Navigation – Plan du site

AccueilNuméros166études et essaisReflections on Two Sticks

études et essais

Reflections on Two Sticks

Gender, Sexuality and Rainmaking
Todd Sanders
p. 285-314


Réflexions sur « deux bouts de bois ». Genre, sexualité et rites d’obtention de la pluie. – Tous les chercheurs qui ont étudié les croyances et les rites d’obtention de la pluie ont noté à leur propos la prééminence du symbolisme sexuel. Troublés par ce phénomène, ils se sont tournés vers les aspects sociaux, politiques et économiques de ces rites. À l’inverse, la perspective adoptée ici privilégie l’analyse des interprétations locales des rites d’obtention de la pluie chez les Ihanzu de Tanzanie, en avançant l’hypothèse que les rites de ce groupe et, par extrapolation, les rites similaires d’autres populations africaines sont tous saturés de symbolisme sexuel et que, de surcroît, ces rites prennent sens par rapport à des interprétations plus larges mettant en jeu les processus de reproduction. Pour les Ihanzu, en effet, toute (re)production (y compris celle des enfants et de la pluie) requiert une combinaison équilibrée des catégories mâles et femelles. Cette caractéristique tranche avec leurs notions ordinaires de genre, lesquelles supposent la hiérarchie entre les sexes et l’inégalité. Cet article tente ainsi de montrer que les notions concurrentes de genre et de pratiques de genre sont mises en oe uvre dans certains rituels et dispositifs de la vie de tous les jours. Enfin, si, à l’instar de tous ceux qui ont étudié les rites de pluie cet article met en évidence leur symbolisme sexuel, il va également plus loin en tentant de cerner, dans le cadre d’un terrain particulier, l’importance de ce symbolisme et les raisons pour lesquelles il prend précisément cette forme.

Haut de page

Texte intégral

Sexual matters bulk large in the art of the natives as they do in his life; but here it may be very difficult to understand his feelings. Since childhood he is a stranger to none of the natural functions. If he chooses a sexual theme for his song he does not go in for poetic freedom or make the most of the occasion. He is everywhere free and has everywhere his occasion. Lewdness is an invention of civilization (Koritschoner 1937: 52).

It is not hard to imagine the weight that the opposition between masculinity and femininity must bring on the construction of self-image and world-image when this opposition constitutes the fundamental principle of division of the social and the symbolic world (Bourdieu 1990: 78).

1It was yet another sweltering day in Ihanzu, north-central Tanzania. Although it was mid-February, the sun shone down on us with an unseasonable and unreasonable vengeance, the rains long overdue. Having wound our way through the bush and up the hill on a narrow, serpentine path earlier in the morning, we now sat in a small, dusty clearing. We had come to this secluded sacred site—most with great hopes, many in sheer desperation—to carry out an ancestral offering for rain.

2The sheep had already been sacrificed, and the grandchildren were now moving into the centre of the clearing to start the fire. The grandson grasped a long, slender stick between his outstretched palms. Placing its tip into a small hole in a second flat stick that the granddaughter held on the ground, he twirled the firedrill determinedly. At the same time he theatrically intoned an address to the spirits.

3At that moment, as I scribbled hurriedly in my notebook, an elderly man who had been sitting silently next to me all morning leaned over. “Do you see those two sticks?”, he enquired meekly. Ever the anthropologist, I continued to write, thinking this was a decidedly inopportune moment to deliberate over a couple of sticks. Smoke began to rise from the point where the sticks met, a certain sign that the fire would soon ignite and that the addresses would be over.

4When the addresses finally finished, I turned to the man. “What was that about the sticks?”, I asked. Smiling wryly, he spoke: “The long, slender stick, that’s the male. And the short, fat stick with the hole in it, that’s the female.” Apparently my expression betrayed puzzlement. He elaborated: “Look: the male is active, the female passive. Still, can either do anything alone? No, of course not. But if they co-operate, if they have sex, fire is born.” Grinning as if he had just given away the game, he leaned back and said no more. Although this elder no doubt felt otherwise, for me, this was the beginning, not the end, of my musings over the relationship between rainmaking, gender and ritual power.

5As the months passed, I participated in many similar ancestral offerings. And as before, other men and women spontaneously explained to me in nearly identical ways the centrality of the firedrill to those rites and, of course, the gendered and sexual symbolism that came with it. The aim of this article is to come to terms—if somewhat belatedly—with those two sticks. More broadly stated, my goal is to delve into the relationship between the cultural construction of gender, sexuality and rainmaking rites and beliefs in Ihanzu.

6Even though my interest in rainmaking, gender and fertility derives from one particular ethnographic encounter 1, the theoretical implications move well beyond it. Most anthropologists concerned with rainmaking in Africa, as good empiricists, have noted the centrality of gender and sexual symbolism in these rites and beliefs. More often than not, however, they have been puzzled by this, and have therefore either de-emphasised the topic or ignored it altogether, focusing instead on more instrumental social, political and economic aspects surrounding rainmaking (e.g., Beemer 1935; Colson 1948, 1977; Evans-Pritchard 1938; James 1972; Larson 1966; O’Brien 1983; Wollacott 1963). Important though these topics undoubtedly are, my contention is that our understanding of rainmaking in Africa will remain decidedly impoverished until we pay more careful attention to locally-informed understandings.

Rainmaking in Africa: A Brief Overview

7Rainmaking in Africa has long interested Westerner observers, academic and otherwise 2. And for as long as such observers have reported on African rainmaking rites and beliefs, they have reported on their blatant sexuality and the presence of things gendered. This has been the rule not the exception.

8All across the continent, for example, we find gendered rainstones (Cooke & Beaton 1939: 182; Hartnoll 1932: 738, 1942: 59; James 1972: 38; Middleton 1971: 196; Packard 1981: 69; Rogers 1927), some of which are themselves allegedly “capable of reproducing” (Avua 1968: 29). Also common are gendered rain pots (Cory 1951: 51n; Hauenstein 1967b: 13); male and female rain drums (Weatherby 1979); male and female rain statuettes (Ntudu 1939: 85; also Johnson 1948: 41, 96 [plate]); and even male and female rains (Bleek 1933: 308; Ginindza 1997: 152; Holas 1949; Marshall 1957; Sanders 1998). And this is just the beginning.

9More often than not, rain rituals themselves, much like the implements used in them, hint at the cultural salience of gender and sexuality (e.g. Håkansson 1998: 276; Hauenstein 1967b, 1967a; Hoernlé 1922; Jellicoe et al. 1968; Lindström 1987: 77; Murray 1980; Schoeman 1935; Ten Raa 1969; Vijfhuizen 1997). Songs sung during such rites are often sexually explicit or, if not, highly suggestive (Evans-Pritchard 1929; Jacobson-Widding 1990: 71, n7; Krige 1968). So, too, are the ancestral addresses for rain made during such rites (Leakey 1977: 203).

10By the same token, the opposite of rainmaking-that is, “rain-breaking”—is frequently associated with sexuality or the breach of taboos using genitalia. In some places a man may destroy the rain by urinating on a fire (Marshall 1957: 237) or “by raising his posterior to the clouds” (MacDonald 1890: 130).

11It is not difficult to imagine why gender and sexuality might feature centrality in most, if not all, African rain rites. For in most places and most cases, rain equals fertility on the grandest scale: without rain, people, plants and animals whither and die. In short, rain is life. It might thus seem banal—if not palpably tautological—to point out that rain rites, which are after all rites of fertility, are full of fertility symbolism. What is altogether more puzzling is why so few scholars have dwelt on this 3.

