Demanding from Others: How Ancestors and Shamans Govern Opacity in the Kalahari

ABSTRACT Among Jú|’hoànsi in the northern Kalahari, there are differences in the way people address their suspicions about what others might be thinking or feeling – in other words, in the way people confront the opacity of other minds. Among friends, playful forms of mockery allow people to express their suspicions or ill-feelings directly, without the fear of causing harm. Among relatives, by contrast, suspicions and ill-feelings are typically concealed. While people may talk about other minds when those minds are not around, they refrain from direct confrontation. To confront one’s relatives is to make them feel ‘pain in their hearts’, and to do so is to risk losing them to sickness or damaging enduring arrangements of care. Ancestors and shamans, who can see and hear more than others, play a crucial role here in governing opacity: exposing suspicions and ill-feelings when people feel they cannot speak of them.

appear somewhat outside of it. They are seen to help people understand certain contradictions at the heart of egalitarianism or decentralised politics, but tell us little about the conditions of social life and how these conditions shape the values we live by.
From an analytical perspective, the ancestors are certainly ambiguous. But to a Jú|'hoàn person, the dust-devils who tell of their presence have only one obvious meaningthey bring to light the breakdown of relations among relatives. There were many occasions throughout my ethnographic research within the region, when dustdevils would pass by and people would speculate about the suspicions, misunderstandings, or ill-feelings that motivated them. 'That is an ancestor' (g||aoan hin to'a), my Jú|'hoàn interlocutors would remark as we watched one go by. 'They want to kill their relatives, to make them sick' (Hi kare !hun ju, n ǂai t'am |kau ju). The dust-devils are ancestors responding to moments when their living relatives have pain in their hearts (!ká t'am | kau), either because others have wronged them or because they, suspicious or feeling 'pain in their hearts', wronged others. The ancestors long for the company of their living relatives, and they seize any opportunity to remove them from acrimonious situations they may find themselves in. 'It's dangerous!' Tshae, a renowned shaman whose role it was to confront the ancestors, explained to me ' … if they keep feeling pain in their heart, they could fall sick. The ancestors could come for them and they could die. ' This paper looks closely at the relationship between these forms of sickness and how they are addressed and, in the process, addresses a range of assumptions made within the regional literature about social life in the Kalahari and the values that sustain it. Most notably, the assumption that these are societies that practice 'universal kinship' and that demand sharing is the primary means through which people reckon one another as 'kin'. These assumptions are embedded within wider tendencies within the literature to frame social life primarily in terms of the values associated with decentralised societies or egalitarianism. While this has allowed for a much better understanding of egalitarianism, and decentralised politics, it has left other forms of sociality overlooked. These forms of sociality matter because they tell us something about the wider economic context or historical processes that support the development of these varying forms of governance.
Consider, for a moment, the context for what Jú|'hoànsi take to be the most active time of year for the ancestors: the dry season, when months without rain have left the landscape awash with pale, dry grass and brown, leafless shrubs. This is 'one of the most difficult environments on earth' (Barnard 2019: 43) and winter is a period of notable uncertainty and suffering (||háí tih). This suffering is partly due to a scarcity of resources, and partly due to the strain this scarcity puts on social relations. The dry season is not the only time of scarcity in the Kalahari, with unemployment levels now high and many dependent on destitute rations throughout the year, scarcity affects people throughout the year. It is a time, however, when things get noticeably worse, and with that, a time when dust-devils are more numerous. They are an index of the health of relationships among relatives, at any given time, and where one finds them gives some indication of which relatives have 'pain in their hearts'.
Ancestors dwell on the edges of territories (n!oresi), somewhere beyond the horizon to the east from where the hot winds come. As one looks out over the sparse landscape, a dense grove of trees appears on the horizon. These vast stretches of rolling, fossilised sand dunes appear mostly emptyinterrupted only occasionally by small villages of no more than one hundred people or by salt pans and waterholes that sustain the region's many birds and animalsbut Jú|'hoànsi know the terrain to be made up of territories. These territories are the hunting, gathering, or living spaces of particular families and the dwelling places of their ancestors (from whom one inherits n!ore rights). The territory is thus an index of social relationships that have developed there, as much as dust-devils are an index of the suspicions, misunderstandings, or ill-feelings that threaten to break social relationships down (but which people struggle to speak of).
Ordinarily, ancestors seem to pay little attention to their living relatives. When their relatives feel pain, however, or they become the subjects of suspicion, misunderstanding, or ill-feeling, they become active. 'It starts with cursing or when someone refuses another person', a renowned shaman had lamented, ' … and from then people feel pain in their hearts. That is when the ancestors come to take their relatives away with sickness.' While suspicions, misunderstandings, and ill-feelings are a regular part of social life in the Kalahari, as much as elsewhere, and affect relatives as much as friends, what is distinct is the way that relatives avoid confronting one another and the way that the ancestors intervene when such avoidance causes them pain. It is customary for relatives to avoid confronting one another, and to do so by refraining from expressing suspicions and ill-feelings directly. Relatives, in other words, cultivate a certain opacity of mind. Being a relative means being opaque about one's suspicions, misunderstandings, and ill-feelings, where being a friend does notor at least in a markedly different way.
The situation is thus decidedly more complex than the regional literature has suggested up to now. There are not only different ways of reckoning kinship, but demand sharing is not the only means through which kinship is established or maintained. Central to this discussion are differences in the way that people confront (or refrain from confronting) one another about suspicions, misunderstandings, and illfeelings. These differences have implications for the central proposition of this special issue: that tools of legibility and approaches to intention management correlate with forms of governance. We find not only that there are differences in the way people establish or maintain their status as relatives and in what it means to be a 'relative', but that these differences can be at least partially accounted for by considering experiences of uncertainty and the way these shape relationships of care in different ways in the Kalahari.

Navigating Uncertainty and Cultivating Opacity
The role that healing dances play in restoring health and well-being in the Kalahari is well-documented in the regional literature. So is the relationship between periods of sickness and the breakdown of social relationships (Lee 1967(Lee , 1979Marshall 1969Marshall , 1976Marshall , 1999Katz 1982;Biesele 1993;Katz et al. 1997;Guenther 1999). What is perhaps less well-documented, and certainly under-theorised, is the role that ancestors and shamans play in dealing with the problem of the opacity of other minds and uncertainty in everyday life. There are also few analyses showing how experiences of economic uncertainty shape social relationships in the Kalaharigiving rise to specific obligations to share but also to specific ways of managing the suspicions, misunderstandings, or ill-feelings that arise when people choose not to uphold them. Uncertainties, in other words, not only generate opacity by making it harder to watch over people and to know other minds, they shape the way people manage it. Few analyses have taken seriously how these issues play out.
