability to run families properly (Midweek Sun 2012). AIDS was cast as one of many symptoms of this purported social rot. Specifically, initiation was intended as a means of promoting botho – often translated as ‘humanity’ but also literally ‘personhood’, a powerful moral standard of dignity, humaneness, respect, and civility (see Livingston Reference Livingston2008) – as an antidote to these iniquities. In other words, this initiation was deeply preoccupied with regenerating a collective ethics, and was intended to inspire collective reflection on shared and intransigent ethical dilemmas – a preoccupation familiar from family negotiations of dikgang.

Our uncertainty about how the ritual events would play out had precedent: that process of collective reflection was not only historically contested but deliberately left open to interpretation. In Pnina Werbner’s (Reference Werbner2009) account of Tswapong girls’ puberty rituals, the elders’ explanations of what was happening and what it meant were multiple and highly variable, mapping out a landscape of interpretive – and ethical – possibility that the new initiates would no doubt reinterpret in their turn, adapting it to the circumstances in which they lived. As far back as 1909, W. C. Willoughby commented in his reports on ‘Becwana’ initiations that ‘the significance of the ritual is not known even by the tribes that preserve it’ (Reference Willoughby1909: 228). Ignorance and confusion around initiation rituals are, of course, one way of distinguishing the initiated from the uninitiated, and of privileging those with ritual knowledge (LaFontaine Reference La Fontaine1985: 16–17). However, I suggest that the interpretive open-endedness that characterises Tswana initiations, even among the initiated, is also a potent site for the collective exercise of the moral imagination.

If initiation ‘shapes an ethical subjectivity’ (Werbner Reference Werbner2009: 441), then, it also marks a zone of contestation over what ethical frameworks ought to apply and what sorts of subjectivities are desirable – and these are political questions as well as ethical ones (Werbner Reference Werbner2014b). These questions are worked out, in part, by collectively negotiating appropriate relationships between the self, the losika (or family), and the morafe (or polity). In this sense, initiations provide an important antecedent and context to the NGO and government interventions I have described in this book. They offer an insight into the ways families have figured in and managed such interventions in the past, and with what implications for the relationships between lelwapa and kgotla, losika and morafe.

I do not propose, in this chapter, to attempt a full analysis of Tswana initiation on the basis of a single homecoming celebration.Footnote ³ However, following Maurice Bloch (Reference Bloch1992), I take it that the reintegration stage of such a rite of passage might be illustrative of its legacies and implications, and may therefore have much to say about the relationships between self, losika, and morafe that initiation mediates. Specifically, I suggest that the homecoming condensed a long-standing interdependence between these domains, in which being able to mobilise labour and contributions from within families was key to establishing, asserting, and centralising political power in the morafe, and in which family histories, relationships, contributions, and care were highlighted and reanimated in turn. It also enacted a series of distinctions: between the initiate and his family; among malwapa; between lelwapa and kgotla; and, ultimately, among merafe and between the morafe and the nation state (Setlhabi Reference Setlhabi2014b). And, as elsewhere, these interdependencies and distinctions, and their ethical implications, are made evident in the generation and management of dikgang.

The initiation was carefully veiled in secrecy: initiates – past and present – were explicitly forbidden from discussing what the process entailed (see Setlhabi Reference Setlhabi2014a on secrecy and bojale). But between pestering the elders at home, prompting their age-mates among our neighbours, and considerable speculation, the Legae siblings and I had cobbled together a few ideas. Mmapula and Dipuo explained that the men would learn the history of the morafe and its songs and practices – although, based on the send-off event, Mmapula was concerned that they would be learning generic Setswana songs rather than those particular to the morafe. They would learn to hunt. Rra Ditau, our neighbour and builder, had tipped us off that initiates also learned minor witchcraft – of the sort that was necessary to protect oneself, one’s cattle, and one’s family, or to identify and divert malicious threats and attacks sent by others. And, perhaps most importantly, the men would be circumcised and ‘doctored’ with herbs thereafter – although official statements on the current initiation had carefully aligned themselves with the Safe Male Circumcision campaigns to curtail HIV and AIDS and noted that trained doctors would be involved (Midweek Sun 2012). When they returned, the initiates would be recognised as men. Indeed, once they were back in the village, initiates were greeted and congratulated with shouts of ‘O tla nyala!’ – You will marry! – although many were already married and had had children long ago. In fact, the initiates ranged in age from their early twenties to their late forties, there having been no initiations for so long.

None of the Legae brothers had opted to participate. Tuelo, the youngest, had originally planned to go along and had attended preparatory meetings, but at the last minute he backed out. Kagiso was adamantly disinterested. ‘Ga ke motho wa dilo tse,’ he said dismissively – I’m not a person for these things, implying that with their dalliances in witchcraft and tradition they were inappropriate for a born-again Christian. Modiri and Moagi registered no particular interest. Oratile and Kelebogile were ambivalent when toying with the idea of participating in the women’s initiation planned for the following year. Kelebogile was up for it until her mother told her that she had had to sit quietly and without reacting next to a snake at her own initiation, at which point Kelebogile changed her mind abruptly. Neither Mmapula nor Dipuo put any pressure on their children to participate; indeed, both official discourse and village conversation seemed to stress the importance of initiates choosing to participate for themselves, although they required fairly hefty ongoing sponsorship from their families (to which we will return). Only Mmapula’s malome from the main village – or rather the son of her malome, who had inherited the responsibility – had decided to attend. We were hoping to find him among the men at the homecoming.

By early afternoon, word had spread that the mophato would soon arrive. The milling spectators converged on the main road into the kgotla in anticipation, their phone cameras readied, jostling one another with an air of companionable merriment. Someone wedged herself through the crowd to stand next to me. I glanced up, surprised to find Mmabontle giving me a mischievous look. She was an old friend from Dithaba with whom I had worked at the orphan care centre for some time, but hardly someone I had expected to see there. After some affectionate teasing and banter, I asked whether she had come specifically to see the mophato.

‘Ee, I’m here for Tharo,’ she explained. This statement came out of nowhere, and I was baffled. Tharo was a young man we both knew from the orphan care project in Dithaba, but then we both knew plenty of young people that way, and it seemed an odd reason for her to come all the way to Maropeng. She watched my confusion for a moment with a knowing look. ‘Don’t you know we’re related?’ she added casually, knowing the discovery would give me a shock, and laughing with satisfaction when it did. She explained that after doing some research into a ‘small house’ her father had had – another family outside his marriage – she discovered the link. ‘My father was his grandfather. Anyway,’ she continued, with her characteristic nonchalance, ‘when we heard Tharo was to be initiated, we contributed to buy a cow. I bought him blankets and contributed for some food,’ she added, referencing the costs incurred during the initiation itself.

It was a very generous contribution, given that Mmabontle was already looking after her own and her sister’s children on a fairly meagre income. Tharo’s older sister had been complaining bitterly to me of the initiation’s expense the week before, calculating the combined cost of food, blankets, shoes, and the shorts, beads, and creams with which the men decorated themselves on their return at well over P3,000 (£250) – more than most reasonably employed people in Dithaba made in a month. The cost had been a source of some strife at home, making Mmabontle’s contributions timely – they would have added as much as P1,000 (£85). ‘Hei! They don’t tell you how expensive these things are in the beginning,’ Mmabontle said. ‘You just see them coming to you saying they need more money to feed mogwera [the initiate]. Even these boys they don’t know how much it costs. But what can you do? If the boy wants to be initiated, you see what to do. Look, I made him a purse,’ she added, showing me a small drawstring pouch she had sewn from scraps of cloth to give to him for collecting coins from people who wished to speak to him that day.