12Take, for instance, Elizabeth Colson’s landmark paper on rainmaking among the Tonga of Northern Rhodesia (today Zambia). In it she describes at some length Tonga rain rites, but only to tell us that she is “frankly uzzled as to why some are considered appropriate to the occasion. In one the dancer limped sadly about the shrine… Other dances were obscene” (Colson 1948: 279-280).

13In one of the few, full-length monographs devoted to the topic, Isaac Schapera repeatedly records but then either ignores or fails to appreciate the significance of sexuality in rainmaking. He was told that:

“‘In the old days’ widows and widowers were   doctored in groups outside the village’, in the presence of the whole tribe. Their clothing was removed, and when they were absolutely naked the doctors smeared their bodies with the juice of ‘stinging bulbs’ (digwere tsedibabang). They were then made to cohabit together sexually in public; if they refused… they might be killed on the spot. They were thought to be holding off the rain, and unless they were so treated the country would become ‘spoiled’ by the heat of the sun” (Schapera 1971: 123).

14Regrettably, Schapera seems to have made no further enquiries into the local logic that lay behind this and other statements like it 4. Nowhere in this excellent monograph, in fact, does he provide any coherent account of how the themes of sexuality, gender and fertility might be linked to rainmaking rites and beliefs more generally 5.

15Still others writing about rainmaking seem to shy away from sexuality altogether, even though the topic is, by their own accounts, not absent. In their monumental study, The Realm of a Rain-Queen,  Krige & Krige (1943) have almost nothing to say about sexuality in rainmaking rituals, except this:

“In ritual, sex is symbolized to some extent, but only as an aspect of fertility, not as a sensual pleasure, while obscenity is inconspicuous and must sometimes be rationalized, as if it were questionable even under the auspices of the ritual. The fertility theme is itself pushed far back, so that it is unrecognizable in the symbolic background” (Krige & Krige 1943: 290).

16We might be justified in wondering just how sex is symbolised as an aspect of fertility if at the same time it is “unrecognizable in the symbolic background”. Or why obscenities, if they are really so inconspicuous, must be rationalised at all. And to whom, exactly, must they be rationalised (except to their ethnographers)? But whatever the case, the noteworthy point is that it has been common for scholars writing about rainmaking to note the presence of gender and sexuality, only to dismiss these themes as insignificant for its comprehension. We can only guess at how many others might have failed completely to report similar themes in rain rites for their apparent irrelevance. It was only recently, after all, given impetus from feminist scholarship, that sexuality was “rediscovered” in anthropology as a topic worthy of theoretical consideration (Lindenbaum 1991; Tuzin 1991; Vance 1991).

17My aim in this article is to show how, amongst the Ihanzu of north-central Tanzania, people’s ideas about gender, sexuality and transformative processes inform and give meaning to certain rain rituals and everyday practices and vice versa. On the one hand, in the Ihanzu everyday world, gender representations and practices are multiple, contradictory and contested. There are three Ihanzu notions of gender: male superiority, female superiority and gender equality. Of these, the first is most commonly foregrounded in men’s and women’s day-to-day lives. Even so, men and women frequently find themselves engaged in a tacit struggle over the relative position of the genders. Resolution is always situationally defined and is thus fleeting. This is the state of affairs in the mundane world.

18In the ritual realm of rainmaking, on the other hand, matters are different. Disputes over the relative statuses and powers of the genders are temporarily resolved; the gendered state of affairs is agreed to by all. It is here that the ideology of gender symmetry or gender complementarity eclipses all others and becomes hegemonic (Ortner 1996). To be sure, this eclipse is not total. In rain rites there are still traces of gender asymmetries to be found. Competing gender representations are comprehensible only when defined one against the other (Moore 1994: ch. 3). Yet, on balance, when it comes to Ihanzu rain rites, men and women stress gender complementarity, by which I mean interdependent and equal 6.

19The reason people foreground notions of gender complementarity during rain rites, both conceptually and in practice, has to do with their ideas about reproductive processes broadly defined (also Sanders 1997, 1998, 2000). In Ihanzu eyes, cosmological transformations result only from the complementary combination of the cultural categories “male” and “female”. In other words, power comes in differently gendered pairs; and these pairs must combine as equals to ensure efficacy (Heald 1995). This is as true for making rain as it is for making babies. Underlying all Ihanzu transformative processes we find the same cosmological model, what might be termed a “procreative paradigm” (Herbert 1993) 7.

20Such broad notions of reproductive processes are sometimes formulated discursively, as the elder’s statement about “male” and “female” firesticks above reminds us. Yet it would be unwise to speak of culturally-specific ideas of fertility and transformative processes as “folk models”, if this requires we reduce such understandings to verbal exegesis alone (Holy & Stuchlik 1981; Jacobson-Widding et al. 1990). To limit our understandings in this way is to ignore the fact that many rituals are meaningful precisely because they do, rather than say things (Moore 1999). In this context, it is their illocutionary force that is at issue. Rituals, for those who perform them, are culturally-appropriate ways of acting upon the world. They are meant to make things happen. And this is particularly so in the context of rainmaking, where the issue is not simply one of symbolic representation, but of animating the cosmic and divine powers of the universe and effecting change (Buxton 1973: 358-359). In fact, it is precisely this practical engagement with the world that enables such rites convincingly to range across, and draw their cosmic powers from, a number of separate yet interrelated cultural domains. Making babies requires the genders to combine equally. So, too, does rainmaking. Before showing how and why this is so, we need to elaborate on Ihanzu notions of gender, and how they are operationalised in different everyday and ritual settings.

The Ihanzu of Tanzania

21The Ihanzu of north-central Tanzania are a Bantu-speaking, matrilineal people who number around 30,000. Although some keep cattle, goats and sheep, they are first and foremost agriculturalists and largely imagine themselves as such. Sorghum, millet and maize are the staple crops and, together with an assortment of wild greens, milk and dried fish, provide the bulk of their diet.

22Agriculturalists and part-time pastoralists the Ihanzu may be, but farming and herding in this arid region have never been easy. Annual rainfall averages a meagre 20-30 inches. Even in the best of years rainfall is unpredictable and patchy. Consequently, droughts and famines are not uncommon. Furthermore, there are no year-round rivers that might alleviate or lessen these difficulties. That the rains arrive on time—or indeed, that they arrive at all—and fall regularly is, quite literally, a matter of life or death for all concerned. This given, it should come as no surprise that rainmaking rites and beliefs feature prominently in Ihanzu today, as they have done for well over a century.

23There are two Ihanzu ritual leaders (akola ihĩ) whose job it is to organise and orchestrate all rain rites. One of these leaders is male, the other female. To understand the significance of this royal gendered duo, and the pivotal role they play in the Ihanzu cultural imagination, it is helpful to explore briefly Ihanzu gender notions and practices.

24There are three discernible Ihanzu notions of gender, each of which is salient for men and women in different contexts 8. The first such notion, that male is superior to female, permeates much of the Ihanzu lived-in world and informs people’s ideas and ideals about an array of daily practices. This is most commonly the picture Ihanzu men and women provide when asked to sum up gender relations among them.

