Lorato and I leaned against our square-edged spades, looking out across the dry, yellowed patches of farmland to the brick-red hills beyond. The afternoon heat was merciless and the landscape shimmered with it. We had been clearing a rocky, steep slope at the top of Lorato’s plot of the plant life that had colonised it over the years, in preparation for digging the foundation of the house she would build there.

The plot sat high on the slope of a hill that separated it from much of the rest of the village, and it commanded a rare view. It had belonged to Lorato’s mother Keitumetse, who had begun developing it years previously, not long before her death. Close to where we stood, the contours of a foundation trench could be discerned in the tall grass, partly backfilled over the years with gravel and stone swept down the hillside by the rains. After Keitumetse’s death, Mmapula had made a point of transferring the plot into Lorato’s name – an uncommon gesture at a time when family squabbles over the inheritance of land and property were rife. Meanwhile, a few stacks of unused cement bricks, window frames, and other material that Keitumetse had acquired for building had been taken back to the family plot – a 20-minute walk away – and incorporated into its continuous building projects.

Several years had passed, and, as Lorato entered her mid-twenties, the local land board had begun to put pressure on her to develop the land – or lose it. The Ministry of Lands and Housing oversees land boards in every district; their role is to manage the land of the local morafe, or tribal polity. Historically, this role had been undertaken by village chiefs, who apportioned land to their headmen, who in turn distributed plots so that recently married men could settle among their paternal kin (Schapera Reference Schapera1940: 95). After independence, this function was centralised at district level, and land tended to be apportioned in a more geographically arbitrary way than before, depending on which areas of the village the land board had marked for expansion and development. Currently, both men and women, married or otherwise, can apply for plots, and Batswana can apply for plots anywhere in the country, regardless of morafe. Building, in turn, is no longer simply about establishing a marital home near the husband’s kin; it is also about opportunities to move away from one’s parents and siblings, whether one is married or not, to live independently, and even to generate income through rentals or resale (see Griffiths Reference Griffiths2013 for further detail on these trends).

When I first lived in Dithaba, new plot owners bore the responsibility to mark the corners of their plots with fenceposts (an echo of precolonial practices of marking off land with ‘doctored’ pegs; Comaroff and Comaroff Reference Comaroff and Comaroff1991: 134). Then, within five years of taking possession, they had to fence their plots fully and build at least one structure – even an outhouse. But demand for plots skyrocketed in the village, especially as people from around the country sought places to live, rent out, or develop and resell within commuting distance of the capital. The growing availability of mortgages sped up the hitherto slow process of building, and also drove the commoditisation of land. The standards of what constituted ‘development’ accelerated proportionately. By the time Lorato started her building project, the plot had to be fenced and a full house had to be under construction to prevent the land board from simply reassigning it to someone else when her five-year window of opportunity expired.

Mmapula was quite concerned that Lorato should retain the plot and had set aside a small amount of money from her farming income – an amount roughly equivalent to the building supplies she had acquired at the time of Lorato’s mother’s death. It was unlikely to go far. Lorato herself was equally concerned. ‘It is the only thing I have left of my mother,’ she reflected, with a note of discomfort at her admission, a brief and rare articulation of her loss.

The situation had started to come to a head while I was staying with the Legaes. The unspoken request in Lorato’s and Mmapula’s accounts of the plot was no less plain for its omission. After much weighing up of options, reflection, and consultation, I offered to help finance the building through a series of loans, partly sourced from family and friends. Once built, we agreed, the house could be rented out until the loans were repaid. Having recently landed a short contract post with the government, Lorato committed to contributing a significant proportion of the funding. The money available, however, was still not a great deal, and the only way to build the house affordably was to do as much of the work as possible ourselves. By the time we stood taking in the view, we had already been digging and hauling truckloads of river sand for making bricks at home, and we would spend much of the coming months lugging cement, quarrying dense pit sand, ferrying water, and backfilling concrete as the house progressed. We were sometimes helped in these heavy tasks by the Legae sisters and often by the children of the yard; of the brothers, only the youngest, Tuelo, assisted – and only on condition of being paid.