In his notes from 1909, Willoughby describes a similar economy of contribution mobilised to feed the initiates. He records that the initiates were housed together by ward and that the women of the ward would prepare and bring out food to them daily, dropping it at a safe distance before retreating. The initiates would eat this food together, along with their initiators. Remembering that wards tended to be patrilines, in Willoughby’s account, both the provision and the consumption of food were active expressions and experiences of kinship, extending from the village into the initiation camp. While the logistics of feeding no doubt worked differently in 2012 – Tharo’s older sister had complained that the initiators insisted on being given packs of Russians, as the popular spicy sausages were known, as well as money – the contributory expectations and process were similar. And for Tharo and Mmabontle, it provided an opportunity for the public acknowledgement and performance of what would previously have been an unknown kinship. Much like the youth posted out on national service (Tirelo Sechaba, or ‘Work for the Nation’, a programme now lapsed), the initiates ‘were situated … as household members receiving care, as engaging in self-development, and as forming links with and for the nation’ (Durham Reference Durham, Cole and Durham2007: 119) – in this case, the morafe.

As Mmabontle and I chatted, older men in blue work overalls and hats moved towards us along the road, pacing back and forth and snapping long, slender sticks against the pavement like whips to clear the route. The spectators moved quickly out of the way; the initiators were rumoured to thrash people if the occasion demanded it. Then we heard ululations and excitement from the top of the road, and, in the distance, above the heads of the crowd, we saw handkerchiefs dancing on the ends of long staffs. Before long, the mophato was trooping past us, each man covered completely in new, heavy blankets, incongruous with their floral prints. It was impossible to see any man’s face, much less recognise him. It reminded me of the way I had seen women covered in blankets for patlo, as their relatives and in-laws whispered advice in their ears. One initiate was driven by in a car, the rumour chasing up the line behind him that he was ill.

The men were herded into the cattle kraal attached to the main kgotla, the high stone walls of which made it impossible for them to be seen. Anyone who tried to climb something nearby to get a look was angrily chased off by one of the initiators. No one was admitted into the kgotla, and so we all waited around in some confusion. Eventually, smaller groups of men – still bundled head to toe in blankets – began to emerge from every exit, heading off in different directions. The crowd scattered, people running to attach themselves to one group or another, following behind them with enthusiasm. I lost Mmabontle, and, like many others, followed one group and then tagged after another, clueless about what was happening until someone explained to me that each group was going back to its home kgotla – of which each ward in the sprawling village had its own.

Lorato joined me soon after she knocked off work, to try to find Mmapula’s malome (whom the whole family took as malome as well). Many of the wandering spectators were not entirely sure in which kgotla they might find their relatives – nor even, in some cases, where the ward kgotlas were. Some simply followed the initiates, although there was no way to recognise anyone unless you knew – having bought – their blanket. After many phone calls home to Dithaba for suggestions, we eventually traced our malome to a yard in the neighbourhood Mmapula and Dipuo had grown up in. We didn’t find the open, stone-walled circle that would ordinarily signify a kgotla, nor even the more old-fashioned semicircle of stripped logs jammed upright in the ground, but we surmised that the yard must have been that of the headman. Like many larger yards in Maropeng, it had a thatched rondavel at the edge of the lelwapa, which had been requisitioned for the men; the lelwapa, or courtyard, had been partitioned and enclosed with a fence of thin hedge branches. We greeted the hosts and elders perched around the edge of the lelwapa individually, most of whom were familiar from past funerals and weddings I had attended. A man sitting in the entrance tried to demand money from everyone who entered, in exchange for the right to speak to the initiates – an act of contribution that would be demanded again the following day by the initiates themselves (a practice also described in Tswapong by Werbner Reference Werbner2009: 453) – but visitors didn’t always oblige.

The initiates were ranged inside with their backs to the thin fence and their legs drawn up, looking tired and ragged, clothed only in cut-off shorts. The gatekeeper told us to greet everyone quickly and move out, but at the insistence of our uncle and a couple of his friends we sat in front of them to chat awhile. To my surprise, I found Tharo among them too, grinning and asking me to bring him a bottle of Coke the next day as he had been craving it. As we chatted, it became clear that most of the other men had been connected to the ward through family history (see also Willoughby Reference Willoughby1909: 230). Given that wards were historically settled by a single patriline, that congruence suggested that most of the men would have been related. While the initiation worked to stratify families and thereby shore up political hierarchies, ‘[t]he ranking system always revealed unknown family histories’ (Setlhabi Reference Setlhabi2014b: 469) as well, demonstrating and reviving unexpected relationships among and between extended kin groups, while transmitting and reproducing them intergenerationally (Werbner Reference Werbner2009: 454; Willoughby Reference Willoughby1909: 230). And yet, specific relationships remained opaque, especially to those of us who were uninitiated. Given his presence there, I surmised that Tharo must also be somehow related to the Legae family, as well as to Mmabontle’s.

We were back in Maropeng again the next morning to see the official welcoming and naming of the mophato by kgosikgolo Seboko. The main kgotla was packed: the large, thatched stage was crammed with dignitaries, and grandstands erected around the open meeting area were jammed with people standing and sitting, many having clambered up onto walls and the roofs of vehicles. The initiated men came trooping in from the various corners of the village at a stomping trot, kicking up clouds of dust around their jostling staffs, glistening red with a mixture of soil and Vaseline they had applied to their bodies. Their hair had been shaved to their scalps and coloured back in, sharp-edged, black, and iridescent. Plastic beads rattled, draped diagonally across their chests. Some blew on the hollowed, twisting horns of kudu antelopes, symbols of a successful hunt. Their initiators circled them with thrashing whips, keeping the crowd back, herding the men back into the cattle kraal, where they stood out of sight until being called in front of the paramount chief.

The official programme of the event unfolded in something of a blur, everyone jostling for space and talking excitedly over the top of one another. It was uncharacteristically brief. Unlike kgotla events for Independence Day and other celebrations – which usually featured long-winded speeches from district bureaucrats, local counsellors and members of parliament, the chief, pastors, and whoever else might be available – only the kgosikgolo spoke. She named the mophato ‘Matsosangwao’ – ‘those who lift up culture’ – emphasising the importance of rediscovering culture as a route to dignity and botho.Footnote ⁴ She described the historical importance of mephato in defending the village, and later in advancing development projects for the community’s benefit; and she emphasised the initiates’ new-won status and the civic responsibilities that went with it, urging them to work for the betterment of the village and to support one another in times of need. The crowd listened impatiently. When the ceremony concluded, the men were trooped back to their respective ward dikgotla, from there to return to their homes. The men from Dithaba and other, more distant villages stayed the night and undertook the entire event again on a smaller scale in their home communities on the following day.

Figure 10 The Matsosangwao mophato returns home.

This series of events around the mophato’s return suggests an interpolation of the morafe into the role of the family in the process of self-making, as we have seen it unfold throughout this book. The main kgotla called, sent, and moved the initiates around the village in ways that were opaque to the uninitiated, and briefly housed them as well – much as they had been called, sent, and housed together in the bush. Initiates were required to mobilise contributions of money, food, and labour to support them during their time away, and to support their initiators too; and the kgosikgolo’s speech emphasised the continuing contributions they would be expected to make to one another and to their villages and morafe. Both the contributions and the organisation of initiates by ward enacted and performed a wide range of kin relationships, including some that were previously unknown, making the bagwera and their kin networks newly recognisable to themselves and each other. The men’s initiation also rendered them recognisably marriageable on their return. All of these undertakings resonate with the interlinked practices of kin-making and self-making we have explored, and in many ways they seem to usurp them from the mogwera’s natal family. At the same time, initiation does not serve to extend or produce kinship among initiates who were not otherwise related, even metaphorically. The kgotla, in other words, produces kin and selves in the same ways families do; but, in the process, it actively distinguishes morafe from losika, the realm of the domestic from the political. And it is the simultaneous enactment of kin work and distinction from kin that underpins the kgotla’s claim to pre-eminence.