25This discourse asserts the frivolity of the feminine, suggesting that women are sometimes simple, often irresponsible and largely lacking in foresight. Many daily activities are justified on these grounds alone. The when’s, where’s and what’s of planting, for instance, are almost always decisions made by men, often on the grounds that women cannot understand the complexity of these matters and would therefore make either the wrong decisions, or no decisions at all. Similarly, ownership of and control over livestock fall roundly into the male domain, again, because women are often said to be incapable of caring for them over the long run. Men own nearly all livestock. They inherit it, use it for bridewealth and make all important decisions concerning its sale and slaughter. To a large extent, livestock— but most especially cattle—is “men’s business” and women are excluded from it. Livestock are thus male-coded 9. In the political domain, too, it is generally men and not women who play an active role. Whether for witchcraft, theft or something else, local vigilante (nkĩlĩ) meetings are always men’s concerns, never women’s.

26Whatever salience this first disparaging discourse about male-female relations might possess, it is, in other situations, thoroughly undermined by another suggestion: that females are superior to males.

27Together a man and his wife farm their fields. Yet once the grain is harvested and has been safely stowed in the grainstore, a wife gains close to total control over its allocation within and outside the household. With her grain, a woman must budget from one harvest to the next. She must decide how much she can afford to give away to needy neighbours, kin and others, and still have enough to provide for herself, her husband and children 10. She must additionally take decisions about when and how many times to brew beer, the single largest contributor to household income, the foundation of day-to-day social life in the villages as well as the bedrock of the local cash economy. Beer brewing is associated with women as it is always they, not men, who brew. It is also women who control the proceeds of any given brew. These are most often used for the reproduction of their own households.

28For all these reasons, in domestic contexts, women are often symbolically associated with grain and especially with stiff porridge and beer, the indispensable grain products of everyday life. More than this, in a very real sense control over grain is a source of women’s power and status. Women control the ebb and flow of the local economy and village sociality by controlling grain and its by-products.

29In this context women and men alike generally recognise female superiority. Not infrequently men portray themselves (often accurately) as domestically incompetent and therefore at the mercy of wives, mothers, sisters and other women. As one middle-aged Ihanzu man put it:

“Sometimes women are more powerful than we [men]. We depend on them. [. . .]. A man may live in a house alone but without a wife he is helpless, or at least he has a lot of difficulties. He must fetch his own water, cut his own firewood, cook his own food. [. . .] We depend on women like we depend on rain. This is why I say that, if you look at it from another angle, perhaps it is really the women who are the important ones, the most powerful.”

30Ideas of female superiority are sometimes couched in biological “facts”. As elsewhere in the region (Beidelman 1964: 377, 1993: 39), the Ihanzu link female bodies with wetness (atotu) and male bodies with dryness (akalamũku). Men and women claim these biological differences are obvious, evinced by women’s “moist” vaginas and “fatty” bodies and men’s “dry” penises and lean bodies. Wetness is associated with rain and fertility and is thus highly auspicious. Dryness, in contrast, connotes precisely the opposite. The fact that women are seen as “naturally” wetter than men suggests that women are in some sense “naturally” better than men when it comes to individual and social reproduction upon which all depend. On occasion, I have heard men loath this apparent fact.

31Thus, in everyday contexts, the Ihanzu hold competing views on the nature of gender relations. In many contexts they see men as superior to women. Indeed, this is the dominant gender discourse, and it pervades a number of domains of daily life. But in certain contexts—namely those related to reproduction of the household and images of male and female bodies—it is palpable to all that women rule the roost. In these situations, both on an ideational level and in practice, women appear to hold the upper hand. The relative status of the genders in Ihanzu is therefore situationally-defined. Under such circumstances, to gloss Ihanzu gender relations monolithically as “patriarchal”, “matriarchal” or anything else would be highly misleading, all the more so when we consider a third and final Ihanzu notion of gender: gender complementarity or symmetry.

32In still other contexts, male and female each has his or her own unique abilities and tasks, none of which is thought to be inherently more significant than any other. No relative evaluation of gender categories is implied. In such contexts male is said to be active, female passive; male leads, female follows; male is above, female below. In all cases, it is the interdependence and mutually defining character of gender categories that is at issue. Male and female complement and give meaning to each other. There are several cultural and social arenas in which people strongly emphasise this particular representation of the genders. One of the most evident is in procreation beliefs.

33The Ihanzu believe that both men and women possess fertilising fluids (manala) which are required in equal amounts to create a child. These fluids, frequently referred to euphemistically as “waters” (mazĩ) or “seeds” (mbeũ), are virtually identical, but are nevertheless in some sense gendered. Men have male fluids; women have female fluids. In concert, though not separately, the two are potent. When it comes to human reproduction in the abstract, all I spoke with agree that male and female are equal (Sanders 1998). This is similarly the case for coitus.

34It is correct and proper that men initiate sex. Women should not. Male is active, female passive. Men, also, should be on top, women underneath. In making such statements, many men and women insist that sex is about male and female interdependence—the correct order of things—and that no gender hierarchy is implied. Sex, in other words, people often take as evidence of the underlying equality and mutually interdependent nature of the cultural categories “male” and “female”. Both genders must be combined as equals for successful reproduction. Only when joined are the genders ritually powerful and capable of bringing about change.

35At life’s end, as at its beginning, notions of gender complementarity are manifest. When people in Ihanzu die their souls (nkolo) are said to enter the underworld (ũlũngũ), immediately becoming ancestral spirits (alũngũ; sing. mũlũngũ11. These spirits do not dissolve into any androgynous collectivity (Bloch 1987: 326-328) but, quite the contrary, maintain their distinctive gender identities. Both male and female spirits are equally thought capable of afflicting the living (see also Kohl-Larsen 1943: 293). In spite of suggestions to the contrary (McKnight 1967), in Ihanzu at least, there is no apparent preponderance of afflictions by either maternal or paternal spirits, nor are there any discernible differences between types or severity of afflictions caused by male and female spirits. More often than not, in fact, it is male and female spirits together that are ultimately responsible for any given affliction. This is as true for personal affliction as it is for spiritual intervention in matters of rain.

36To sum up so far, both at life’s beginning and at its end male and female are often viewed as equal and complementary. It is only in-between that things are much less decided, the actual statuses and relative powers of the genders being questioned and contested.

37Resolution is context dependent and hence, at best, temporary. Household grain—and more especially, grain transformed into stiff porridge or beer—embodies ideas of female power and status, in the same way that livestock, as part of the male domain, symbolically asserts male power and status. Procreation beliefs and death, on the other hand, tend to emphasis symmetry between the genders. So, too, as we shall see in the remainder of this article, do rainmaking rituals and beliefs.

Ancestral Rain Offerings

38The Ihanzu have a number of different rain rites. Some are performed annually, at the onset of each new farming season. Others are performed only in those years when the annual rites fail to bring the rain. Whatever type they may be, all such rites are informed by the same underlying logic of reproduction, what I have elsewhere called gender complementarity (Sanders 1998, 2000).

39The focus here is on ancestral offerings for rain (mapolyo a mbula),  one among several types of Ihanzu rainmaking rites 12. These rites take place only when the rains have utterly failed and it has been divined that the royal Anyampanda clan spirits have demanded such an offering. Offerings take place over two days, but the entire ritual sequence often lasts a month, sometimes longer. It is only the two Anyampanda royal leaders, and no one else, who can bring such rain offerings to fruition.