In the introduction to this special issue, Buitron and Steinmüller ask what the correspondences are between specific forms of governance and whether people choose to 'cultivate opacity'. Within contexts where individual accountability is not central to the exercise of power (such as where political regimes are structured around valuing autonomy or where people take collective responsibility for preventing individual acts of wrongdoing), people are said to cultivate opacity. By 'cultivating opacity', they mean not speaking openly about the minds of others. This builds upon the writing of Robbins and Rumsey (2008: 408), who reflect upon how in the Pacific (and likely elsewhere), we find people 'ruminating on how difficult it is to see into the hearts and minds of others'. This gives way to a close analysis of how people approach intentionality, theory of mind, and empathy, but also to a close analysis of the way certain approaches to opacity align with particular political regimes. Picking up on Stasch's (2008: 450) observation that among Korowai it is 'a matter of claiming authority to know or determine' the minds of others, this special issue draws attention to different approaches to the problem of opacity and forms of governance that shape them.
Taking a broadly comparative approach, Buitron and Steinmüller frame these differences as those between 'state' and 'non-state', or 'centralised' and 'decentralised' societies. In contrast to decentralised forms of governance, centralised forms of governance require forms of legibility that make it possible to control or manage individuals or to make them personally accountablewhether for good deeds or for transgressions. At their most idealised, centralised forms of governance tend to simplify social life to a series of values and duties, where decentralised forms of governance tend to focus on the way that values and duties are interpreted or pursued differently, depending on the context. Promoting transparency, with the aim of predicting intentions or making individuals accountable for their actions, is said to be key to statecraft. Cultivating opacity, with the aim of reducing individual accountabilityeither because people value autonomy or because they choose to emphasise collective or group accountability, is said to be key to more egalitarian forms of sociality. This paper, based on fifteen months of ethnographic research among the Jú|'hoànsi of north-eastern Namibia, 1 complements this discussion but offers a distinct framing. Rather than compare a famously decentralised society, the Jú|'hoànsi (better known in anthropology as the !Kung), against centralised or state societies more generally, it examines the role that cultivating opacity plays in some areas of social life and not others. By 'cultivating opacity', this paper refers not to the avoidance of mindreading more generally, but specifically to the way people refrain from confronting one another directly about suspicions, misunderstandings, and ill-feelings. 2 The areas of social life or types of relationships in question are those between relatives, on the one hand, and friends, on the other. Among friends, playful forms of mockery allow people to express their suspicions or ill-feelings more directly, without the fear of causing harm. Among relatives, by contrast, suspicions and ill-feelings are typically concealed. While people may talk about other minds when those minds are not around, they refrain from direct confrontation. Since broader values of fairness and autonomy characterise friendship as much as kinship in the Kalahari, these differences provide a compelling starting point for interrogating, as this special issue does more broadly, whether tools of legibility and approaches to intention management correlate with forms of governance. They also challenge what are key assumptions about kinship and sharing in the Kalahari: that people practise 'universal kinship' and that they establish and maintain the status of 'kin' by practising demand sharing (Barnard 1978(Barnard , 2016Widlok 2017).
The expectation that people should share has long been a feature of social life in the Kalahari. Being a Jú|'hoàn person means sharing what you have with others. Jú|'hoànsi are well-known for sharing in this way (Marshall 1961;Lee & DeVore 1969;Sahlins 1972;Lee 1979;Barnard 1992, and more recently Widlok 2017). They, like other 'hunter-gatherers' or 'foragers' (as they are commonly recognised) are described as an 'egalitarian society' (Woodburn 1982) one that practices 'demand sharing' (as described by Peterson 1993) and values autonomy. This means that rather than engage in unsolicited giving, people in egalitarian societies 'allow others to take what is valued' (Widlok 2013) and they do this whenever the balance of power, prestige, or wealth shifts and some appear to have more than others. They are also a community who support people in choosing when to do so. This means that rather than choose when to give, Jú|'hoànsi choose instead when to present themselves to others to be demanded from. In theory, this supports those who wish to accumulate wealth what Jú|'hoàn speakers describe as efforts to 'ruin others' (kxuia ju). In practice, people tend to present themselves to othersthey make themselves legible (when they have nothing to give) and they allow others to make demands of them (when they do). People do this because they need one another, but also because being a good person (ju jan) means allowing others to take what is valued. 'You are a good person,' (a o ju jan) my Jú|'hoàn interlocutors would say, 'if you help others, if you are free and open' (ka a hui ju, ka a o koeqe te sara). These forms of governance are key not only to achieving the levels of equality that have been observed here among Jú|'hoànsi but for stabilising what are undulating forms of abundance and lack. They are key, in short, to navigating uncertainty.
This captures what it means to be a 'true' or 'proper' person (whether you are a Jú|'hoàn speaker or someone who behaves like one), but it also flattens what are more complex arrangements of care and ways of reckoning kinship. Among friends, unlike relatives, suspicions and ill-feelings are expressed more openly. They engage in playful forms of mockery with those who are suspected of avoiding their commitments to others, or of trying to accumulate power or prestige (see Laws 2019a). While there is a degree of opacity in these encounters, leaving open the possibility that suspicions are unfounded or ill-feelings unjustified, there is no doubt among those who engage in these forms of mockery that real suspicions or ill-feelings are what motivate them. There are also differences in the way that people are expected to care for one another. While 'all Jú|'hoànsi are family', and being Jú|'hoàn means helping each other, not all Jú|'hoànsi kxaea khoe ('have each other'; in other words, are relatives). As Lorna Marshall (1976: 214) and Elizabeth Marshall-Thomas (2006: 223) write about their experiences carrying out ethnographic research between 1950 and 1958: despite 'our Ju/wa names and the implied relationships' and sharing 'tirelessly', 'we were not an integral part of the social fabric and no one felt the urge to include us as, indeed, we did not belong'. One cannot become a relative simply by engaging in demand sharing and this applies to outsiders as much as it does to Jú|'hoànsi.