This demonstration of the kgotla’s efficacy in self-making and kin-making was especially potent in a context where other means of making-for-oneself were so fraught and difficult to achieve, particularly for men. But its appeal extended to married as well as unmarried men, settled with families or otherwise. More than marking a specific, fixed stage of transition in the lifecycle, bogwera provided an additional, experimental means of self-making – itself an open-ended, cumulative process. The fact that none of the men at home felt obliged to participate – especially those, like Kagiso, with confidence in the ways in which they were already making-for-themselves (through business, church leadership, and marriage negotiation) – underscores the extent to which initiation, revived after so long, was more an optional and alternative approach than a necessary prerequisite to personhood. At the same time, there was some effort made to reassert the value of initiation in self-making. During the entire month that the mophato was out in the bush, weddings and parties were banned, bars were asked to close early, and churches were asked to keep their services quiet (a gesture that suggests the comparable roles of each in the making of persons; see Suggs Reference Suggs2001 on bars and making men). For two nights before the mophato returned, a village-wide curfew and blackout was maintained. The emphasis on maintaining silence, invisibility, and secrecy for the duration of the initiation, and during the subsequent gradual, controlled process of revealing or emergence – as the men returned to the village first covered in blankets, then partly obscured in malwapa (courtyards) scattered all over the village, then resplendent in red body cream and beads at the main kgotla – is reminiscent of the emergence into recognition that pregnancy provokes for women, a permanent sort of recognition to which men otherwise have limited access. By demonstrating its ability to bring about this unique sort of self-making, the kgotla again distinguishes itself from and elevates itself over the lelwapa.

The family is actively backgrounded in this process, if not altogether concealed. Unlike the careful description of relationships that characterised the party – whether during invitations, speeches, or introductions among guests – the public ceremonials of the homecoming obscured and understated kin networks. No one was quite sure where the initiates were going when they left the kgotla; even when they arrived in the yards of familiar (and familial) wards, no one was quite sure whether or how they were all related, and no formal effort was made to describe those relationships. Family queued to see their initiates and paid money to speak with them; the men ate and slept separately. Speeches focused on the men’s achievements and responsibilities, their new roles in the morafe, and their new relationships to one another, rather than to kin. The men were demonstrably distinguished from their families, and the morafe was likewise distinguished from the lelwapa and bound to the kgotla.

At the same time, as narratives such as Mmabontle’s suggest – and they were common currency among spectators as we waited for the mophato’s return, trying to piece together what was unfolding – kin relations permeate bogwera and are crucial to the initiate’s success. An initiate’s family must be willing and able to cobble together money, food, clothing, and other resources sufficient not only to send the initiate off, but also to address immediately any need expressed by his initiators in his name during the initiation, and to welcome him home again – often with lavish celebrations (also described by Willoughby in Reference Willoughby1909). In supporting a man’s initiation, his family demonstrates its ability to cooperate, to provide, and to sustain its members in their self-making projects – opening opportunities of marriageability and the reproduction of the family in turn. As we saw in the example of Mmabontle’s new-found kinship with Tharo, who constitutes an initiate’s family becomes newly evident in who contributes to his sponsorship and upkeep, who clothes him for his homecoming, who takes him gifts upon his return, and who throws him a party. Initiation is thus enabled by kin, performs kinship, and becomes a kin-making process as well. And, indeed, the scope of kinship is unexpectedly expanded in this process: a ‘small house’ is absorbed into the relations of the ‘big house’, long-standing but long-forgotten ward-based patrilineal relations are rediscovered and reanimated, and so on.

Family – and specifically the lelwapa – also has a critical role to play in reintegrating the mophato. The initiated men are considered dangerous when they return from their isolation in the bush. They have great potential to cause damage – hence the preparatory interventions of witchcraft to ease their return into the village, the distance at which people are kept as the mophato travels to the kgotla, their covering in thick blankets, and the imposed curfews. And, of course, they pass through the kgotla – or, at least, its cattle kraal – first. But then they are returned homewards – specifically to one lelwapa in their ancestral wards, which, given that they were historically settled by kin, returns the men to perhaps their widest network of family, and thereby throws the history of their kin relations into relief. Those who could find them there were those who shared and knew those relationships, or were able to discover them (as we did) from family; alternatively, they were those who – being family – had provided the men’s blankets and could identify and follow them accordingly. As such, those who visited the initiates and contributed money to speak with them – a gesture of re-establishing kin economies of contribution, perhaps, acknowledging that the initiates had accumulated a new sort of value – tended to represent the widest possible extension of kin. It is in the space of the lelwapa that the men bathe, shave, and beautify themselves in preparation for their recognition as a mophato the following day, in a sort of preliminary domestic transformation that will allow them to move via the kraal of the kgotla to its central arena – cattle becoming men.Footnote ⁵ The lelwapa, in the case of Maropeng’s contemporary bogwera, is a key space both for containing and mitigating the danger the new initiates present and for rendering them safe again – for re-domesticating them. After their initial return, and before they can be named and officially recognised by the chief, the age regiment is literally contained in the lelwapa; and, in that sense, so too is the political construct of the morafe.

I suggest that this obscured but permeating involvement of kin in initiation is key to the kgotla’s project of regenerating a collective ethics. A school textbook the Legae children showed me listed the first task of initiation as ‘Go ba fa molao wa Setswana’ – to give them (the initiates) the law (Makgeng Reference Makgeng2004: 206; see also Werbner Reference Werbner2009: 450 on ‘laying down the law’ in Tswapong girls’ puberty rites; McNeill Reference Howell2011: 92–101 on the role of Venda laws, or milayo, in initiation). It reminded me of a similar explanation Mmapula had given me of patlo,Footnote ⁶ the gender-segregated, nominally secret session in which a marrying woman is advised by the married women of her own and her spouse’s family on her obligations as a wife, during which she is also ‘given the law’ (men undergo a similar session). Giving initiates, or marrying spouses, molao wa Setswana involves equipping them with an ethical framework to effectively engage dikgang, and thereby make both themselves and their families. While the kgotla constructs and retains the ultimate authority to rule on disputes that cannot be addressed by the family, it also positions initiates to better address dikgang at home and thereby avoid that eventuality. Much of Tswana customary law is geared towards managing disputes of kinship; on this reading, kinship emerges as one major means of the law’s transmission, interpretation, implementation, and change over time (Reece 2021a). Just as the family generates and permeates the village and morafe, the kgotla and the law permeate the family – and the distinctions marked between the two are made in terms of dikgang.

This distinction was already emerging in the bogwera itself, which generated a range of dikgang for initiates’ families to navigate. As we have seen, the expectations that emerge from initiation are likely sources of dikgang among kin, and they must be managed to ensure that they do not interfere with the initiate’s success – or, indeed, the wider success of the mophato. As Tharo’s older sister discovered, mobilising the resources necessary to support him occasioned pushback, shortfalls, and disappointment among his family, much like those that characterise the contribution economies of kin – requiring both her careful management and Mmabontle’s help to ensure that they were suitably addressed. More seriously, the sick man who came back from the bush by car – reputedly fallen ill because of a reaction to the herbs used to heal the circumcision – was nursed for a week at home, until he died, raising the fraught question of whether he or the family had been targeted for supernatural attack, and by whom. While bogwera provoked these issues, they were not addressed in or by the initiation, nor indeed by the kgotla; they had to be managed by kin – who are also, of course, one likely source of the problem.Footnote ⁷ The dikgang produced by the initiation offered a sobering reminder of both the potential threat kin posed and their singular capacity for managing that threat, on which initiate and morafe alike relied.

In Maropeng’s contemporary bogwera, then, the kgotla permeates and animates extended kin networks, bringing their histories and relationships to light; creating new opportunities for self-making and kin-making through kinship practice; and equipping families to negotiate dikgang. At the same time, the losika permeates and animates the morafe, drawing it into being, containing and domesticating it. The mophato, in other words, tapping capacities that Durham associates with youth more broadly, proves ‘key to regenerating household and community interdependency’ (Reference Durham, Cole and Durham2007: 103), enacting and embodying both the ties and the distinctions between losika and morafe. Rather than simply demonstrating that the morafe encompasses or supersedes the family (pace Comaroff Reference Comaroff1985: 98), initiation underscores the ways in which the losika builds the kgotla as well, as the proverb at the beginning of this chapter suggests – an iterative process in which both losika and morafe produce and reproduce one another as collective ethical subjects (see Lazar Reference Lazar2018). In this sense, the Tswana public mirrors the Tswana person – it is brought into being through, if in marked tension with, the family, which it brings into being in turn.