40To initiate the offering (kũkũmbĩka),  which is done the evening of the day of the divination session, a few “grandchildren” are summoned to the male ritual leader’s homestead. These grandchildren (always classificatory grandchildren) play a central role in all such rain offerings. They must initiate all ritual activities. It is imperative, too, informants stress, that at least one grandson and one granddaughter participate. Both genders must be present (cf. Kenyatta 1959: 245).

41The grandchildren place some white sorghum flour and water, together with some ritually-significant tree branches, into a special long-necked calabash (mũmbũ). While addressing the royal clan spirits, they set the calabash in the doorway of the male ritual leader’s homestead. Both grandchildren address the spirits; the grandson’s addresses always precede the granddaughter’s. Such addresses are brief, usually stating the obvious—“We are now putting your beer here in the doorway”, etc.—and are made throughout the offering whenever the grandchildren initiate a new task. There are three addresses, above all others, that are of decided importance in these offerings. I will discuss these presently.

42Several days later the grandchildren begin brewing beer. Unlike in everyday contexts, however, where it is only women who brew beer, for these offerings granddaughter and grandson must brew beer together. Over the course of a few weeks, at each stage in the brewing process—addressing the spirits, digging the beer brewing trench, collecting firewood and so forth—these two must co-operate (kiunga) and “reside together harmoniously” (wikii ũza palũng’wi). For reasons that will become apparent below, it is worth noting that the beer brewing trench (ilũngũ) must be dug on an east-west axis, which again differs from everyday beer brewing.

43On the final day of brewing, the grandson and granddaughter set aside a small amount of ritually significant beer (kinyaũlũngũ). Just before sundown, they visit a few significant places where, as before, they address the spirits. The places normally visited are, first, the royal clan offering sight itself and, second, the graves of a few former ritual leaders, both male and female.

44The following morning, the male ritual leader leads the way from his homestead to the royal offering site, a small mountain called Ng’waũngũ in the village of Kirumi. Everyone follows him in single file, walking slowly, deliberately, speaking either in quiet tones, or not at all.

45Most of the party ascends the mountain with the male ritual leader to a small clearing. The female ritual leader, for her part, remains at another clearing at the bottom of the mountain with a number of elderly women. It is their job to prepare castor-seed oil (mono) to be used later in the offering. Throughout the offering these women sing songs they conventionally sing at the birth of twins (ipaha),  the women’s rain dance (isĩmpũlya) and girl’s initiation ceremonies (mũlĩmũ). Most of these songs are blatantly sexual. Many are considered obscene.

46In the principle offering clearing, people sit in distinct groups. Diviners, together with the male ritual leader and some male elders, are expected to sit at the highest point in the clearing. Most elderly men sit slightly below them. Women sit further down the clearing and in the path, while those who have never before attended such an offering, young men and women, sit in the lowest position. Picture 1 shows the relative positions of men and women at one such offering I attended. These particular seating arrangements are not accidental (Obst 1912: 116n), and follow a specific cultural logical I will explore below.


Picture 1. — Ancestral Offering Seating Diagram


49When everyone is seated, the grandson continues up the path with some beer (sometimes with a rainmaking assistant) to one of two sacred clan caves. After removing his shirt, he sprays some of the ancestral beer over the cave entrance and briefly addresses the spirits.

50The grandson then collects a number of tree branches and places them in the centre of the clearing, cut ends to the east, leaves to the west. The two grandchildren lay down the sacrificial animal—most commonly, a black sheep born at night—with its head to the west, atop the leafy branches. Together the grandchildren smother the animal until it passes out, at which time a Muslim slits its throat 13.


Picture 2. — Grandson starting fire with firedrill, while granddaughter holds the base


Picture 3. — Granddaughter starting fire with firedrill,
while grandson holds the base


54After skinning the animal, the grandchildren start a fire “in the traditional way” (kijadi) by twirling a long, slender firedrill (kĩlĩndĩ) into a hole in a smaller, stationary hearth (kiziga) (see Pictures 2 & 3). While so doing, each addresses the spirits aloud. One such address, the one that began this article, went as follows:

“You, Mũnyankalĩ, who come from your senior house and are going to your junior one, you have passed through Ihanzu and have seen we are carrying out an ancestral offering, an offering in the cave. We have an offering for rain. We are offering [you] water [i.e., beer], and a sheep that was born at night. Take good news to the place you are going; and the bad, toss it into the waters of Lake Victoria” 14.

55These addresses are always made to Mũnyankalĩ, the name used for the sun in ritual contexts. Being neither God nor spirit, Mũnyankalĩ may be understood as “a visible and tangible symbol of a supernatural world about which nothing can be known” (Adam 1963: 22). In these addresses, reference is customarily made to the fact that Mũnyankalĩ—who is unequivocally said to be male—is moving from one wife’s house in the east to another wife’s house in the west (Adam 1963: 11-12, 22; Kohl-Larsen 1943: 303-305). As he does so, grandchildren urge him to make it known to the spirits that an offering is in progress in Ihanzu. Typically, too, they tell Mũnyankalĩ to remove whatever evil there might be, like witchcraft, and to cool it in the waters of Lake Victoria, the (perhaps mythical) homeland of the Ihanzu people.

56Following this comes one of the most significant addresses of offering: giving meat to the spirits (kũtagangĩla). Collecting several roasted pieces of the sacrificial meat, the grandchildren make an address to each of the four cardinal points, tossing, before each address, a piece of meat in that direction. As always, the grandson’s address precedes the granddaughter’s.

57This sequence of addresses completed, the diviner (or in some cases, diviners) reads various entrails in an oblong, divining bowl (ntua) (Picture 4). The bowl must be oriented east-west, like the beer trench and sacrificial animal. The entrails invariably tell of the spirits’ gratitude for the offering, of plentiful rains and of other things auspicious (Adam 1963; Obst 1923: 221-222).


Picture 4. — Diviner examining the sacrificial animal’s lungs
to see whether the spirits have accepted the offering


60Following the divination session, the women from the base of the mountain dance and sing their way up the path, through the main clearing, to the cave. Both grandchildren take a handful of chyme from the sacrificial sheep and join them. At the cave, one at a time, the grandchildren address the spirits and toss the chyme around the entrance, “to cool” (kũpola) the spirits. The grandson then descends to the clearing. The female ritual leader, the granddaughter and a few other elderly women of the royal clan remove their clothes and, carrying the half-gourd of castor-seed oil, enter the cave. After the granddaughter addresses the spirits, the women anoint some ancient cave drums with oil 15. They enter a second, nearby sacred cave too, naked, where they anoint an enormous ancestor-snake (usually described as a python) that allegedly lives there. The women then don their clothes and join the other women in the clearing below.

61At this point, the grandchildren roast and dole out meat from the sacrificial animal. Most eat in their respective groups, while a few ad hoc groups visit and feast on the royal graves visited the previous day by the grandchildren. The loin, which people explicitly associate with giving birth, is always eaten on a female ritual leader’s grave, while the a front leg is eaten on a male ritual leader’s grave. When all have eaten, they return to the male ritual leader’s homestead.

62Once there, people sing rain songs, many of which are sexually explicit or suggestive. In preparation for the second major address of the offering—giving beer to the spirits (kũlonga shalo)—the male ritual leader sits in the doorway to his house. The grandson initiates the offering. He stands in the centre of the courtyard holding a ritual whisk (nsing’wanda); at his feet sits a divining bowl (again, placed east-west) filled with ancestral beer and water. Dipping the whisk into the bowl and splashing it to the east, he begins his theatrically intoned address to the spirits, repeating this procedure to the west, south and then north. When he finishes, the granddaughter does the same. The day’s events are brought to a close by addresses from the royal leaders’ classificatory father and, finally, a jester. People continue to drink beer, often late into the night, and are expected to sleep on the ritual leader’s homestead.