This is not to say that 'relatedness' is not a meaningful category hereafter all, being Jú|'hoàn means being a 'person' (jú) who behaves in ways that are 'true' or 'proper' (|'hoàn). It is to say that sharing the same ancestors, and in turn sharing rights to the same territory, means having specific commitments and expectations. 'If you are someone's relative,' my Jú|'hoàn interlocutors explained, 'you give thought to one another' (a ǂ'ang |'an a juasi). To be in such a relationship is to 'have a happy heart' (!ká n|'ang)it means knowing that one will always be cared for and never be shamed or alienated. Relatives are not only expected to share what they have (and not only if people make demands of them), but to avoid confronting those they either suspect of wrongdoing or of harbouring ill-feelings towards them. They share what they have because it is the right thing to do, but they cannot be made to do so, nor should they be confronted about what they have to share or whether they have refrained from sharing. They are relatives, people who 'have each other' (kxaea khoe), and relatives avoid confronting one another with suspicions, misunderstandings, and ill-feelings.
When framed in this way, being struck with sickness by one's ancestors serves as an effective means of 'governing opacity'. In other words, it brings suspicions, misunderstandings, and ill-feelings into the open when people feel they cannot speak of them, but when speaking of them is central to ensuring people maintain their commitments to one another. Being struck with sickness by ancestors, and having that sickness (and its cause) carefully exposed by shamans is, like playfully mocking one another, an effective means of reaffirming the commitments people have to one another and that sustain social life in the Kalahari. Where one involves cultivating opacity, however, the other involves addressing opacity more directly. Within the broader framing of this special issueon the correspondences between tools of legibility and forms of governancethis paper asks whether, and to what extent, efforts among Jú|'hoànsi to cultivate opacity correlate with certain forms of governance that distinguish friends and relatives. 'Your relatives,' I was told, 'are people you who love you very much and they want you to be happy' (A juasi n!obe are a te kare ka !ká n|'ang). As relatives, when it comes to sharing, they need not make demands: they either take freely from one another or they expect that people will share with them without ever needing to ask. The same cannot be said for friendships. Though friends regularly care for one another, a friend is also 'a person you really cannot trust.' 'These people', one of my interlocutors explained, ' … share with you one day, or hang around with you when you are drinking, as if they are good and they ask you for things, then next thing they are gone, and they are doing bad things '. 3 What this suggests is that, rather than simply value autonomy and sharing and apply it equally to anyone they meet, Jú|'hoànsi have commitments to relatives that are quite different to the commitments they have to friends. These differences, in turn, shape the way that people address the doubts, suspicions, or ill-feelings they feel towards one another. Relationships between friends are characterised by demand sharing and generalised reciprocity, and people are committed to one another more loosely and in the moment of encounter. Relationships between relatives are characterised by more enduring and fixed commitments to the needs or desires of others by visiting, listening, giving, or taking. Understanding why it matters that certain relationships of care endure, and therefore why it is important to cultivate opacity, calls for a close analysis of a wide range of ethnographic materialboth past and contemporary. On the one hand, the emphasis placed on not confronting one's relatives appears to be due to the closeness of these relationships. Across the northern Kalahari region, whether at their territories or elsewhere, relatives very rarely live apart or move away from one another. They are also people you should generally 'fear' and 'respect', in line with long-standing patterns of 'joking' and 'avoidance' that shape relationships within families and between those who share names (see Marshall 1957). On the other hand, avoiding confrontation appears oriented towards maintaining closeness, either in the face of economic uncertainties and obligations to share that draw people apart or because of new political and legal uncertainties surrounding communal land and resources. Not only does gaining territory rights depend on 'having relatives,' but actually benefitting from these territory rights (and the opportunities they offer for work or subsistence) means being on good terms with one's relatives. With the formal registration of ancestral territories after the fall of the apartheid regime, it both became possible to gain access to one's ancestral territory and harder to establish new onesharder, in short, for families to move apart. This paper approaches the task of working out how tools of legibility relate to forms of governance by looking closely at the values that shape how and whether people confront one another, and at the conditions under which such values are established and maintained. With respect to relatives, cultivating opacity is a way of giving people the benefit of the doubt. In other words, it is a way of recognising that acts of wrongdoing often follow misunderstandings and ill-feelings that are collective or intersubjective problems. With respect to friends, uncertainties are not only inevitablea case of not knowing where resources are or where they may appear next. They are an important feature of forms of autonomy that give people the freedom to carefully manage their commitments to people they feel they cannot trust. It falls upon the individual to carefully manage these commitments, to make themselves legible as well as vulnerable to others, and to use mockery as a way of reinforcing that others should do the same. The next section turns to economic uncertainties, both as they play out today and might have played out in the past, and analyses how they shape different arrangements of care and tools of legibility. The section thereafter turns to how ancestors and shamans bring suspicions, misunderstandings, and ill-feelings into the opento how they 'govern' the opacity that arises when people avoid confrontation. The final section looks more closely at avoidanceanalysing cases of sickness and the specific commitments or expectations they index. This brings us back to uncertainty, to the way is shapes social life in the Kalahari, and to how it helps us understand tools of legibility.

Arrangements of Care in the Kalahari
The region of north-eastern Namibia where this research was conducted is known locally as N||oq'an!'ae, and by outsiders as Nyae Nyae. It is a communal conservancy in the Kavango-Zambezi transfrontier conservation area, south of Khaudum National Park and west of the border with Botswana and the Dobe region. It is estimated that 3000-4000 Jú|'hoàn speakers live here (see Biesele & Hitchcock 2011: 6), either at their ancestral territories in the conservancy or alongside other language groups at Tsumkwe (Tjum!kui), the region's administrative centre and municipal town. Under the apartheid regime of South Africa, following the Odendaal Commission of 1964, the region was declared a 'homeland for the Bushman nation' (ibid: 36). About a decade before, the apartheid government established Tsumkwe as an administrative centre and urged Jú|'hoànsi to move to the town and become sedentary. Many of the region's inhabitants did so, seeking to benefit from the distribution of rations, provision of medicine, and opportunities for employment. With the advent of the Namibian War of Independence, many Jú|'hoànsi were recruited as trackers by the South African Defence Force (SADF) and Tsumkwe became an important military stronghold for the apartheid regime. Many scholars (see Lee & Hurlich 1982;Gordon & Sholto-Douglas 2000;Biesele & Hitchcock 2011) have noted the damaging effects that colonisation and militarisation had on the region's inhabitants. In the years leading up to independence, supported by the ethnographic filmmaker John Marshall, a 'back to the land' movement began which sought to address these effects. Jú|'hoànsi, hoping to regain their ancestral rights, formally mapped their territories and devised strategies for resettlement.