At the same time, the limits on this mutual involvement – like the temporary wall erected to contain the newly returned initiates in the lelwapa – are equally clear. As Jean La Fontaine notes, ‘Initiation defines boundaries’ (Reference La Fontaine1985: 16). After a man’s initiation, the kgotla acquires a narrow access to him, and through him to his family – a right to call him to service or work and to demand contributions from him and his kin – as do a man’s co-initiates. And the kgotla is drawn into a narrow connection in turn, whereby it may be called upon to resolve intractable family conflicts and disputes. But the kgotla does not, for example, enter into family conflicts without being called to do so (usually as a last resort); it seldom accesses the space of the home at all. Even historically, it did not force initiates to leave paid work or neglect their obligations to plough and harvest in order to undertake the work of the village (Schapera 1955 [Reference Meier, Alber, Martin and Notermans1938]: 110). In this sense, a relationship of relative, voluntary parity is established. As collective ethical subjects, losika and morafe are imbricated but separate; neither one absorbs or supersedes the other. I suggest that it is in reasserting this deep mutual embeddedness and these clear distinctions, and in regenerating this parity, that bogwera achieves the social change sought by the kgosikgolo at the outset, in and through the family – in ways that similar attempts by NGOs and social work offices struggle to achieve.

Having examined how the Tswana family manages its interdependencies and boundaries with the motse and morafe, and how the morafe manages those interdependencies and boundaries in turn, Chapter 15 explores how government ministries, local NGOs, and international civil society and donor groups manage similar dynamics. These agencies foreground idioms and ideals of kinship to naturalise and legitimise their work, to establish relationships among themselves, and to encompass the families in which they intervene. They, too, explicitly and implicitly open up spaces of contestation around alternative ethical frameworks and alternative subjectivities, posed in terms of alternative relationships between the self, losika, and morafe. And, in doing so, they, too, work to establish themselves in and through the Tswana family. But at the same time, they are permeated and driven by an unmarked range of kin dynamics. These dynamics – and their accompanying, divergent ethics – simultaneously animate and frustrate their projects, aligning them with but excluding them from the families they serve, undermining their claims to precedence, and interfering with their projects of social change.

The Deputy Permanent Secretary for Botswana’s Ministry of Foreign Affairs stood at a makeshift podium, incongruous in his sharp business suit among the trees. Flanking him to his right sat a small phalanx of similarly well-dressed officials, suited or uniformed, the women wearing high heels despite the deep sand. To his left ran a long, open white tent, under which a handful of elite personages sat on office chairs at long tables covered in cloth and Botswana-blue bunting, fronted by an impressive display of baskets, gourds, and woven mats. Facing the tent, across an open performance area, three rows of Canadian high school students wearing tailored shirts and skirts of blue German-printFootnote ¹ cloth shifted uncomfortably on small iron chairs brought from a local primary school for the occasion. Everyone else – a crowd of people from the nearest village, including elders, young men and women, and gaggles of children to whom the speaker gestured inclusively but vaguely as ‘the community’ – sat and stood around the edges, behind the ranks of officials and Canadians. Children darted in to check the proceedings, and back out to play in the surrounding bush.

The Deputy Permanent Secretary was outlining the president’s goals for national development, and appreciating the Canadian group for situating their work so well within them. ‘That these students can demonstrate this kind of love and care for other human beings gives me hope that coming generations will inherit a more caring world,’ he continued. ‘I wish to pay a special tribute to the parents of these young people … we hold in high esteem parents who can allow their small children to travel to a far place and live among strangers for a week.’ He spun together development goals, love and care, inheritance, global humanitarianism, parenthood, and cultural exchange as effortlessly as he had envisioned ambassadors in families in his opening lines. His audience listened impassively.

We were an unlikely group in an unlikely spot. We sat in a semi-cleared, wooded area next to a deep, dry riverbed, tucked behind a range of unusual rock formations in a remote corner of the country. A well-respected national NGO had acquired the area as a campsite in which to host its therapeutic retreats for orphaned children. Its programme had been modelled explicitly on the tradition of initiation, which had long since lapsed in most of the areas the NGO served (including, until not long before, Dithaba); a group of children participating together from one community were even called a mophato. But unlike the bogwera undertaken in Maropeng, the retreats were also cast explicitly in funding proposals as a means of ‘creating kin’. I had helped broker the government’s partnership with the NGO in my previous incarnation at Social Services, and I had attended training sessions and part of a retreat in the past. The programme now spanned the country and was being implemented by government social workers in half of the nation’s district councils. It had already enjoyed a long history in Dithaba, where the NGO had been working for years with many of the children and families I knew.

The Canadian students, looking alternately bored and bewildered as the speeches continued, had fundraised to help build a meeting hall – modelled on a kgotla – to be used for ceremonies at the new campsite. They had come for a week to help finish its construction before making a short tour of the country, and an agreement had been struck to mark the occasion with an official opening event. And so a remarkable number of senior civil servants – from the tribal administration and schools in the nearby village; the district council and land board in the main town a couple of hours’ drive away; and the Department of Social Services, the Ministries of Foreign Affairs and Local Government in distant Gaborone – had made their way along the red, sandy roads and down the narrow track that led into the site. Many had come from the capital, a day’s drive away; some had come during the week to camp and help with the work of finishing the site and preparing for the event, much as they might have done for a wedding or funeral. The head of the country’s orphan care programme had even been tasked with chaperoning the Canadian group for their entire stay. As I had enjoyed long-standing relationships with both Social Services and the NGO, and being Canadian too, I was invited to tag along.

The Deputy Permanent Secretary finished his speech and made way for the first of six local choirs performing that day. Dressed in matching T-shirts printed with the choir’s name, they danced and sang their way into the performance area to the shouts and ululations of the audience, some of whom came forward to dance with them in encouragement. The choir, singing a greeting song for bagolo (the elders), initially faced the podium and tent – until an enterprising social worker, no doubt noticing the disappointed expressions of the Canadian contingent, induced them to move so that they could be seen by everyone at the same time. They sang, ‘Modimo, o thusa bana ga ba na batsadi’ – God, help the children without parents. It was the first reference to the children for whom the campsite had been built. The song painted a vivid picture of orphans’ helplessness, vulnerability, and isolation, as well as the threat they posed to the nation’s future. The choir sang boldly and danced energetically, at one point prostrating themselves – as if they were the helpless children about whom they sang – until a well-dressed man came forward from the ranks of dignitaries to drop cash in the dirt in front of them. They refused to go on performing until money had been left by others as well, at which point they gathered it up triumphantly, ululating.

The story I have told about Tswana kinship so far has gravitated around the home, or gae – the expansive, multiple, and interlinking spaces in and between which families and selves are made. As we have seen, social workers and NGOs, and the programmes of intervention they run, have claimed an increasingly prominent role in that context, with mixed success. I have suggested that the work of these agencies and the families they serve adheres to a certain common logic and practice, which links them intimately. Both agencies and families focus their energies on enabling and managing movement, for example; both prioritise building as an important gesture of self-making and kin-making; and both locate care, in part, in the provision of specific sorts of material goods (food, clothing, cash, and so on). Both are concerned with managing the recognition of relationships (as we will see further below); both take the care and circulation of children as a primary responsibility; and both rely on the public performance of success to solidify their relative priority in relation to one another. Given that most social workers and NGO staff or volunteers at the projects I have described are Batswana, share experiences and understandings of kinship with their clients, and are even bound up with the communities they serve through kinship ties, the close alignment between the services they provide and the needs they seek to address should come as no surprise. At the same time, the preceding chapters have detailed how social work and NGO practice serve to disrupt, invert, and muddle Tswana kinship practice in each of the spheres above – knocking it out of sync, stretching or collapsing its boundaries, and in some cases displacing it altogether. These disruptions have been most evident in the sort of dikgang (conflicts, risks, or issues) that arise and in the availability of responses to them. Such disruptiveness is only possible because of the close links of ideology, experience, and relationality that organisations and kin enjoy; but it also speaks to a fundamental divergence.