63The next morning, the final day of the offering, the grandchildren arise early. The granddaughter sets off cooking stiff sorghum porridge, while the grandson roasts the meat remaining from the previous day’s sacrifice. The grandson and granddaughter divide the stiff porridge in half and put the two portions into two separate calabash bowls. They also divide the roasted meat in this way. Each takes a bowl of stiff porridge and meat.

64At sunrise, in the centre of the courtyard, the grandson begins the third and final significant address of the offering. He tosses a piece of roasted meat to the east, immediately followed by a piece of stiff porridge. As the previous day, he addresses Mũnyankalĩ in the east. He repeats the sequence to the west, then to the north and south. The granddaughter follows the grandson, doing the same.

65Addresses completed, the grandchildren feed small pieces of roasted meat and stiff porridge to the male and female ritual leaders who sit idle throughout. They then feed a few others from the royal clan, as well as a group of senior men and the diviner. From inside the house the grandchildren bring out bowls of roasted meat and stiff porridge—two each—which are then served to the group of younger men, and finally, the women.

66Following the meal the grandchildren distribute ancestral beer likewise: clan elders, younger men and then women. A few elderly men and women, including the male and female ritual leaders, then enter and drink beer inside the house. The beer is served from two ritually-significant long-neck calabashes or mũmbũ,  also the word for “womb”. These calabashes wear white beads around their necks, beads that are today associated both with the ancestors who wore them in abundance (Obst 1923: 222), and with the powers of fertility the ancestors control (see Picture 5).


Picture 5. — Two grandchildren inside the house displaying their beer-filled,
ritually-significant calabashes or “wombs”


69The grandchildren briefly address the spirits, one last time, and cover over the beer trench they used to brew the beer. This rite officially marks the end of the offering.


70Ihanzu ancestral offerings for rain are far from simple affairs. This much is clear. If one takes into account ancestral beer brewing, these rites can easily last a month, sometimes longer.

71Equally evident is that Ihanzu ancestral offerings are replete with fertility and sexual symbolism. Throughout, ritual participants play, sometimes in rather striking ways, on themes of gender difference and complementarity.

72One of the more obvious ways this is done is by using two differently gendered grandchildren to initiate and conduct each and every step in the ritual process. They must “co-operate”, so people say, and “reside together harmoniously” in everything they do. It is also they, and they alone, who must use the gendered firedrill: to cause the sticks to have “sex” so fire is “born”. Grandchildren’s addresses similarly allude to locally-inflected understandings of gender and sexuality by focusing much attention on Mũnyankalĩ, the this-worldly male symbol who moves each day between his two wives’ houses. Finally, the songs sung also hint at the salience of fertility and sexuality in these rites.

73Yet—and this is my point—to note simply that these rain rites, or any others, are full of fertility and sexual symbolism, as many Africanist scholars have done, provides us with few clues as to why this might be so. Nor does it explain why ritual participants feel such rites are both appropriate and efficacious. In other words, the question of interest is “Why do Ihanzu rain rites, or similar rites elsewhere in Africa, often play on themes of gender, sexuality and fertility?” The answer, I suggest, is because such rites are linked to, and broadly informed by, local conceptions of reproductive processes. As noted above, in Ihanzu eyes, this implies the notion of gender complementarity. It now remains to trace explicitly the linkages between Ihanzu notions of reproductive processes on the one hand, and ancestral offerings for rain on the other. More to the point, the task is to demonstrate that Ihanzu rain rites foreground themes of gender symmetry or gender complementarity, while simultaneously de-emphasising themes of gender hierarchy and difference so prevalent in day-to-day life. Since Mũnyankalĩ features centrally in the principle ancestral addresses during all ancestral offerings for rain, we would do well to begin with him.

The “Sun God” and Spatiality

74Ihanzu men and women have surprisingly little to say about Mũnyankalĩ, surprising, that is, given his centrality in these rain rites. First, everyone seems to agree, he is male 16. And second, the sun is his visible, this-worldly representation. No one I spoke with could say more. This apparent paucity of information, however, is itself instructive. For the little people do say goes a long way towards explaining why they need say no more.

75Mũnyankalĩ is male and, as the sun, moves from east to west each day between his two wives’ houses. All spatial references during the offering, including the orientation of the beer trench and divining bowl, as well as the tossing of meat and porridge, people relate explicitly to Mũnyankalĩ’s daily movements across the sky.

76Each day, Mũnyankalĩ moves from east to west, a movement associated with cycles of life and death. For the men and women of Ihanzu, the east and morning are associated with birth, upwards, growth and renewal. Mũnyankalĩ is reborn each morning. For this reason the east is auspicious. Life-giving rains come from the east. Each year the royal rainshrine is opened in the morning, when the sun is rising. And divination sessions, which usually occur in the morning, face east. All ancestral offerings, too, take place in the morning.

77The west, on the other hand, is inauspicious. It is commonly associated with death, downwards and decay. Mũnyankalĩ dies here each evening. Ihanzu graves have east-west orientations, and the head is invariably placed to the west. When, on occasion, the wind causes rain to fall from the west, as infrequently happens, people say this is bad rain (mbula mbi) that brings aphids, lightning and other things unpropitious. When the rainshrine is closed each year, it is done in the evening, when the sun is setting in the west.

78Ancestral addresses to Mũnyankalĩ, like all east-west spatial references in these rites, therefore serve as a spatial commentary on an idealised gender order. This is an order, as the Ihanzu see it, of gender complementarity. Male and female work and live together harmoniously. Here, male (the sun) is “active”. Female (his wives) is “passive”. Also worthy of note is that there is never a time—in the daily cycle, in the individual life-cycle or in the larger cosmic movement between the living and the dead—when male and female are apart. They remain together but distinct.

79Other spatial orientations also suggest ritual participants are actively working at an idealised gender order. Recall, for a moment, the relative positioning of men and women at the offering site itself (Figure 1). Men sit above, women below. Between these two differently gendered groups lies the fire, a ritually transformative agent which is frequently used as a metaphor for sex, much like the gendered firesticks that started it. Within these groups, the male ritual leader sits in one of the highest positions; while the female ritual leader remains for most of the offering at the lowest position. This is not only of anthropological interest, but of local interest too. At one such offering I attended, several young women who had not previously attended an offering initially sat above the elderly men. They were sharply rebuked by other women and men and made to relocate to their proper position “below”. These particular seating arrangements show a concern with strategically placing the genders to reflect an ideal of gender complementarity. Remember that “above” and “below” are ideally considered “male” and “female” respectively and that no gender hierarchy is implied in this context.

80Likewise, during these offerings, grandchildren address both male and female spirits. Male and female spirits (alũngũ) are generally said to co-operate in the underworld. All this-worldly spiritual afflictions, drought included, are thought to be caused by male and female spirits working together. It is for this reason that, in Ihanzu, male and female spirits must be addressed and placated together to heal the ill or to bring rain.

Ritual Sequence

81If the contents of ancestral addresses and the spatial layout evoke an idealised order of gender complementarity, then so, too, does the temporal sequence of these addresses. Grandson and granddaughter make all addresses jointly, but never simultaneously. The grandson must go first, the granddaughter second. This alludes once again to an idealised gender order of complementarity where male precedes female. Both work together; each fulfils his or her role. The overall sequence of addresses operates in a similar fashion.