The promise of a mixed-subsistence economy, in addition to the promise of political self-determination, motivated most families to return to their territories. As one of my Jú|'hoàn interlocutors remarked; 'John said they would drill boreholes for anyone that moved back to their territories and help us become farmers so that we had food'. There were numerous challenges along the way, however, making the reality decidedly more complex. Not only were there no boreholes for water at their territories but hunting and gathering was 'a difficult life', especially during periods of drought. People living at their territories struggled with predators and drought, and there were major disagreements over how best to generate the funds to maintain infrastructure. Establishing the region as a communal conservancy was considered one way to address both economic and political challenges. After establishing themselves as a distinct 'traditional community', who recognise their own 'traditional authority', conservancy status gave people the right to hunt and gather, rear small herds of livestock, and tend gardens at their ancestral territories. It also gave people the right to generate an income by selling licences to professional hunters, entertaining tourists and collaborating with researchers, and harvesting and selling veld products. Since the region became a conservancy in 1998, it has generated enough revenue each year to maintain boreholes, conserve fauna and flora, invest in and maintain infrastructure (including vehicles), support a wide range of community projects, and pay annual dividends to members.
Despite these successes, political and economic uncertainties remain. There are not only ongoing threats to political self-determination in the form of incursions by neighbouring groups looking to utilise rangelands (Hays 2009; Begbie-Clench 2016; Hays & Hitchcock 2020), but economic challenges in the form of widespread unemployment. The conservancy employs several members in management or as rangers. A few other members have employment in government ministries, non-governmental organisations, or as tour guides. The majority, however, depend on state destitute rations, child welfare grants, pensions (either their own or other family members'), piecework, or selling hand-beaded or -carved crafts to tourists. Some of these sources of support are regular, allowing people to take credit with petty traders (albeit at high interest rates), but still they are 'never enough' (ka |oa ǂ'aun). Speaking of his own circumstances, one of my Jú|'hoàn interlocutors described the circumstances of many when he said: You see the way I just hang around. The time drags on as I go around looking for small jobs, and then when I do a small job, and it brings me a small thing, I take it, but it is never enough.
Getting by from one day to the next depends on 'roaming in order to live' (zula ka ku | xoah) (also see Boulton 2019). They roam both in the hope of finding work or selling crafts, and to foster or maintain arrangements of care (see Dawson 2019 for a focus on unemployment or Schnegg 2021 for a focus on food sharing within the broader southern African context). Jú|'hoànsi describe their circumstances in terms of either 'good luck' and 'fortune' (kxani) or 'bad luck' and 'misfortune' (!xò). Whether heading out to hunt and gather, or in search of work, you hope either for good fortune, or that those who are fortunate share what they have with you.
This sharing takes a wide range of forms, depending on what is being shared, who is sharing, and where sharing takes place. After a few months of ethnographic research, these differences became apparent. After asking about them, one my Jú|'hoàn interlocutors explained that: If you are asking a stranger to share with you, we say you beg (g ǂara). We say this because it is like you know that person will refuse, so you behave as if you are ashamed. If you are asking a Jú|'hoàn person, we say you demand (||an). You just say give me (na), and they will give to you if they can. If you go to their place, however, sometimes you will feel fear (koaq) or shame (taokhomm), because you will feel like you are begging. At home, with your family, you do not need to ask, they will just give to you or you can just take if you need it if it is there.
The form that sharing takes, in other words, is shaped by whom you encounter and where you encounter them. While early ethnographic analyses of sharing among the Jú|'hoànsi (notably Marshall 1961Marshall , 1976) drew attention to these differences in approaches to sharing, later analyses have tended to overlook differences in favour of a focus on 'demand sharing'. This is due partly to the way early analyses described sharing in terms of 'reciprocity' or 'exchange', stretching these notions to 'cover transactions that are clearly not reciprocal at all' (Widlok 2017: 17; also see Woodburn 1998). In Sahlins' (1972) model, for example, we find 'generalised reciprocity' at home among relatives, 'negative reciprocity' on the periphery among strangers, and 'balanced reciprocity' between them, among friends or acquaintances. And yet, in the Kalahari, one speaks of 'reciprocity'of repayments or of exchangesonly when carrying a debt ( ǂo) or in a gift-giving (xaro) relationship (see Wiessner 1977Wiessner , 2002. Typically, sharing means responding positively to the demands of otherseither because they desire it, or because they need it. It is not a form of 'exchange' or 'reciprocity', because what or whether one returns anything depends upon one's ability to do so. The issue is partly one of generosity, of being 'close-hearted' (!ká tò'm), and partly one of fairnessif 'it is there', as my Jú|'hoàn interlocutor suggests, it should be shared with others.
This emphasis on making demands within the context of disparities between people has generated new ways of thinking about sharing, and served as an important rebuttal to the tendency to think of sharing in terms of generosity or charity. It has also contributed to a tendency to overlook differences between approaches to sharing. These differences are important because they draw attention to the contextual factors that shape social life in the Kalahari (including whether, and if so how, people confront one another about suspicions, misunderstandings, or ill-feelings). The economic uncertainties facing people in the Kalahari today compel them to 'roam'in search of work or of people who may care for them, but also in order to make themselves legible. As my host often stated, 'I have been at work for a long time with you. I have to spend today visiting people or they will start to wonder about me.' Since getting by from one day to the next depends not on acts of generosity or reciprocity but on being present to make and respond to demandsmaking oneself legible emerges as crucial to arrangements of care. If reciprocity features in these encounters, it is not in the act of sharing but in the act of making oneself present to be demanded from.
Speaking of relationships of demand sharing, Widlok (2013: 20) argues that rather than being determined by one's existing status as relatives, 'the categorisation of a particular person with a particular kinship term often … follows the interaction rather than the other way around'. Following the discourse of universal kin classification (Barnard 2016), the process happens spontaneously: 'any strangers who might have cause to engage in marital alliance, or possibly even the trade of material goods, would be fitted into kin relations' (Barnard 2016: S148). As much as relatives are said to be made in this way they can be unmade. This fluidity of kinship status is important because roaming fosters arrangements of care between people, but it also makes them vulnerable to those who 'cannot be trusted', who 'share with you one day … then next thing they are gone … doing bad things' (see Laws 2019b: 93). One's status as kin thus serves as a marker of one's commitment to certain, collective intentionsin this case, to care for one another and to do so indiscriminately.