What generates this divergence? In this chapter, I turn my attention to the dynamics evident within and between NGOs, government agencies, and donors to pursue that question. While the opening ceremony was a singular event, it condensed the attitudes and assumptions that pervade the work of these agencies in Botswana and that animate the relationships among them. It also draws together the trends we have seen in practice in their programmes over the course of this book. Following the clues of their unexpected resonances with kin practice in previous chapters, and the trail of dikgang, I ask whether and to what extent we might better understand these institutional endeavours in kinship terms.

While these institutions may cast themselves as iterations of a recognisably modern, liberal, and perhaps ‘Western’ political project (in the sense used by McKinnon and Cannell Reference McKinnon, Cannell, McKinnon and Cannell2013), I suggest that we might reconceptualise them as being fundamentally informed by kinship ideals and practices, and as being in constant, unmarked negotiation with both. Unlike the morafe initiation, however, the work of these organisations both ignores and rejects the possibility of their interdependencies with kinship. Indeed, in performance and practice, they cast themselves in opposition to kinship and the family, which become corrupt, dysfunctional remnants of an immodern era – requiring the intervention and benevolent guidance of these agencies. And this opposition, like the distinctions made by the morafe, is a question of ethics: it seeks to escape, avoid, or transcend the fraught interdependencies of community life, and thereby offer equal service to all. Assuming the distinctions between the domains of politics and kinship are given, and that the realm of the political naturally encompasses that of the family (Ferguson and Gupta Reference Ferguson and Gupta2002), these organisations focus instead on deploying a kinship idiom to naturalise and depoliticise their claims, to forge links, and to contest hierarchies among themselves. But, as they do so, it becomes clear that the shared, universal terms in which they think they are working are shifting and unpredictable – suggesting both that there may be more than one sort of kinship at stake, and that it may permeate their institutional practice in unexpected ways. Paradoxically, in failing to recognise the imbrications of their political projects with kinship, to negotiate and produce appropriate distinctions between those domains, these modern agencies prove decidedly ‘immodern’ (Lambek Reference Lambek, McKinnon and Cannell2013).

Humanitarian and development interventions have been convincingly described in terms of their anti-politics (Ferguson Reference Ferguson1994; Ticktin Reference Ticktin2011), but seldom in terms of the work to which kinship and families are put in their depoliticisation. I suggest that the family provides a key depoliticising, dehistoricising, and universalising space in and through which an international humanitarian community – a global family – can construct itself (see a description of refugees in these terms in Malkki Reference Malkki1996: 378). As Erica Bornstein noted in her work on World Vision in Zimbabwe, the health and safety of the family mark a universal moral good that transcends national politics, opening up new avenues for NGOs, states, and donors to reconfigure and extend their power (Bornstein Reference Bornstein2005: 97–118). In both the speech of the Deputy Permanent Secretary and the choir’s performance, deploying the discourse of family is a powerful means of downplaying (or justifying) fundamentally political aims. The family provides a powerful metaphor that government, NGOs, and donors can – and do – tap into as a means of naturalising their work, relationships, and power. But attempts to operationalise kinship to further the ends of governance are frequently foiled by the ‘superfluity … and excess’ of kinship (Lambek Reference Lambek, McKinnon and Cannell2013: 255; cf. Ticktin and Feldman Reference Ticktin and Feldman2010: 5). Kinship is, after all, more than a metaphor; and I argue that it features just as powerfully in the daily practice and lived experience of ‘official’ spaces as in their programme delivery. Government and NGO programmes that intervene in the family, attempting to contain and reshape it, are themselves suffused and animated by kinship ideals and practices. These ideals and practices are neither clear nor consistent; they are left unmarked and opaque. In this sense, kinship is as crucial to understanding development and humanitarian programmes as development and humanitarianism are to understanding kinship.

In this chapter, I explore these possibilities by focusing on the ways in which relationships within and among NGOs, government, and international donors are publicly performed and delimited. I argue that the ceremony described above simultaneously enacts multiple notions of kinship; and I suggest that these multiple notions have also been contested and at work in the NGO and social work office described in previous chapters. This multiplicity exacerbates the superfluity of kinship, which tends to overwhelm, outstrip, and evade the constraints imposed by both workplaces and bureaucratic systems. Keeping this multiplicity in mind, I ask whether kinship can be ‘encapsulated in and by the state’ (Lambek Reference Lambek, McKinnon and Cannell2013: 257; see also Ferguson and Gupta Reference Ferguson and Gupta2002 on assumptions about the state’s encompassment and verticality) and by other transnational political agencies; or whether it not only permeates but also generates and animates those agencies.

The choir finished its rousing performance, weaving its way off the sandy stage and singing until its members broke formation and dispersed among the audience. From the podium, the master of ceremonies thanked them with great enthusiasm and warmly welcomed the lead teacher of the Canadian school group to speak next.

The lead teacher was a contentious figure, having offended many government and NGO representatives over the course of the week with his brash, demanding manner. The previous day he had insisted on separating water for his students from the water supplied for everyone else, suspecting theft; senior government figures watched with bemused resignation as he first berated the NGO director and then instructed his students to relocate dozens of water bottles from the kitchen into their tents. Now at the podium in his custom-tailored German-print shirt and a baseball cap, he consulted with the translator to ensure that he would be translated phrase by phrase. After speaking about what the retreat campsite – which he framed as a ‘humanitarian project’ – represented for bonds between Botswana and Canada, the teacher thanked the host NGO and government departments and ministries in a perfunctory, non-differentiating fashion. He added offhandedly, ‘We consider everyone here to be like surrogate parents for us.’ The translator followed with ‘Re le tsaya jaaka batsadi ba rona tota tota’ – we take you like our real, real parents.

He then called all 29 of his students in front of the podium – although it meant that their backs were to the dignitaries and most of the community, and they faced only the VIPs under the tent – and presented them as the best Canada had to offer. They were a visibly mixed group, as the line-up was meant to emphasise, of largely South Asian, South-East Asian, Chinese, and mixed European descent. He intoned: ‘A country without its culture is lost.’ It was an accidentally apt echo of the words of Botswana’s first president, Seretse Khama, who warned that ‘a nation without a past is a lost nation, and a people without a past is a people without a soul’ – a sentiment that has shifted to incorporate a warning against the loss of culture instead of just the loss of history (Parsons Reference Parsons2006; see also Dahl Reference Dahl2009b). Indeed, a similar sentiment pervaded the revival of initiations back in Maropeng, as well as the NGO’s own initiation-oriented model. Attached to such a diverse group of children, however, from a place no one knew much about – but that presumably had greater prosperity and fewer social ills to cure – it caused obvious confusion. The teacher elaborated a vision of what defined Canada as a nation: multiculturalism, a history of peacekeeping instead of war, the assurance of equality for all. ‘We teach our children to celebrate other cultures and values,’ he explained, describing his students as the future leaders of Canada. He added: ‘They are an example of what youth should be throughout the world … committed to making change.’ The students tried to look grave and inspiring. Behind them, many in the crowd looked politely baffled. On the one hand, it seemed, the audience was being encouraged to preserve their culture; on the other, they were being encouraged to adopt a rather inscrutable but ostensibly successful Canadian model. On the one hand, these children had respected and taken their hosts as parents; on the other, they seemed to suggest that parents were incidental or unnecessary to the exemplary individuals these children had already become. I thought back to the teacher’s comment to his students late the night before, which I had overheard from across the campsite: ‘I’ll be honest with you, I don’t really care about Botswana or Botswanans or whatever. The important thing here is you guys, and the experience you’re getting.’

The Canadian teacher stepped down from the podium, leaving it to the last and most highly ranked speaker – the Assistant Minister of Local Government. His ministry oversaw everything from Social Services to district councils and village kgotla administrations. He made his way out from under the VIP tent, dressed in sharp khaki trousers and a multi-pocketed photographer’s vest and flashing a good-humoured smile. He waved away the translator jovially and settled in at the podium, beginning with an unexpected injunction: ‘I would like to invite you all to rise, and observe a moment of silence for those orphans we have lost to HIV and to abuse.’