82The grandchildren make numerous addresses throughout the offering. These are usually brief, and do little more than state the obvious (i.e., we are making an offering; we are brewing beer; we are digging a trench; etc.). There are three addresses in particular that people single out as more significant than all others. In the order they occur these are, firstly, giving meat to the spirits (kũtagangĩla); secondly, giving beer to the spirits (kũlonga shalo); and thirdly, on the final day, giving the spirits meat and stiff porridge simultaneously. In light of the symbolic associations between female and grain on the one hand, and male and livestock on the other, the gender symbolism is once again manifest. So is the emphasis on gender complementarity.

83Grandchildren make the first of these addresses by tossing meat which, given its association with men’s control over livestock, is symbolically male-coded. This address condenses within it the male elements of Ihanzu cosmological universe. The second significant address is made with sorghum beer. Given the everyday symbolic associations between grain, beer and women, this second address might be seen as embodying feminine elements of the cosmos. Taken together, the two addresses realise the idealised gender order in which male precedes female. Lest there be any ambiguity as to the relative status of gender categories in these two addresses, the final address to the spirits, made on the final day of the offering, is a blatant assertion of gender equality. Here, the grandchildren combine as equals masculine and feminine elements by tossing meat and stiff porridge in the same address. It is at this point instructive to turn more directly to the principle ritual officiants, the grandchildren themselves.

Grandchildren and Gender Complementarity

84Grandchildren play a central role throughout ancestral offerings for rain. Indeed, without a granddaughter and grandson co-operating and working together, such offerings cannot take place. As we have seen, both the content and sequence of their addresses are gendered. So, too, is the firedrill that features so centrally in these rites, as the opening vignette to this article suggests. The firedrill is “above”, “active” and thus male; the hearth, the stationary piece of wood with a hole in it, is “below”, “passive” and thus female.

85But what makes granddaughters and grandsons appropriate, above all others, as ritual officiants on such occasions? What distinguishes the grandchild-grandparent relationship from other possible relationships, and makes it ritually significant? To answer this question requires a brief discussion of this important relationship.

86In Ihanzu it is not uncommon for grandparents to look after their young grandchildren. Children may reside for short or long periods with their grandparents when, for example, parents carry out migrant labour in other parts of Tanzania. Or, in other cases, children may grow up living with grandparents, especially in those cases where the latter require regular assistance with everyday domestic chores.

87Grandparents do not discipline their grandchildren. This they leave to the child’s parents. Today as in the past, the grandparent-grandchild relationship in Ihanzu is one of easiness and mild teasing (maheko), “affection and equality” (Adam 1963: 23). This equality is manifest structurally in several ways. First, grandchildren are frequently named after their grandparents. Second, grandparents and grandchildren often refer to each other as siblings (aheu or ng’waitu). Consequently, the grandparent-grandchild relationship is marked by a notion of equality that is never found, for instance, in parent-child relationships (Turner 1955).

88A corollary of this structural equality is that grandsons may woo their grandmothers and vice-versa. There is, in fact, a good deal of sexual joking between grandparents and grandchildren. Such joking, which is common elsewhere in Africa (Beidelman 1997: 63), often revolves around the idea that a grandson is scheming to steal ‘his’ rightful wife (that is, his own grandmother) away from his grandfather. Not only this, but it is precisely because grandchildren and grandparents are structurally located at opposite ends of the generational cycle that they are considered appropriate for the occasion. For both are, in a sense, sexually “pure”: ideally, neither grandchildren nor grandparents are sexually active, the former because they are too young, the latter too old. And it is this very ideal of sexual “purity” that allows them to “play with fire”—that is, to play with themes of gender and sexuality—most freely and to the greatest effect 17.

89With its heavy emphasis on both equality and sexuality, the Ihanzu grandparent-grandchild relationship thus presents itself as an ideal one for enshrining and acting out notions of gender complementarity necessary for rain offerings. It is equal in a way that other relationships like parent-child simply are not. It is also the most blatantly “sexual” relationship, albeit in a joking manner. Although the husband-wife relationship might be considered a suitable alternative, this relationship is rarely marked by the same notion of equality between the genders required for rain rites.

90Above all else, by acting out the principles of gender complementarity, while at the same time relegating ideologies on gender asymmetry to the symbolic background, ritual participants seek to create an alternative vision of their social and natural worlds: one where male and female co-operate, reside harmoniously together and where, as a result, fertility flows in abundance. Grandson and granddaughter, being differently gendered and equal, are best positioned to do this. With the assistance of others, they act out, evoke and embody the principle of gender complementarity. As with gendered fire sticks, creation, rejuvenation and transformation all require that male and female unite as separate but equal entities. A solitary stick, like a solitary gender, has no meaning.


92In this article, I have shown that Ihanzu ancestral offerings for rain, in varied ways, play on themes of gender and sexuality. The reason for this, I submit, is that such rain rites are informed by a particular cultural logic of transformative processes, a logic of gender complementarity. It is this culturally-specific notion of transformation that makes such sexual symbolism both meaningful and necessary in these rites. If rain rites are to be effective, if they are to bring rain, then the Ihanzu cultural categories “male” and “female” must combine as equals. There is no other way. Ritual participants thus pay close attention to acting out, through ritual practice, this particular proposition.

93Identical cultural notions of reproductive processes undergird Ihanzu men’s and women’s ideas about procreation. To create children, male and female must unite as equals. Men and women each have semen, and these differently gendered semens must combine in equal amounts. Once again, male and female must co-operate to bring about change.

94This cosmic congruence between rainmaking beliefs and ideas about procreation is not the result of one leading directly to the other. Even though people speak of the gendered “waters” required for procreation, as well as the sexuality of rainmaking, no one in Ihanzu would confuse sex with rainmaking. The point worth stressing, rather, is that underlying both (re)productive, transformative processes lies the same notion: that cosmic power comes in differently gendered pairs. And to accomplish their respective goals, these pairs must combine as equals. What is more, such principles need not be explicitly formulated in order to do what they purportedly do. To discuss sex, after all, will not bring a child any more than discussing sexually-laden rain-rites will bring the rain. In both cases action speaks louder than words.

95Although cosmic transformation for the Ihanzu is about uniting masculine and feminine as equals, this is not everywhere the case. Nor need it be. It is of course possible to imagine transformative processes in all sorts of ways, one of which is asymmetrical reproduction where masculine is superior to feminine (e.g., Brandström 1990). Consider, briefly, the Kaguru of Tanzania who share a number of cultural, social and linguistic similarities with the Ihanzu. For them:

“[K]indling a fire clearly alludes to sexual intercourse. In making fire with firestick, Kaguru use an active stick of hardwood and a passive stick of soft and easily combustible wood. The passive stick, the one drilled, is female” (Beidelman 1997: 205).

96When it comes to firesticks, Kaguru thinking about gender and sexual symbolism is strikingly similar to that of the Ihanzu (see also Århem 1985: 13; Jacobson-Widding 1990: 68). Even so, the Kaguru construct fire-making not as a domain of gender equality but as one of male domination (Beidelman 1997: 205). This implies, among other things, an ideology of reproduction or transformative processes that is not about gender equality but hierarchy. These differences, naturally, are empirical not theoretical questions. The point, to reiterate, is that the questions need asking in the first place. How are people’s different and competing notions of gender linked to specific types of transformative processes like rainmaking?