The problem with scholarly contributions to this area of study, however, is that they give the impression that all relationships are organised along these terms, or that these are the only ways of reckoning kinship within the region. These forms of sharing allow people to navigate uncertainty, and encourage people to be good to one another, but they are not the only arrangements of care in the Kalahari. Among those who 'have each other', being present is less about making oneself legible than it is about 'giving thought' to others or allowing others to 'give thought' to you. The emphasis is not on fairness, but on ensuring the continued 'participation in another's existence' (Sahlins 2013: 18). One's home (tjù|'hó), whether in town or at one's territory, is a space of distinctive commensality. Whether food is bought with money generated through employment, made up of destitute rations distributed by the government, or gathered or hunted in the bush, it is something you share with your family. It is also where you care for your family in other ways; by tending to gardens or herds of livestock, assisting with craft production and cultural tourism, or simply providing help or care in other ways. This is not to say that the home is a space of entrapment, or that 'having each other' is opposed to the autonomy of friendship. These are relationships that are described in terms of looking out for each other (sea khoe), helping each other (huia khoe), understanding one another (tsa'akhoe), or trusting and respecting one another ( ǂomakhoe)not strictly in terms of 'obligations'. As my hosts explained, your relatives want you to have a 'happy heart'. They want this because they love you, and because it matters to keep the family together. These obligations are quite distinct from those that exist between friends with respect to demand sharing. As the final section of this paper explores in more depth, there are long-standing patterns of avoidance that maintain arrangements of care within families in the Kalahari. There are also new political and economic uncertainties, however, that give new weight to the 'territory' and to the relationships it indexes. 'You can become a member of a territory if you have family,' my host explained, 'but if you are fighting with your family, you cannot feel free there, like it is your place.' What becomes apparent when analysing the differences between kinship and friendship here is that there are tensions between one's commitments to relatives and one's commitments to broader relationships of care. Where the latter draw you away from the territory, and from your relatives, the former draw you away from town, and from your friends. Expressing his own ambivalence about travelling back and forth, one my Jú|'hoàn interlocutors described how he typically plans only to 'go get tobacco and sugar and come back immediately', but 'zula' makes him 'forget he just came from the village'. When he comes home, he then finds he has nothing to give to his family. At the same time, when faced with demands from friends, many stated that the only reasonable response was to say 'hold any bad feelings you may have towards me' (||Ae na), ' … my family are looking for me, but I will help you next time.' Navigating political and economic uncertainties requires that they turn to those beyond their families. 'When you are around people, they feel good, but when you spend time away they start to wonder if you are thinking of them or maybe doing bad things.' While further research is needed into how socio-economic status shapes relationships between families, as well as between generations, preliminary observations suggest that economic security correlates positively with a reduction in roaming as a feature of social and economic life. Suspicions, misunderstandings, and ill-feelings still feature among families with a relatively secure socio-economic status and people still avoid confronting them. Cases of sickness suggest, however, that these are less frequent and certainly less likely to emerge from problems related to sharing. The next section turns to these cases of sickness, to the situations of opacity they emerge from, and to the role that ancestors and shamans play in confronting them.

How Ancestors and Shamans Govern Opacity
Several months into my ethnographic research, my hosts took me to their territory, far on the north-western edge of the conservancy. They had not been home for months, spending most of their time in town where they could find work, and so had most of their extended family. Among the extended family members who frequently spent time at my host's yard in town, were a young couple with a new baby. They followed us back to the territory, not only to see their family but because their new-born son was sick. Two of their family members at the territory were shamans. They would help him recover and help his parents understand what was causing the sickness. Sometime around midnight, several family members came together to perform a healing dance. The next morning, one of the shamans explained that it was the ill-feeling of their parents that had motivated the ancestors to make their son sick. The couple had formed a family against the desires of their parents, who had hoped to see them marry into other families, and they had become a source of gossip. 'It is like they [their relatives] don't want this child,' so 'the young boy's namesake [the ancestor after whom the child was named] … he wants to take him away from here'. 'These shamans really have n|om,' my host continued, ' … their son will get better now.' There was, in short, a problem of opacity (of not knowing what others were be thinking or feeling about their union), and the ancestors brought these suspicions and ill-feelings into the open.
The healing dance, and the n|om that shamans possess, is key to understanding the way that ancestors and shamans govern opacity. In what is perhaps the most wellknown account of shamanism in the Kalahari, Katz (1982) describes num (n|om) as an 'energy' or 'medicine' that shamans draw upon to heal the sick. Shamans are described as relating 'synergistically' to n|om. As the dance progresses, and the polyrhythm of clapping, singing, and dancing builds in complexity, the shaman starts to sweat. During this process, n|om starts to 'boil' in their stomachstravelling up the spine before reaching the nape. The shaman then begins to shriek and convulse, suggesting they are about to 'die' (!ai)to travel on 'threads' (tsoo) to the ancestors (also see Low 2004: 150-155). At this moment, Katz (1982: 200) writes, n|om starts to 'flow freely among participants at the dance' and heal those who have suffered at the hands of their ancestors. While this description captures the effects that working with n|om has, and the role that shamans play in restoring vitality, it fails to capture what it means to 'really have n|om'. Drawing upon wider ethnographic research from the region (Marshall-Thomas 1959;Marshall 1961;Lee & DeVore 1969;Sahlins 1972; and Lee 1979), Katz develops a 'holistic' argument that aims to demonstrate how the trance dance fits into a larger social system predicated on sharing. The 'tradition of sharing' that prevents any one person from accumulating wealth, power, or prestige (Woodburn 1982), appears here to apply equally to n|om. When healers travel to the ancestors, they are demanding that vitalityin other words, the 'animacy of the lifeworld' (Ingold 2006: 10)be restored (also see Katz et al. 1997). Only when shamans 'become professionals' within the context of capitalism might they 'try to overcontrol or overmanipulate' n|om (Katz 1982: 203). 4 At this point, 'one is no longer dealing with [n|om] but with some 'containable' image of [n| om]', one that is no longer 'beyond the limits of any one person ' (ibid: 197).