His sombre invitation – in English – caught us all a little off guard, although we rose dutifully and bowed our heads. Indeed, for all my years of attending such ceremonies and events, I had never heard such a discursive combination of catastrophes. Holding orphans up for pity over the loss of their parents and the assumed neglect of their overburdened families, and rallying cries to rescue them and the future of the nation, constituted the usual rhetoric. But in the context of successful, free programmes for the provision of ARVs and the prevention of mother-to-child transmission, orphanhood was seldom posed as a cause of HIV infection, and links between orphanhood and death were virtually never made. While abuse was connected with orphanhood frequently enough and had become a major focus of social services discourse, I had never heard it connected to death either. The request for silence was unsettling in the complexity of social ills it subsumed; more than that, it was jarring in its dislocation from the reality to which most of us in the audience were accustomed, in what felt like a dramatic inflation of the stakes of orphanhood in particular.

After the silence, the Assistant Minister continued for a while in English, congratulating the Canadian students, and their parents, for the spirit of love and giving they had shown, and calling upon all present to learn from their example. He did not bother to translate. Before long, however, he had shifted into Setswana – and he began a different speech altogether. The exhortative thrust of this parallel speech was kgokgontsho ya bana, child abuse, and on this topic the Assistant Minister spoke at great length, with great conviction and passion. He confronted his audience: ‘Child abuse is there in our homes and families, though we are turning a blind eye to it and pretending it is not. Men! Uncles! Check yourselves! Check yourselves, look into your hearts.’ It was the deliberate echo of a nationwide HIV and AIDS behaviour change campaign launched a few years previously, dubbed Oicheke! – Check yourself! (USAID 2010). ‘We appreciate these Canadian children for coming to look after our children,’ he continued, still in Setswana, ‘but we have a responsibility to look after our children too, so that one day they might go to Canada to help children there, or even to any other place in the world.’ He did not bother to translate this part of the speech either.

It was a spellbinding oration. And yet the audience did not look altogether engaged. The ranks of community members listened attentively but wore bland expressions. Children continued to run in and out, and choir members joked with one another on the sidelines. The Canadian contingent had begun to glaze over; most looked bored and a few looked frustrated, or perhaps offended. Just at the point when he had almost lost them, the Assistant Minister switched back into English – to describe his hope that, one day, one of the Canadian students before him would meet a doctor on their travels and find that she had grown up in Botswana; had attended a camp run in the very place they sat now; had come to grips with her loss and grief, had found hope, a sense of self and direction, and had made something of her life. The students lifted their heads, and some began to smile warmly. They were, of course, unable to decipher the strange double register that had emerged: in Setswana, families were abusive, irresponsible, corrupted, and broken; while in English, they were sources of love, giving, and hope for the future.

Shortly after the speeches finished, the cooks and several volunteers from the village nearby called the Canadian students to help serve up the enormous meal that had been prepared – a gesture of inclusion that befitted children and young people at such a gathering. Their lead teacher was outraged, refused his meal in protest, and insisted that they all sit and allow themselves to be served like the VIPs, as he felt befitted respected guests. Everyone dispersed soon afterwards, the community members walking up the dusty road back to their homes and the government officials heading off in convoys of white four-by-four trucks. I learned later that the event, and the Canadians’ week-long visit, had in fact cost the host NGO in Botswana more than three times as much as the students had fundraised – running into hundreds of thousands of pula. It cost Social Services as much again, in officers’ hours, petrol, food, and so on; and both the Ministry of Foreign Affairs and the district council would have had similar bills. I was shocked, but my friends at Social Services and the NGO shrugged it off. ‘If someone was giving me only five pula I would still do everything to appreciate them,’ one insisted.

The speeches recounted above show how discursively entangled the family is with the state, and with projects of development, humanitarianism, and international relations – a notable contrast to the careful way in which the morafe distanced the mophato from family, in discourse and performance. At the opening ceremony, community, national, and international relations were all – often awkwardly – cast in the idiom of family, with a special emphasis on parents and children. International diplomacy was framed as a familial fostering of ambassadors; humanitarian work was cast in terms of love, care, and the inheritance of future generations. The NGO took as its explicit mission the creation of kin for and among orphans, implicitly replacing lost parents. The Canadian students were thanked in part through their parents; acknowledged their hosts as parents; and were appreciated for helping raise Batswana children – a network of relatedness within and against which they then defined their culture and nationhood. As Elana Shever notes of national sentiments – to which we might easily add humanitarian and development sentiments more broadly – they ‘rest on a trope of familial bonds as the authentic basis for solidarity, care, obligation, and sacrifice’ (Shever Reference Shever, S. and F.2013: 88). And this trope worked to refigure an otherwise distinctly odd combination of institutional characters in Botswana’s backwoods, loosely and temporarily bound together by circumstance, as natural, unified, and enduring.

At the same time, these discursive formulations worked to separate the event’s participants and to establish the terms on which they could relate. As Didier Fassin notes, compassion performed in public spaces is ‘always directed from above to below’ (Reference Fassin2012: 4), both presupposing and reproducing inequality. The sharpest separation made was between the NGO, government ministries, and Canadian students on the one hand – sources of care, love, and compassion – and the families in attendance, whose lives these figures sought to protect, on the other. This performance, and others like it, ‘was more of a theater for politicians than “for the people”’ (Bornstein Reference Bornstein2005: 112), a matter of contesting institutional hierarchies in which ‘the people’ were always already at the bottom. Thus, the Assistant Minister cast aspersions on his entire Setswana-speaking audience by purporting to publicly expose the abuse in their homes, upbraiding them collectively for their inability to look after their own children as effectively as the Canadian students – themselves children – could. The Tswana families (especially their men, and bo malome) were thereby infantilised, cast beneath the protective elderhood first of the juvenile Canadian contingent, and second of the government and NGO agencies that recruited the Canadians’ assistance. The Canadian teacher’s speech, while accepting the group’s Tswana hosts as surrogate parents, underscored this infantilisation by emphasising the students’ superior agency in addressing issues that afflicted the community.

Meanwhile, both the Assistant Minister and Deputy Permanent Secretary – when speaking in English – were careful to position themselves and their agencies as the equals or elders of the Canadian group, whether thanking the students through their parents or positioning themselves as temporary parents. The insistence on appreciating the Canadian contribution no matter the expense required was, I suggest, a similar assertion of independence and equal agency, and an active refusal of the implicit hierarchies that emerge in gifting and international aid – a corollary to what Durham (Reference Durham1995) describes as the spirit of asking, and a means of absorbing gifts that have not been asked for (see Stirrat and Henkel Reference Stirrat and Henkel1997 on how development gifts reinforce difference and hierarchy). And both of the government keynote speakers deployed parallel professional discourses – one framed around international relations; the other in terms of social work assessments of societal dysfunction and its remedy – that reinforced this claim to equal consideration by establishing a suitable distinction between the corrupted, suspect realm of the family and the advanced, modern realm of the state. As China Scherz notes in reference to the model of sustainable development more broadly, this professionalisation allowed agencies to ‘imagine themselves as separable and separate from those living in the places they work’ (Scherz Reference Scherz2014: 8) – a hallmark of their modernity and their alignment with prominent global expectations in development work. This distinction echoed those made by the Canadian teacher, whose reference to family was peremptory and quickly superseded by a lengthy rumination on the Canadian nation, establishing common ground among the speakers and their agencies from which the families in whose mould they had earlier cast themselves were explicitly excluded. All of the speakers, in other words, were engaged in a form of ideological boundary-making work in separating the realms of politics and kinship (McKinnon and Cannell Reference McKinnon, Cannell, McKinnon and Cannell2013) – although, unlike bogwera, there was no room for interdependence with kin, much less the potential for voluntary parity.