97In conclusion, it is simply not enough to note that African rainmaking rites play on themes of gender and sexuality, patently true though such observations may be. To stop here is to ignore most of what is interesting about rain rites in Africa. Instead, we must interrogate and problematise these issues—which means coming to terms with locally—inflected notions of reproduction and transformative processes, always, of course, situating them soundly within specific cultural and social contexts. For at the end of the day, male and female firesticks may be just two more sticks, like any others. But they also promise to be much more.


London School of Economics and Political Science, London.

Haut de page


Adam, V.

1963  “Rain Making Rites in Ihanzu”, Conference Proceedings from the East African Institute of Social Research, Makerere College.

Århem, K.

1985  The Symbolic World of the Maasai Homestead (Uppsala: African Studies Programme, University of Uppsala) (“Working Papers in African Studies” 10).

Avua, L.

1968  “Droughtmaking among the Lugbara”, The Uganda Journal 32 (1): 29-38.

Beemer, H.

1935  “The Swazi Rain Ceremony (Critical Comments on P. J. Schoeman’s Article)”, Bantu Studies 9: 273-280.

Beidelman, T. O.

1964  “Pig (Guluwe): an Essay on Ngulu Sexual Symbolism and Ceremony”, Southwestern Journal of Anthropology 20: 359-392.

1993  Moral Imagination in Kaguru Modes of Thought (Washington D.C.: Smithsonian Institution Press).

1997  The Cool Knife: Imagery of Gender, Sexuality, and Moral Education in Kaguru Initiation Ritual (Washington D.C.: Smithsonian Institution Press).

Bleek, D. F.

1933  “Beliefs and Customs of the /Xam Bushman. Part 5: the Rain”, Bantu Studies 7: 297-312.

Bloch, M.

1987  “Descent and Sources of Contradiction in Representations of Women and Kinship”, in J. F. Collier & S. J. Yanagisako, eds, Gender and Kinship: Essays Toward a Unified Analysis (Stanford: University Press): 324-337.

Bourdieu, P.

1990  The Logic of Practice (Stanford: University Press).

Brandström, P.

1990  “Seeds and Soil: the Quest for Life and the Domestication of Fertility in Sukuma-Nyamwezi Thought and Reality”, in A. Jacobson-Widding & W. van Beek, eds, The Creative Communion: African Folk Models of Fertility and the Regeneration of Life (Uppsala: Acta Universitatis Upsaliensis): 167-186.

Buxton, J.

1973  Religion and Healing in Mandari (Oxford: Clarendon Press).

Colson, E.

1948  “Rain-shrines of the Plateau Tonga of Northern Rhodesia”, Africa 18: 272-283.

1977  “A Continuing Dialogue: Prophets and Local Shrines among the Tonga of Zambia”, in R. P. Werbner, ed., Regional Cults (London: Academic Press): 119-139.

Cooke, R. C. & Beaton, A. C.

1939  “Bari Rain Cults. Fur Rain Cults and Ceremonies”, Sudan Notes and Records 22: 181-203.

Cory, H.

1951  The Ntemi: the Traditional Rites in Connection with the Burial, Election, Enthronement and Magic Powers of a Sukuma Chief (London: Macmillan).

Dornan, S. S.

1928  “Rainmaking in South Africa”, Bantu Studies 3: 185-195.

Driberg, J. H.

1919  “Rain-making among the Lango”, Journal of the Royal Anthropological Institute 49: 52-73.

Durkheim, É. & Mauss, M.

1963  Primitive Classification (London: Cohen and West Ltd.).

Evans-Pritchard, E. E.

1929  “Some Collective Expressions of Obscenity in Africa”, Journal of the Royal Anthropological Institute 59: 311-331.

1938  “A Note on the Rain-makers among the Moro”, Man 38: 53-56.

Feddema, J. P.

1966  “Tswana Ritual Concerning Rain”, African Studies 25: 181-195.

Feierman, S.

1990  Peasant Intellectuals: Anthropology and History in Tanzania (Madison: University of Wisconsin Press).

Ginindza, T.

1997  “Labotsibeni/Gwamile Mduli: the Power behind the Swazi Throne 1875-1925”, in F. E. S. Kaplan, ed., Queens, Queen Mothers, Priestesses, and Power: Case Studies in African Gender (New York: The New York Academy of Sciences): 135-158.

Håkansson, N. T.

1998  “Rulers and Rainmakers in Precolonial South Pare, Tanzania: Exchange and Ritual Experts in Political Centralization”, Ethnology 37 (3): 263-283.

Hartnoll, A. V.

1932  “The Gogo Mtemi”, South African Journal of Science 29: 737-741.

1942  “Praying for Rain in Ugogo”, Tanganyika Notes and Records 13: 59-60.

Hauenstein, A.

1967a  “Rites et coutumes liés au culte de la pluie parmi différentes tribus du Sud-Ouest de l’Angola”, Boletim do Instituto de Angola 27: 5-23.

1967b  “Rites et coutumes liés au culte de la pluie parmi différentes tribus du Sud-Ouest de l’Angola”, Boletim do Instituto de Angola 29: 5-27.

Heald, S.

1995  “The Power of Sex: some Reflections on the Caldwells’ ‘African Sexuality’ Thesis”, Africa 65 (4): 489-505.

Herbert, E. W.

1993  Iron, Gender, and Power: Rituals of Transformation in African Societies (Bloomington-Indianapolis: Indiana University Press).

Hoernlé, A. W.

1922  “A Hottentot Rain Ceremony”, Bantu Studies 1: 20-21.

Holas, B.

1949  “Pour faire tomber la pluie (Nord du Togo)”, Notes africaines,  41: 13-14.

Holy, L. & Stuchlik, M., eds

1981  The Structure of Folk Models (London: Academic Press).

Hunter, G.

1953  “Hidden Drums in Singida District”, Tanganyika Notes and Records 34: 28-32.

Jacobson-Widding, A.

1985  Private Spirits and the Ego: a Psychological Ethnography of Ancestor Cult and Spirit Possession among the Manyika of Zimbabwe (Uppsala: African Studies Programme, University of Uppsala) (“Working Papers in African Studies” 24).

1990  “The Fertility of Incest”, in A. Jacobson-Widding & W. van Beek, eds, op. cit.: 47-73.

Jacobson-Widding, A. & Beek, W. van, eds

1990  The Creative Communion: African Folk Models of Fertility and the Regeneration of Life (Uppsala: Acta Universitatis Upsaliensis).

James, W.

1972  “The Politics of Rain Control among the Uduk”, in I. Cunnison & W. James, eds, Essays in Sudan Ethnography: Presented to Sir Edward Evans-Pritchard (London: Hurst & Co.): 31-57.

Jellicoe, M., Puja, P. & Sombi, J.

1967  “Praising the Sun”, Transition 31 (6): 27-31.

Jellicoe, M., Sima, V. & Sombi, J.

1968  “The Shrine in the Desert”, Transition 34 (7): 43-49.

Johnson, V. E.

1948  Pioneering for Christ in East Africa (Rock Island: Augustana Book Concern).

Kaare, B. T.

1999  “Saisee Tororeita: an Analysis of Complementarity in Akie Gender Ideology”, in H. L. Moore, T. Sanders & B. Kaare, eds, Those who Play with Fire: Gender, Fertility and Transformation in East and Southern Africa (London: The Athlone Press): 133-152.

Kaspin, D.