We commissioned a neighbour, Rra Ditau, with the building of the house, and he saw it from its design stages through to the finished structure. Already well into his fifties, he lived close to the Legaes and had built the house I stayed in. In his gnarled, worn-out work boots, his green workman’s trousers, his torn shirts, and the soft hat slung back on his forehead, he looked like any other piece labourer in the village. But he had a contemplative gaze, a habit of speaking in riddles, and a sideline as a poet and musician, which gave him an air of philosophical wisdom. He was fond of asking imponderable questions, looking askance at his befuddled listeners, and laughing heartily before changing the topic.

Lorato retreated into the shade of two stunted trees, and I followed. Rra Ditau, who had accompanied us for the clearing, resumed his fight with the recalcitrant weeds, his spade clanging and jarring against the stones.

‘You think I can get married now, if I have my own house?’ Lorato asked, pensively. She enjoyed surprising and provoking people with such questions, but this time she sounded contemplative, as if she had surprised herself. I didn’t see the connection, and asked why it would matter.

‘Ah, you know these men,’ she said, partly contemptuous, partly resigned, as she gazed out at the lands. ‘They want to be the ones who give you everything. They don’t like this idea of women having their own things, their own jobs, their own money. And imagine, a house! Actually, I might not even live here. A man would want me to live at his place.’

I was quiet, puzzling over whether I had inadvertently created a problem by trying to help (a niggling doubt familiar from years of work in the development sector). It was traditional practice for a man to take his wife to live in his natal neighbourhood or village; Batswana are customarily virilocal, and the administrative subunits of villages – wards – had historically marked off extended virilocal families (Schapera Reference Schapera1940: 95). At the same time, in historical practice, a substantial proportion of couples stayed with the wife’s family while waiting to build for themselves (ibid.: 97). Marriage preferences for parallel as well as for cross-cousins, at least in principle, created the possibility of such an entangled field of relationships that a man and wife (and their families) might be related in several different ways at once in any case – making the question of whether they were living virilocally or uxorilocally potentially unclear and prone to variation (Comaroff and Comaroff Reference Comaroff and Comaroff1991: 132).

These days, of course, it is common for a married couple to settle away from both of their natal homes, depending on where work and opportunity can be found. And the practice of settling and building elsewhere is not altogether new: Mmapula and Dipuo had settled away from their natal homes in a nearby town, after all, first at the lands and later in Dithaba itself. Indeed, most of the married couples I knew lived away from both spouses’ natal homes, and many lived apart – even on opposite sides of the country – depending on where one or the other was posted for work. But regardless of where married couples lived for work, they generally still built in the husband’s home village – ko gae – as well.

Many people I knew – men and women alike – had not yet married by the time they began building, although most of them had had children (something we’ll return to in Part III). A house was an asset against hard times, I reasoned to myself, a place to begin a family, a potential source of independence and income; but did these things in themselves inhibit marriage?

The unanticipated social repercussions of building didn’t end with marriageability. A few nights later – helping us offload a truckload of river sand, down to the last grains caught in the ridges of the truck bed – Rra Ditau put his finger on another. We had been discussing a growing unwillingness among the adults at home to loan us the truck for building work, in spite of our having borne much of the cost of its maintenance and upkeep. Unusual claims had been made, including the suggestion that various items we had to buy in town wouldn’t fit in the truck bed (although we had transported similar items before). ‘How do you think they feel,’ Rra Ditau asked Lorato, in his quasi-rhetorical way, ‘about the fact that you are building first, before they do?’ ‘Haish! Ke kgang akere,’ she had answered, shaking her head – it’s a problem, isn’t it? Only Moagi, Lorato’s mother’s younger brother, had already finished building a small house of his own, as well as the one in his parents’ yard in which we stayed. The eldest brother, Lorato’s malome Modiri, had swapped his plot for a combi-van; another of the younger brothers, Kagiso, was on the endless waiting list for new plots. In fact, Kagiso had exerted some pressure on Lorato to give him her plot to build on not long before. Oratile and Kelebogile, Lorato’s mother’s younger sisters, had plots of their own but no houses yet. When Kelebogile had tagged along with us to see the progress of Lorato’s house, she had been disparaging: ‘You’re only at window height! You still have so far to go!’ Reflecting on these tensions, Rra Ditau laughed his philosophical laugh. ‘Well,’ he said in a non-committal way, ‘I guess you’re killing them at home. But you have to build for yourself.’