Exploring what it means to 'have n|om' may seem tangential but it matters because it allows us to appreciate the way that avoidance (and the cultivation of opacity) circumscribes relationships among relatives in the Kalahari, and the role that the ancestors play in confronting it. People refrain from confronting one another or talking openly about their suspicions or ill-feelings, and n|om makes it possible to know them. Among Jú|'hoànsi today, n|om is something you either have (kxae) or are without (koara). A person at the dance, whether they are sick or participating as a performer, does not say that they receive n|om. They control the 'exercise of power' by the shaman (Biesele 1993: 73-76), but they do not 'have n|om'. Only shamans receive n| om, and to do so, as Tshae explained, is dangerousit is to experience pain. As Low (2004) points out, this commitment by researchers to specific visions of egalitarianism as against 'property' has downplayed the extent to which n|om is in fact 'containable'. Some weeks after visiting my hosts' territory, we went to speak to |Kunta Bo, renowned not only as a shaman but as a generous teacher and an excellent orator. 'He really knows everything about n|om,' my host explained, 'he can explain to you very clear what it is.' Sitting together at his home, he described n|om as 'something you cannot see that allows you to see, hear, and travel to the ancestors', to 'visualise' what is happening. Shamans say they receive n|om from the ancestors because the ancestors have the capacity (in the sense of a 'property' of action, see Graeber 2007: 22) to see, hear, or observe everything. Even though working with n|om may have the effect of redistributing vitality, especially the vitality that ancestors take from their living relatives, n|om is not 'vitality' itself (better translated as g|aoh 'power, strength' or |xoah 'life'). Vitality is what follows after healers ǂhoe 'pull out' the sickness that n|om allows them to 'see' or 'find' in the body (see Laws 2019b: 206). This is not to say that n|om is specifically the 'capacity to see', only that the n|om shamans receive from the ancestors is. Ancestors, and shamans (although only during the dance), possess a capacity to see that ordinary people do not. Ordinary people are faced uncertaintyboth as an inevitable part of everyday life and as a consequence of the way people avoid confronting one another. As a general concept, n|om refers to distinct capacities. It is 'potency', as Low (2007: S87) describes it: the 'ability of one thing to affect another in a particular manner'. As my Jú|'hoàn interlocutors explained to me, 'all things have a n|om'. A giraffe can walk long distances, a snake can kill with a single bite, and lightning can generate enough heat to burn a tree in seconds. Most things have some significant quality or capacity that they share only with others like themselves. We see this in the way the term is used elsewhere in everyday speech. Among Jú|'hoànsi today, n|om takes four primary forms. There is n| om that shamans draw upon to heal the sick, n|om that pastors draw upon to supplant the presence of evil spirits, n|om that doctors prescribe at the clinic, and n|om that sorcerers use to kill. Across these contexts, there are people who are described as n|omkxaosi, 'owners' or 'masters' of n|ompeople who are able to 'own' or 'master' the potency of certain beings or things (the 'properties' or 'attributes' that make them who they are and set them apart from others). 5 This ability to own or master the properties or attributes of others is, in fact, itself a property or attributeone that shamans, pastors, doctors, and sorcerers share with one another. The clue is in the name, n|omkxao. The suffix -kxao is used only in connection with verbs to indicate that someone or something is an 'owner' or 'master' of the action that precedes it. 6 Where 'properties' (generally speaking) are classed as nouns, n|om, in this case, is a verb. It is the act of drawing upon (or, perhaps, demanding a share) of the distinct potencies of others through ritual means. Where sorcerers draw upon the properties of snakes, cars, or lightning to kill people; doctors upon the properties of medicine to locate and eradicate sickness in the body; and pastors (typically charismatic) upon the properties of God to supplant the presence of evil spirits, shamans draw upon the properties of the ancestors to be all-seeing, all-hearing, and all-powerful so they may confront the ancestors and the problems of opacity that motivate them. 7 Ancestors, shamans, doctors, pastors, and sorcerers are extraordinary beings. They have a 'special skill or anything out of the ordinary' (Lee cited in Guenther 1999: 191) that means they are not like ordinary people. 8 They are people (ju), but they are other people (ju doresin). 9 Only 'other people' can be omniscient, omnipresent, or omnipotent.
A close analysis of n|om allows us to appreciate the specific capacities that ancestors have and it attunes us to the specific uncertainties that they attend to. The ancestors strike people with sickness not simply because they face uncertainty, but because they have 'pain in the hearts' on account of not knowing other people's minds. Among relatives, there is ordinarily something like an 'opacity doctrine' (Robbins & Rumsey 2008) at work that makes it inappropriate to voice one's doubts and suspicions about others. Since one cannot really, or at least not fully, know the thoughts and feelings of othersthe context for their behaviourit is 'a matter of claiming authority to know or determine' them (Stasch 2008: 450). Ordinary people cannot know one another's intentions in the way that ancestors or shamans can, and so they should not speculate on them. These are forms of generalised avoidance that, as the following section analyses more closely, have a long history in the region. While these certainly help people navigate new uncertainties, they are not themselves contemporary phenomena. In what follows, the broader work that avoidance does in addressing the social basis for acts of wrongdoing, or the contextual factors that shape suspicions and ill-feelings, becomes apparent. So does the role that ancestors and shamans play in this process when they bring suspicions, misunderstandings, and ill-feelings into the open. They remind people of the importance of 'giving thought' to others. This means not only sharing, but exercising a kind of compassion and recognising that it is often suspicions, misunderstandings, and ill-feelings that motivate acts of wrongdoing and should therefore be handled sensitively.

Generalised Avoidance and the Pain That Causes Sickness
There is good reason to assume that in egalitarian societies sharing is based on the right to demand an equal share of wealth, power, or prestige, and kinship is universal. As many anthropologists have emphasised, 'kinship' is not only a description of what we are made up of and how we came to be. It is a recognition of the specific ways in which we are 'intrinsic to one another's existence' (Strathern 2014: 45). Of how, in other words, we should sustain one another. An egalitarian society, in theory, should be one in which all people, irrespective of their place within a hierarchy or a family, can demand an equal share. It should also be one that ensures that people can move and speak freely (to 'vote with their feet', as Lee [1979] puts it). As noted, this captures what it means to be a 'true' or 'proper' person, but it flattens what are more complex arrangements of care and ways of reckoning kinship. Across the Nyae Nyae region today, people regard relationships between those who 'have each other', who share the same ancestors and by extension the same territory, as distinct from broader relationships predicated on sharing.