These discursive deployments and repositionings of kinship are typical of a social welfare, development, and humanitarian genre as well as being familiar ways of speaking about the state. To the extent that they organise means of relating, however, they are more than simply metaphorical. Indeed, a closer look at the unfolding of the event demonstrates uncanny parallels with kinship practice and discourse. Echoes of the family feast – itself reminiscent of wedding celebrations and of the feast we saw in Chapter 13 – are perhaps most obvious: the white tent, housing bagolo (elders) around which the event was oriented (here government ministers instead of parents); the arrangement of celebrants around an open lelwapa-like space; the speeches, introducing key figures in terms of their relatedness to one another; and the collective contributions of money, goods, and work appropriate to a celebration, for entertainment, and of food sufficient to feed a village of guests. Like the family feast, the opening ceremony sought to perform the success of key figures – NGO, ministries, and Canadians – and the generative power of their relationships, while attempting to extend that success and remake those relationships in clear ways that distinguished them from the invitees.

Echoes of other dimensions of kinship practice are evident, too, including all of those we have seen throughout this book: geographical scatteredness and the mobilisation of movement, gravitating to a shared space of care work and contribution; the careful management of visibility, speech, and recognition; the anticipated circulation of children to the campsite for therapy, which was modelled explicitly on bogwera; and so on. But perhaps most significantly, dikgang were produced throughout: around imputations of stolen food and water; refusals to share, help serve, or eat; the public dressing-down of NGO organisers or purportedly abusive families; and many more besides – all of which echo dikgang we have encountered elsewhere, and draw the performance of relational success into question. Where dynamics of dikgang have previously highlighted limits on the ways in which social workers and NGO staff relate to the families they serve, here they suggest a performance of relatedness among rather unusual actors: national government, local government, international donors, and local NGOs. Indeed, we might even discern an attempt to create a collective, ethical subject (Lazar Reference Lazar2018: 268) in the process, one like the family, or indeed the morafe, interlinked and hierarchised, able to self-produce and reproduce. But, if this process is afoot, it is a different sort of ethics at work. It may provoke a collective reflection on who has done what for whom, through which specific relationships and relative seniority are asserted and recalibrated; but it takes the larger question of the correct relationship between self, family, and polity – which was at the heart of the ethics of initiation – as given, a natural matter of verticality and encompassment.

The Tswana family, meanwhile, is marginalised from this process, destabilised, even demonised. Parents and children sit on the edges of the ceremony, moving in and out; unusually, they have no real role to play in the proceedings. The only mention made of them is either in terms of orphans having lost parents to disease or in terms of the collapse and corruption of their relationships, beset by death, loss, abuse, and the constant threat of harm. While appreciation is afforded the Canadian students and NGO for their help, it is the Tswana family that bears the blame and responsibility for its own dissolution. Everything is done for them, but they have done – and can do – nothing for themselves or for the agencies that offer this withering vision. What families may have done for one another is obviated; the standard to which they are held here is one of international rights discourse and the self-improvement imperatives of sustainable development (see Scherz Reference Scherz2014 on the ethics of sustainable development in Uganda and similar dissonances with Baganda ethics of patronage).

In discourse and practice alike, then, it seems that both the state and NGOs are involved in processes that we have seen to be characteristic of Tswana kinship – but in ways that are more about legitimising themselves as political entities and navigating their relationships with each other. They are engaged in a process of state-making, or NGO-making, or perhaps the making of a shared public sphere, through family and kinship processes but also against them, and in ways that exclude actual families. Their legitimacy is modelled on kinship, justified by their intervention in actual families and enacted in kinship idioms, practices, and ideals; but it is geared towards navigating relationships with other ‘super-familial’ actors, at local, national, and transnational levels, where relative influence is highly contested (Bornstein Reference Bornstein2005: 98–9). And this disjunction is especially apparent in the different ways in which dikgang are identified and addressed. As distinct as the spheres of development and humanitarian policy and practice may be (Mosse Reference Mosse2004b), they are thus bound in part by an idiom and logic of kinship. Paradoxically, their deployment of that idiom and logic separates and excludes them from the sphere of the family, over which they attempt to assert authority but to which they enjoy little real access, which means that their programmes are often beset by failure and frustration.

What is the logic of kinship that seems to bind these actors? In the speeches above – as in the disjunctions evident between social work offices, NGOs, and families ‘on the ground’ – a certain mutual misunderstanding seems to be at work. While the Canadian head teacher imagines his hosts as ‘surrogate parents’, for example, his translator understands them as real parents; the links the teacher makes between individuals, culture, and nations against that backdrop visibly perplex his audience. The Assistant Minister’s assessment of family breakdown, and his moment of silence for ‘lost orphans’, strikes a similarly confusing note. While these speakers assume a shared understanding of the biological realities of relatedness and the social relationships they underpin as indisputable ‘facts of life’, with clear epistemological and moral implications (Pigg Reference Pigg, Pigg and Adams2005), this assumption doesn’t quite hold. I suggest that these moments of misunderstanding result from a proliferation and confusion of different notions of kinship at work in the discourses above, and in the intervention practice we have observed. The speeches above weave together, take apart, and move between what we might identify as Tswana and Canadian – or at least Euro-AmericanFootnote ² – understandings of kinship, familiar enough to one another to be mutually recognisable, but disparate enough to be jarring. In this sense, it is worth considering political institutions as ‘site[s] of contention … between competing normative ideas’ (Bierschenk and Olivier de Sardan Reference Bierschenk and Olivier de Sardan2014: 6) of kinship as much as of governance or bureaucracy.

A strongly Euro-American notion of kinship emerges from the very beginning of the ceremony. The Deputy Permanent Secretary of Foreign Affairs cast families as a background, contextual device for the production and reproduction of individuals and nations – prioritising the individuality of persons (Strathern Reference Strathern1992: 10–11). The Canadian lead teacher replicated this discursive technique, perfunctorily appreciating the group’s Tswana hosts as ‘parents’, effacing the students’ own families, and then presenting the youth as successful, agentive individuals, able not only to represent but to reproduce both their own nation and the nations of others. The Assistant Minister, too, in both his English and Setswana speeches, emphasised individuality as the key experience and aim of kinship. He individuated orphans first of all, cutting them off from their families in a way that explicitly prioritised their relationships with their biological parents over any other relatives (Strathern Reference Strathern1992: 12); he portrayed uncles and others outside the parent–child binary as the most insidious figures of the family; and he personalised responsibility for abuse, while suggesting that it will produce abusive individuals in turn. Indeed, having chosen to come halfway around the world to help other people’s children, and having enacted that commitment in a wild, isolated space – notably, in the absence of those children and their families – as an individual enterprise oriented mainly to their own growth, the Canadian students were bringing to life many of the fundamental imaginings on which English kinship is based (Strathern Reference Strathern1992: 12–13): choice, isolation, nature, and, above all, individualism.

What I have glossed as the Canadian or Euro-American imagination of kinship is not, of course, entirely divorced from the Tswana notion of kinship, and links emerge at several points. These connections give the impression that everyone is referencing the same, universal notion of kinship, while also producing the distinct jarring noted above. So, for example, although an emphasis on the parent–child relationship would have felt familiar and ‘natural’ to Canadians and Batswana alike – since Batswana reframe a variety of relationships, including siblingship, in these terms, and since it is the critical nexus for biologised and emotional concepts of Euro-American family relationships as well (Schneider Reference Schneider1980) – the sense of mutual recognition it provides is quickly undermined by the stakes it represents. Thus, in Euro-American articulations of kinship, the parent–child relationship most strongly evinces uniqueness and individualism (Strathern Reference Strathern1992: 12); but in Tswana articulations, it is taken to underline lasting responsibilities of care, intersubjectivity, and mutual dependence. For the Canadian students, the parent–child relationship is fixed, given, and linked uniquely to birth (Schneider Reference Schneider1980); for Batswana, it is multiple, fluid, and linked to responsibilities of care, which may be applied equally to siblings, spouses, or other relationships.

This simultaneous familiarity and divergence also applies to references to love and care. Both Canadians and Batswana emphasised these qualities and used these words in English; both groups recognised them as key concepts in their understandings of kinship; and both assumed that they shared a common understanding of the terms. However, in Frederick Klaits’ thorough description, the Tswana association of love with lorato involves ‘action and sentiment directed toward enhancing the well-being of other people’ (Klaits Reference Klaits2010: 3); it involves ways of speaking and acting that work in people’s bodies (Durham Reference Durham and Klaits2002a: 159). Care, or tlhokomelo, emphasises the provision of material goods and work (Klaits Reference Klaits2010: 4). Both of these terms have sentimental dimensions, but they are expressed and generated in bodily, material, and work-oriented ways. The dominant tone of these terms for the Canadians, in contrast, is more likely to be emotional and private (Strathern Reference Strathern1992: 12) rather than materialised or enacted; and it will likely have been clearly separated from work (Schneider Reference Schneider1980).