We dropped Rra Ditau back at his yard that evening and went in to greet his wife, who was cooking fat cakes in a deep pot of oil over the fire. We sat on one of the long benches against the stacks of old four-and-a-half-inch bricks that gave rough, low walls to their isong. Mma Ditau was congratulatory about the building project. ‘You are becoming a woman now,’ she affirmed to Lorato, smiling. ‘You are becoming a person!’ Lorato was sceptical and asked why building conveyed such sudden status. ‘To have your own yard where you decide what to eat, people take you seriously!’ Mma Ditau explained, bending to examine the fat cakes in the hot oil. Lorato herself – like many others I knew who had begun to build – had often framed her dreams of having her own house in such terms: being grown up, being free of the constraints and conflicts of home, and being able to eat what she liked. When she wanted to illustrate to people just how adult, independent, and self-directed she was, she often said, ‘Ke a aga’ – I’m building – which invariably earned her reactions of surprise and respect.

But it was a burdensome dream. ‘I’m too young to be taken seriously. I don’t want people to take me too seriously!’ Lorato exclaimed, looking dismayed. Mma Ditau laughed generously.

Building a house is a considerable achievement: a testimony both to the material resources and to the personal relationships that one can mobilise for the task. Batswana have long considered it an achievement fundamental to developing as a person, independent of – if still bound to – one’s natal family (Schapera Reference Schapera1940: 103), and to founding a family of one’s own. Go aga lelwapa means to build a house and to build a family, after all. The Setswana verb for building, go aga, echoes etymologically in the words for peace, harmony, and reconciliation (kagiso, kagisanyo, agisanya – see Klaits Reference Klaits2010: 31), each of which in turn connotes helping one another to build. Building relies heavily on a range of relationships and materially instantiates and perpetuates them (Morton Reference Morton2007). Indeed, building is in many ways symbolic of living; as an interlocutor of Julie Livingston’s pointed out, ‘without building there is no life’ (Livingston Reference Livingston2005: 15; see also Klaits Reference Klaits2010: 85). But like most such achievements, it is fraught and generates dikgang; and these dikgang derive from a new uncertainty in the very relationships the builder has put to work in constructing her lelwapa in the first place – or that she might call upon in the ongoing process of building in the future. These uncertainties, in turn, are exacerbated by the new distance the builder is establishing between herself and the people she has relied on – largely family – by building apart. I would argue that it is this production, acquisition, and management of dikgang, beyond the work of mobilising relationships and materials for construction, that gives building its salience for Tswana self-making and personhood – and also for Tswana kinship.

But Lorato’s story also underscores another critical dimension of building, of the spatialities of kinship and personhood, and of the dikgang these produce: their temporality. In all three of the exchanges described above, the problem was not simply that Lorato was building a house, or where or how it was being built; the problem was with when it was being built. Lorato was building not only before marriage but before having children – a time when her major responsibilities were still to her natal yard (especially since she had landed a steady job). She was building before most of her mother’s siblings, including Modiri, her mother’s brother or malome. She was building for herself before she had built for her parents – something many of her mother’s siblings had done (as well as the two-and-a-half that Moagi had built, Kelebogile and Kagiso had tiled the house, installed plumbing, and made various other major infrastructural additions; see also Livingston Reference Livingston2005: 15). And, as neighbours frequently commented, she was building fast; most of the house was completed in under a year (although, importantly, it was never entirely finished). Lorato was building out of sync, out of turn, and out of time; and these distemporalities were all potential sources of dikgang, especially with her mother’s siblings.