To borrow Gibson and Sillander's (2011: 1, my emphasis) phrasing, they are an 'open-aggregated community' in which 'all groups beyond the domestic family are loosely defined, ephemeral, and weakly corporate, and in which membership is fluid, elective, and overlapping'. The 'domestic family', those who 'have each other', care for one another in distinct ways. When they fail to do so, they are not discounted as they are among the Nayaka (Bird-David 2017: 217), they simply become the subjects of suspicions, misunderstandings, and ill-feelings. Similar to the Korowai of West Papua, Jú|'hoànsi 'reside far apart on separate clan-owned lands, and they do so explicitly in the name of freedom from each other's surveillance, desires, and claims to property'. But they do so while still travelling 'fairly constantly across ownership boundaries to visit one another, participate in collective projects, and share each other's company' (Stasch 2008: 448). By focusing on those areas of kinship that are fundamentally flexible and negotiable, we have swung 'the pendulum too far in the opposite direction until the other end of the kinship spectrum, that concerned with formalisation, normativity and fixity in turn disappears below our gaze' (Miller 2007: 537).
As we have seen above, these differences are evident in approaches to sharing but also in the way people avoiding confronting one another. The phrase 'between you and your mother' (itsa a taqe ||'ami), commonly used by Jú|'hoàn speakers, supports the observation that there are differences in the way people relate to one another. The way it used in everyday speech acts also neatly captures the effects these differences have on social life. When using the phrase, the term 'mother' can be replaced with any kin term. During my ethnographic research, the phrase was used on numerous occasions and always between friends in the event of wrongdoing. On one such occasion, one my Jú|'hoàn interlocutors called out to a friend they saw passing by the yard on their way to a nearby bar. The friend passing by had just been paid and was, presumably, heading out for a drink. He often did so, my interlocutor told me making it hard for his relatives who depended upon his income each month.
'Yeah, between you and your mother!' my interlocutor shouted humorously. 'You know how it is with us,' he explained, 'his family will say nothing.' On another occasion, the phrase was used to point out that a friend was 'doing bad things' (du | xuri) as if no-one would confront them. Either way, the term is used to draw attention to acts of wrongdoing and the way that relatives refrain from confronting one another about them. This is not to say that people are 'enjoined never to speak to or even gaze upon the other' (Graeber 2007: 16), as is customary for relationships of avoidance elsewhere. Rather, it is to say that they are committed to 'give thought' to one another in order to sustain their relationships with one another. As the cases presented in this paper suggest, this avoidance only goes so far in sustaining relationships in the Kalahari. The economic uncertainties that people face, make visiting and sharing with one another crucial not only to being a good person but getting by from one day to the next. Problems of legibility emerge not simply because sharing takes the form of demanding a share of what is visible, but because navigating economic uncertainty means that people are increasingly out of each other's sight. Though not the focus of this paper, there are also political and legal uncertainties surrounding one's rights to territory. In the aftermath of apartheid, claiming ancestral rights became a matter both of proving one's long-term presence within a territory, and being members of a distinct community who live and work together. Faced with incursions by neighbouring groups looking to utilise rangelands, and with new economic pressures that call for people to make their territories commercially viable, the 'territory' and the relationships it indexes have come to resonate in new ways. Suspicions, misunderstandings, and ill-feelings appear more likely, but overcoming them also appears more urgent.
Within this context, cases of sickness emerge as 'externalised idioms' (Howell 1981: 142) for the way people feel they should organise their relationships to one another, to themselves, and to the environments they live in. The ancestors who watch over their living relatives, respond both to moments when people have 'pain in their hearts' and to moments when people cause them harm and they are not aware of it. When viewed this way, striking a relative with sickness appears to be an act of care. There was no doubt among my Jú|'hoàn interlocutors, however, that the ancestors were 'not good' (|oa jan). They appear ambiguous, like typical trickster figures (Guenther 1999: 42), but to Jú|'hoàn speakers they are not. 'An ancestor is a person who only wants to kill people', Tshae explained, ' … they don't care that their relatives don't want to go to where they are', to the place people go to when they die. Like ancestors among the Vezo of Madagascar, ' … they miss the food, the drinks, the tobacco, the sex, the dancing that they enjoyed in life' (Astuti 2017: 110) and they look for any opportunity to be reunited with their loved ones. We see that among Jú|'hoànsi, however, they are not capricious. They strip their living relatives of vitality only when the breakdown of relationships among the living makes it reasonable to do so. The ancestors are attributed with causing the sickness that follows the breakdown of relationships between relatives. But there are also sicknesses caused by sorcerers, sicknesses caused by the Christian devil and its entourage of malevolent spirits, and sickness caused by certain viruses and bacteria (see Laws 2019b). 10 All form part of a local aetiology that associates sickness with the breakdown of social relationships or the transgression of certain values.
Through the mediatory work of shamans and potency of n|om, people are not only able to overcome the sickness caused by their ancestors, but to understand the transgressions or ill-feelings that motivated them. As shamans enter trance, they start to speak to the ancestors and parts of their conversations become audible to those who are present at the dance. The ancestors are part of the social fabric, and they know the many ways to be a 'good person'. They are late grandmothers and grandfathers of the people they strike with sickness, and they taught their living relatives what they know about 'giving thought' to othersby visiting, watching, and listening to them. As the trance continues, details emerge about the circumstances that preceded cases of sickness. Sometimes these include transgressions of what it means to be a 'good person'. At other times, they are merely suspicions, misunderstandings, or illfeelings. Since relatives are obligated to not confront one another about the pain they feel in their hearts, the ancestors, by watching over them and intervening in these moments, play a crucial role in confronting suspicions, misunderstandings, and ill-feelings. We see that shamans, far from simply restoring the vitality that ancestors strip away, play a crucial role in exposing the circumstances that motivate the ancestors in the first place. Shamans become all-seeing, all-hearing, and all-powerful with the effect of bringing doubts and suspicions into the open. They do this by drawing painfully upon the capacities of ancestors to do the same. The issue that shamans confront, then, is the problem of how to redistribute vitality, but also how to circumvent feelings of suspicion and doubt when people feel they cannot speak of them.