What become clear in these observations are the fluid, almost invisible ways in which the Batswana speakers in particular shifted back and forth between Tswana and Euro-American understandings of kinship. This subtle shifting, I suggest, is indicative of the multiple ways in which Botswana’s government policy, social workers, and NGO staff see families; and of the extent to which these different visions grow out of fundamentally different ways of being family. The ways in which social workers and NGO staff see their clients show strong elements of Tswana notions of kinship, but they also show strong Euro-American influences. This combination is perhaps unsurprising: the Ministry of Local Government, under which the Department of Social and Community Development operates, is a survivor of the colonial era, and many of its acts and policies – including a particularly outdated one on adoption (RoB 1951) – hark back to that time. So, too, do the principles that underpin those frameworks. The curriculum for social work taught at the university was also of British inspiration aligned with international standards of social work. And, of course, the work of social workers and NGO staff is framed by international conventions, policy frameworks, and ‘best practice’ promulgated by the United Nations and prioritised by European and American development and aid agencies, with a bent towards Euro-American ideals of kinship (see Mayblin Reference Mayblin2010 on international conventions on child labour). The ethical register in which NGOs and social workers assess Tswana families, then, is by necessity an assemblage of the sort described by Scherz (Reference Scherz2014) for Uganda, entangled with quite different notions of what families are and ought to be, and with the political-economic contexts in which those notions have changed and unfolded over time.

Where kinship seems to provide a common basis of mutual understanding – a natural, shared ideal, a common emotional register, a familiar set of practices, a ‘fact of life’ (Pigg Reference Pigg, Pigg and Adams2005) – it instead provides a multiplex, muddled, and contradictory field of experience. In this sense, kinship describes a powerful but unstable register that simultaneously binds together and fractures the political, institutional realm. Kinship both saturates and evades the political, not because it taps into a naturalised, universal process, but because it doesn’t – although these political projects expect it to do so. Where kinship is invoked to naturalise and stabilise institutionalised claims of power, its multiplicity and excess instead makes them awkward and unnatural, and destabilises them. Kinship, then, does not simply escape or overwhelm bureaucratic attempts to contain it; it drives those attempts, permeates their logic, and disrupts their practice from within, rendering them ineffective for reasons that are difficult to grasp. And it is in this sense that I suggest kinship may be understood to generate and animate the purportedly modern, liberal political spheres of governments, NGOs, and donor agencies alike. Not only is the village in the home, but so too are a global array of political communities.

As Jacques Donzelot (Reference Donzelot1979) has shown in his history of ‘policing’ and philanthropy in France, political actors – including state and non-state agencies – have long prioritised access to families and provision for their welfare as key means of extending, stabilising, and reproducing power over time. I suggest that a similar project has animated the work of local, national, and transnational political actors in Botswana since at least the colonial era. Tswana families, in their turn, are constantly working to acquire and incorporate new resources and relationships, to enable the self-making of their members and the reproduction of kin groups over time. In Botswana’s time of AIDS, NGOs and government are important sources of those resources and relationships. The family and the state, NGO, or foreign donor are thus deeply reliant upon and implicated in each other; each establishes its relevance and sustains its growth through the other. But each also poses risks to the other that require containment and management. Efforts to generate social change find traction if they tap into this ‘immodern’ (Lambek Reference Lambek, McKinnon and Cannell2013) interdependency, and can create the distinctions that enable a collective ethics; those that reject that interdependency, falling back on assumptions that the political sphere is naturally distinct from and encompasses the domestic, struggle to do so.

The kgotla’s attempt to create social change explicitly in and through families, on one level, resonated with the attempts of NGOs and state actors to do the same. But the bogwera enacted a deep interdependency between the family and the polity, as well as marking sharp distinctions that reinforced the capacity of families to engage and resolve dikgang – rather than taking over, blocking, or frustrating that role. Part of what was being reclaimed in the initiation was a particular relationship between the self, losika, and morafe, the shape and limits of which had been blurred over years of increasing programmatic interventionism on the part of competing public agencies. Mobilising the ethics and practices of kin-making allowed the kgotla to regenerate the collective ethical subject of the morafe through a regeneration of the collective ethical subject of the losika, simultaneously reasserting both as subjects ‘that can take action on the world in order to transform the world’ (Lazar Reference Lazar2018: 268).

By contrast, at the ceremony described above, the families of the motse, or village, were ranged around the outside of the event, an undifferentiated mass of variously engaged witnesses to the agencies’ main act. While the NGO’s attempt to recreate initiation in its programming showed an awareness of the transformative potential described above, in the event, the government ministries and Canadian students seemed instead to take these families as context and backdrop: a potential challenge, an audience to whom exhortations might be made and for whom responsibility must be borne, but an entity marginal to the performance itself. While these families were a source of far-reaching dikgang, in the speeches of the opening ceremony they were denied the capacity to engage those dikgang and regenerate the collective ethics that might address them, while restoring their relations as kin. The agencies being celebrated implicitly retained that power for themselves. And yet, it was these very families – and the shadow audience of Canadian parents behind them – against, through, and within which that performance was defined, and to which it was oriented. It was those very families – and the diverse and contradictory range of kin practices and ethical engagements they involved – against, through, and within which the everyday work of those same NGOs and ministries was conducted. Just as we found the village and morafe defined against, through, and within the family, here we find a transnational array of political agencies unexpectedly defined in the same way. But in ignoring the imbrication of their politics with the families they serve, the efforts of these agencies to shore up their power and to create social change through their collaborations fall short.

It is not simply that powerful national and transnational political, economic, religious, or other forces are exerting unidirectional influence on the Tswana family and creating upheaval – as Schapera (Reference Schapera1940: 346–57) claimed in the colonial era, and as development and humanitarian discourse suggests now. And it is not simply that the Tswana family is evading those influences or exerting its own counter-influences. Rather, the ‘domains’ of family and politics (McKinnon and Cannell Reference McKinnon, Cannell, McKinnon and Cannell2013) – produced in governance, development, and social sciences discourse – are intrinsically interdependent, in practice as much as in idiom: each can only be meaningfully and fully understood in terms of the others. This interdependence becomes especially clear in the context of dikgang, where the moral and ethical terms in which those domains are established is up for grabs, open to interpretation and reflection. Given that the distinctions made between domains underpin the production of collective ethics, and in turn the ability to act upon the world, both these interdependencies and distinctions ought to be key objects of ethnographic enquiry when attempting to understand the production of social change.

Anthropological analyses of development, humanitarianism, and public health have tended to ignore the family, taking for granted that the domestic is distinct from and incidental to the political, and reproducing that distinction in turn. And yet families are a key sphere in which humanitarianism, development, and public health concerns inevitably converge (pace Redfield and Bornstein Reference Redfield, Bornstein, Redfield and Bornstein2011: 4). Families are targeted by such a diverse and vast array of interventions in part because they provide a context, discourse, set of practices, and ethical framework through which the states, NGOs, and other agencies that run those interventions can produce and reproduce themselves, while simultaneously elevating themselves and naturalising their power and their relationships. The hierarchies generated between family, NGO, and state in turn provide a framework for reproducing, depoliticising, and naturalising global inequalities between nations. At the same time, kinship practices, ideologies, and ethics are shifting, in constant reformulation, and they saturate the work of these agencies in ways that blur and alter the distinctions those agencies seek, invert and denaturalise the hierarchies they assert, and ultimately disrupt the work they undertake, in part by excluding them from families. These dynamics account, in part, for the unintended consequences for which such interventions are notorious; any serious attempt to make sense of the complex legacies of intervention – especially in contexts of crisis – requires that we expand our frame of reference to incorporate the family accordingly.

***

INTERLUDE: The Incident Concluded