Like the dikgang explored in Chapter 1, the potential dikgang posed by Lorato’s building distemporalities were framed and anticipated in terms of intergenerational disruption. Inheriting her mother’s plot and beginning to build was part and parcel of a gradual process in which Lorato was both becoming an adult and shifting to occupy her mother’s familial role (as described in the Introduction), both of which were fraught intergenerational transitions. But, whereas her mother had been the eldest Legae daughter, Lorato’s relative youth and inexperience meant that she was drawn into her mother’s generation as a younger sibling. As we will explore further in Part II, Tswana sibling relationships are often cast in parent–child terms – as, too, are marriages. Part of Lorato’s transition was eased by the fact that, in these terms, she remained the child of her mother’s siblings. But by building in advance of her malome and her mother’s other siblings, and in advance of a future potential husband as well, Lorato was upending those relationships – a child become parent. Building before having a child of her own (and therefore still a child herself), but also before building for her parents, exacerbated her uncanny position, to the extent that she herself was uncomfortable with it.

These temporal and intergenerational dilemmas had a number of sources. First, there was the matter of early inheritance: Lorato was only 14 years old when her mother died. Inheriting property so young is unusual among Batswana and is a possibility that only really began to arise with the advent of the AIDS epidemic. In fact, Lorato might not have inherited the plot at all; Mmapula might have retained it, sold it, or given it to another of her children, and she would have been well within her rights to do so. Given that both Mmapula and her other children were, at the time, favourably situated with plots – and in a context where complaints of property grabbing from orphans had become a hot topic of discussion everywhere from the kgotla to social workers’ offices and in the popular media – Mmapula made the decision to transfer the plot to Lorato. Both the orphan care NGO in which Lorato was registered and the local social worker’s office assisted in the process. But formalising the inheritance wasn’t sufficient to normalise its distemporality; as Kagiso’s pressure demonstrated, for as long as the plot was undeveloped, it remained potentially subject to claims from older kin – in the Tswana sense, Lorato’s parents – who were ready to build, as well as from the land board itself.

In consultation with other arms of government, the land board had suspended its usual development requirements in cases like Lorato’s. No specific new deadlines for development were given, although it was rumoured that inheritors such as Lorato might have only five years to develop from the age of majority (18). Given the scarcity of jobs and the expense of building, even this apparent leeway was insufficient – especially as applications for plots in Dithaba began to outstrip the availability of gazetted land, and the land board began reclaiming and reassigning plots that had not been suitably developed. Government-linked charitable organisations such as the Masiela (Orphans) Trust Fund got into the building game in anticipation of these scenarios, mostly where orphaned children in destitute families had inherited land (Masiela Trust Fund 2015); NGOs also built houses ad hoc for child clients in difficult circumstances (as we will see in Chapter 3). People like Lorato and her family had few options beyond their connections to someone like me, whom they had met through their involvement with NGOs.Footnote ²

Charitable organisations, NGOs, and associated individuals were all able to mobilise much larger immediate capital than many builders could, a situation that – in concert with land board pressures – could speed up a building process that was otherwise undertaken over years, as and when materials and labour were available. Whether because they needed to prove the timely disbursal of funds to donors (as many NGOs did), or whether they had only a limited time to be involved in the work (as I did), these additional figures were all working on different clocks – and therefore knocking builders like Lorato out of their proper time. In this sense, the untimely death of Lorato’s mother inserted Lorato and her family into what could be glossed as a transnational humanitarian project on the one hand, and a national development project on the other, in some unpredictable ways – thereby introducing unprecedented influences on the spatio-temporalities of her family, their intergenerational relationships, and her own self-making trajectory.