Sharing Responsibility
Throughout this paper, ancestors, and the shamans who act as mediators in their relationships with the living, emerge as armed with the capacity to see and to hear what is opaque to most. They use these capacities in order to attend, albeit indirectly, to problems of opacity. These problems of opacity emerge not only when relatives are drawn apart from one another, but when they avoid confronting one another. These problems of opacity have particularly high stakes in the Kalahari, where people depend heavily upon their friends and relatives making themselves present to one another to be demanded from or so they may sustain ongoing relationships of care. As they move between these spaces, problems of opacity become sources of suspicion or misunderstanding, and people are compelled to confront them. Among friends, uncertainties are typically addressed through mockery. This is a playful way of confronting suspicions and doubts without casting firm judgement and was something people did with me as much as with each other. Among relatives, by contrast, these forms of mockery are regarded not as playful but as antagonistic. As the phrase 'between you and your mother' suggests, these are relationships within which people are expected to trust and respect one another. While 'there are always personal stories and previous emotional encounters that colour the relation, often in contradictory ways' (Beatty 2013: 420), there is a discourse of tolerance that circumscribes the way people talk about their relatives. This prevents people from confronting their suspicions, but it also promotes an ethic of avoidance that dissuades people from telling each other what to do or from expressing disappointment when people do not help one another.
These patterns of generalised avoidance are key to sustaining relationships in the Kalahari, but they are also sources of tension as people feel unable to address the suspicions, misunderstandings, and ill-feelings that cause them pain. This culminates in the ancestors striking them with sickness or death in order to take them away from the acrimonious situations they find themselves in. Despite their own desires to share the company of their living relatives, the ancestors use these capacities only when there is acrimony between people and when it appears justified to do so. To say they are capricious overlooks that the ancestors, as much as their living relatives, are aware of the practices of avoidance that have both important, and challenging, consequences for social life in the Kalahari. Mediating these relationships between the living and the dead are shamans, who draw upon the capacities that ancestors possess to see and to hear what is ordinarily opaque. Through this process, the pain that motivates the ancestors become clear. So do the thoughts, feelings, and (at times) actions that cause people harm. Shamans, in response to the moves that ancestors make, thus govern opacityensuring relatives address social tensions without doing so directly. By decentralising confrontation, and making it an issue for ancestors and shamans, people maintain harmony with one another, but they do so without avoiding the problem of accountability. What appear at first to be transgressions, emerge as the outcome of suspicions, misunderstandings, and ill-feelings brought about by failures to 'give thought' to one anotherfailures to attend to specific relationships, not simply to the values that govern them. Notes 1. The data presented here were gathered over the course of fourteen months of ethnographic research in the Nyae Nyae Conservancy in the Otjozondjupa region of Namibia between October 2014 and December 2015. In addition to carrying out participant observation, focused on the way/s people navigate experiences of uncertainty, the author conducted a series of semi-structured interviews with 103 community members, of which approximately 22 address the themes presented here. The quotations provided are largely drawn from this corpus, and where they are not, they are records of more informal conversations between the author and those quoted. A summary of the aims and terms of research were translated into Jú|'hoànsi and Afrikaans and made available in both text and audio format. In accordance with the guidance offered by the Jú|'hoàn Traditional Authority of Tsumkwe District East and other key members of the community, the author did not request that participants formally sign an informed consent document. Research participants gave verbal consent and the author agreed to maintain contact. Follow up research on the topic was carried out in 2016, 2017, 2018, and 2019. 2. In a recent paper analysing 'day talk' and 'night talk' among Jú|'hoànsi within this region, between July-August and November 1974, Wiessner (2014) writes that there are marked differences in the conversations that Jú|'hoànsi have at night and those they have during the day, most notably in relation to talk about others. She notes that 'complaints and criticism' make up a large proportion of day talk but not night talk. This supports the claims this paper makes that there is not a generalised opacity doctrine among the Jú|'hoànsi. It is worth noting that it is not clear whether verbal criticism, complaints, and conflict (a) relate to relatives or to more distant kin, and (b) could be described as direct confrontation. Analysing these conversations more closely would provide compelling insights into how social life may have changed since then. Adopting a similar methodological approach in future research would also provide a more robust body of evidence against which to explore the hypothesis that cultivating opacity correlates with specific arrangements of care and that people maintain distinctions between 'kin' and others in the Kalahari. 3. While it is beyond the scope of this paper to explore this further, several productive parallels may be drawn between these expressions of mistrust within the context of economic uncertainty and broader writing on sub-Saharan Africamost notably on 'hustling', 'bluffing', and the 'informal economy' (notably Cole 2004;Newell 2009;Neves & du Toit 2012;Di Nunzio 2017;and Devlieger 2018). 4. Most healing is performed among a shaman's relatives, but the growing presence of outsiders and the decreasing numbers of shamans within the region has seen shamans seeking payment for their services (both from outsiders and from Jú|'hoànsi who are not their family members). This paper would benefit from further research addressing commercialisation, as well as the absence of healers, and its effect/s on 'governing opacity'. 5. This reading is supported further by Jú|'hoàn cosmogony (see Biesele 1993: 116-138), which speaks of a time when the animals were not 'branded'. During the 'creation' time, it was n| om that gave animals their distinctive characteristics. 6. Jú|'hoàn speakers use this suffix sparingly. They refer to someone as the 'owner' or 'master' of certain actions only when they have performed the action, or set of actions associated with the title, over a long period of time or particularly well. This translates into both a reluctance to praise people for acting in a way that is 'good' and a tolerance or acceptance of behaviour that is 'not good'. 7. Jú|'hoàn speakers have a repertoire of other dances in which the emphasis is not on seeing and speaking to ancestors, but in which they draw upon the capacities of, and transform into, other beings, most notably as lions (see Katz et al. 1997: 24-25;Marshall 1999: 238). 8. The Jú|'hoàn language has a wide range of 'praise names' or 'avoidance' language that they use to refer to humans, non-humans, animals, plants, weather features, and some objects that they respect (typically due to their distinctive character). Most notable among these are sorcerers, elephants, black mamba snakes, and winds. 9. Within everyday speech, this quality of 'otherness' becomes especially clear in the use of thirdperson pronouns. Other people are described not with the singular third-person pronoun ha or the dual and plural third-person pronouns sa and si!a, but the third-person pronoun hia pronoun used when referring to people who speak without clicks, to animals and birds, or to ancestors (see Laws 2019b: 153). 10. The term used to describe the ancestors (g||aoansi) is used in the singular (g||aoan) when describing the Christian devil. Where some church-goers regard these as the same, they are generally described as distinct figures. While missionisation has been extensive, the distinction between the Christian devil appears to map onto earlier distinctions between the ancestors as 'destroyers' (Schapera cited in Marshall 1976: 23), 'tricksters', or 'lesser gods' (Guenther 1999: 98) who dwelled in the sky along with the 'creator god', ǂKao n!a'an.