Several months later, Lorato’s house was nearly finished – a state that turned out to be perpetual, as most building in Botswana is – and we sat on the wide stoep, taking in the view. Her neighbour immediately down the hill had recently finished a small two-and-a-half of his own, and its clean corrugated tin roof glared in the sun. I asked whether she had ever spoken to him.

‘He’s late,’ she said, using the sensitive Setswana idiom for death.

I was taken aback. The house had been finished less than a month. The neighbour had only recently moved in, having never really stayed at the plot before, although it had a pre-existing structure. She explained that he had died in his sleep. It was several days before his body was found.

I asked what had happened – whether it might have been witchcraft born of jealousy, on account of the new house. But Lorato shrugged and shook her head, unconvinced. ‘Gareitse,’ she said – we don’t know. ‘But that’s why I don’t like the idea of staying alone.’ As much as she had dreamed of building for herself as an escape from the pressures of staying at home, to stay alone – and therefore to be seen to have been building for herself (Klaits Reference Klaits2010: 86) – was not only unconscionable, but also potentially dangerous. In her case, the risk was greater because of the intergenerational tension it threatened. ‘They are going to want to teach me a lesson, you know, at home,’ she added, almost as an afterthought. While Lorato’s situation was in many ways unique, she could nevertheless predict the dikgang that would emerge and the way in which they would be interpreted: as a lesson to her about her proper place, generational role, and claims to self-making.

Lorato did stay in her house for a short while, almost experimentally – not alone, but with Lesego and Tshepo, two of her teenage cousins,Footnote ³ who came to help and who were similarly eager for some space away from the family. The adults at home accepted this arrangement in principle and for the time being, but they were insistent that all three girls should make themselves available to help at the lands and at home as usual. They lasted less than two months. Partly, it was too difficult to keep everyone fed on Lorato’s meagre income; partly, all three missed being in the bustle of home. But, above all, juggling obligations at their natal yard with piece jobs and the work the new place required was too onerous for the distances and time involved. The situation had created an ongoing battle with the family, who continuously berated all three girls for neglecting their duties at home – teaching Lorato a lesson, as she anticipated they might. The distemporalities of Lorato’s building project, the profusion of overlapping, ongoing – and gendered – obligations in disparate places they entailed, the instability of the relationships that might have supported her, and her own indeterminate and tenuous generational position made staying away ultimately too difficult to manage. While the new house had seemed to present an opportunity to escape the burdens and dikgang of living at home, in fact it simply added to them and made them more difficult – eventually impossible – to navigate. Lorato was as yet unable to sustain, through space and over time, the relationships, responsibilities, and dikgang that living apart entailed.

Deborah Durham notes that Batswana link ‘the inability to manage people and relationships’ (Reference Durham2004: 594) with childhood, while David Suggs notes that women’s adulthood depends on others’ ‘believing they have competence … [in] the establishment of managerial household independence’ (Suggs Reference Suggs2001: 108). I would connect the management of people, relationships, and household independence to the management of dikgang arising from the relationships on which households rely. If self-making is, in part, the continuous acquisition, navigation, and successful negotiation of dikgang – a process that the perpetuity of building might be said to symbolise – then Lorato’s failed attempt to set up house for herself marked a setback in making for herself and in making the generational transitions that implied. Lorato returned home, and the house remained empty.

Figure 4 Lorato’s house.

Lorato’s frustrated experience suggests that the spatio-temporalities of humanitarian and development projects, whether undertaken by government or by NGOs, have unexpectedly important roles to play in the spatio-temporalities of Tswana personhood and kinship. In the final chapter of this Part, I trace the spatialities of organisations in Dithaba that work directly with families like Lorato’s, and examine the ways in which they both echo and subvert the spatialities of the families they serve.