1996  “A Chewa Cosmology of the Body”, American Ethnologist 23 (3): 561-578.

Kenyatta, J.

1959  Facing Mount Kenya: the Tribal Life of the Gikuyu (London: Secker & Warburg).

Kohl-Larsen, L.

1943  Auf den spuren des vormenschen (Deutsche Afrika-Expedition 1934-1936 und 1937-1939) (Stuttgart: Strecher & Schröder).

Koritschoner, H.

1937  “Some East African Native Songs”, Tanganyika Notes and Records 4: 51-64.

Krige, E. J.

1968  “Girls’ Puberty Songs and their Relation to Fertility, Health, Morality and Religion among the Zulu”, Africa 38 (2): 173-198.

Krige, E. J. & Krige, J. D.

1943  The Realm of a Rain-queen: a Study of the Pattern of Lovedu Society (London: Oxford University Press).

Larson, T. J.

1966  “The Significance of Rainmaking for the Mbukushu”, African Studies 25 (1): 23-36.

Leakey, L. S. B.

1977  The Southern Kikuyu before 1903, Vol. 1 (London: Academic Press).

Lindenbaum, S.

1991  “Anthropology Rediscovers Sex”, Social Science and Medicine 33 (8): 865-866.

Lindström, J.

1987  Iramba Pleases Us: Agro-pastoralism among the Plateau Iramba of Central Tanzania (University of Göteborg, Ph.D. Thesis).

Ludger, K.

1954  “Rainmakers in Teso”, The Uganda Journal 18 (2): 185-186.

MacDonald, J.

1890  “Manners, Customs, Superstitions, and Religions of South African Tribes”, Journal of the Anthropological Institute 20 (2): 113-140.

Marshall, L.

1957  “N!ow”, Africa 27 (3): 232-240.

Mauss, M.

1973  “Techniques of the Body”, Economy and Society 2 (1): 70-88.

McKnight, J.  D.

1967  “Extra-descent Group Ancestor Cults in African Societies”, Africa 37 (1): 1-21.

Middleton, J.

1971  “Prophets and Rainmakers: the Agents of Social Change among the Lugbara”, in T. O. Beidelman, ed., The Translation of Culture. Essays to E. E. Evans-Pritchard (London: Tavistock Publications): 179-201.

1978  “The Rainmaker among the Lugbara of Uganda”, in Systèmes de signes: textes réunis en hommage à Germaine Dieterlen (Paris: Hermann): 377-388.

Moore, H. L.

1994  A Passion for Difference: Essays in Anthropology and Gender (Cambridge: Polity Press).

1999  “Gender Symbolism and Praxis: Theoretical Approaches”, in H. L. Moore, T. Sanders & B. Kaare, eds, op. cit.: 3-37.

Moore, H. L., Sanders, T. & Kaare, B., eds

1999  Those who Play with Fire: Gender, Fertility and Transformation in East and Southern Africa (London: The Athlone Press).

Murray, C.

1980  “Sotho Fertility Symbolism”, African Studies 39: 65-76.

Ntudu, Y.

1939  “The Position of Rainmaker among the Wanyiramba”, Tanganyika Notes and Records 7: 84-87.

O’Brien, D.

1983  “Chiefs of Rain–Chiefs of Ruling: a Reinterpretation of Pre-colonial Tonga (Zambia) Social and Political Structure”, Africa 53 (4): 23-42.

Obst, E.

1912  “Die Landschaften Issansu und Iramba (Deutsch-Ostafrika)”, Mitteilungen der Geographischen Gesellschaft in Hamburg 26: 108-132.

1923  “Das abflußlose Rumpfschollenland im nordöstlichen Deutsch-Ostafrika (Teil II)”, Mitteilungen der Geographischen Gesellschaft in Hamburg 35: 1-330.

Ortner, S. B.

1996  “Gender Hegemonies”, in S. B. Ortner, Making Gender: the Politics and Erotics of Culture (Boston: Beacon Press): 35-80.

Packard, R. M.

1981  Chiefship and Cosmology: an Historical Study of Political Competition (Bloomington-Indianapolis: Indiana University Press).

Pender-Cudlip, P.

c.1974 God and the Sun: Some Notes on Iramba Religious History (Unpublished Manuscript at British Institute in Eastern Africa, Nairobi).

Petermann, W.

1985  Regenkulte und regenmacher bei bantu-sprachigen Ethnien Ost- und Südafrikas (Berlin: EXpress Edition GmbH).

Rogers, F. H.

1927  “Notes on some Madi Rain-stones”, Man 27: 81-87.

Sanders, T.

1997  Rainmaking, Gender and Power in Ihanzu, Tanzania, 1885-1995. Ph.D. Thesis, London, London School of Economics and Political Science.

1998  “Making Children, Making Chiefs: Gender, Power and Ritual Legitimacy”, Africa 68 (2): 238-262.

1999  “Doing Gender’ in Africa: Embodying Categories and the Categorically Disembodied”, in H. L. Moore, T. Sanders & B. Kaare, eds, op. cit.: 41-82.

2000  “Rains Gone Bad, Women Gone Mad: Rethinking Gender Rituals of Rebellion and Patriarchy”, Journal of the Royal Anthropological Institute (n.s.) 6 (3): 469-486.

Schapera, I.

1971  Rainmaking Rites of Tswana Tribes (Cambridge: African Studies Centre).

Schoeman, P. J.

1935  “The Swazi Rain Ceremony”, Bantu Studies 9: 169-175.

Snyder, K.

1999  “Gender Ideology, and the Domestic and Public Domains among the Iraqw”, in H. L. Moore, T. Sanders & B. Kaare, eds, op. cit.: 225-253.

Tanner, R. E. S.

1956  “An Introduction to the Northern Basukuma’s Idea of the Supreme Being”, Anthropological Quarterly 29 (4): 45-56.

Ten Raa, E.

1969  “The Moon as a Symbol of Life and Fertility in Sandawe Thought”, Africa 39 (1): 24-53.

Turner, V.

1955  “The Spatial Separation of Adjacent Genealogical Generations in Ndembu Village Structure”, Africa 25 (2): 121-137.

1967  “Color Classification in Ndembu Ritual: a Problem in Primitive Classification”, in The Forest of Symbols: Aspects of Ndembu Ritual (Ithaca: Cornell University Press): 59-92.

Tuzin, D.

1991  “Sex, Culture and the Anthropologist”, Social Science and Medicine 33 (8): 867-874.

Vance, C. S.

1991  “Anthropology Rediscovers Sexuality: a Theoretical Comment”, Social Science and Medicine 33 (8): 875-884.

Vijfhuizen, C.

1997  “Rain-making, Political Conflicts and Gender Images: a Case from Mutema Chieftaincy in Zimbabwe”, Zambezia 24 (1): 31-49.

Weatherby, J.

1979  “Raindrums of the Sor”, in J. B. Webster, ed., Chronology, Migration and Drought in Interlacustrine Africa (London: Longman & Dalhousie University Press): 317-331.

Wollacott, R. C.

1963  “Dziwaguru - God of rain”, NADA 40: 116-121.

Wright, A. C. A.

1946  “A Rainmaking Ceremony in Teso”, The Uganda Journal 10 (1): 25-28.

Zimon, H.

1974  Regenriten auf der Insel Bukerewe (Tanzania) (Freiburg, Sw.: Universitätsverlag Freiburg).

Haut de page


1 This article is based on fieldwork carried out in Ihanzu, Tanzania, between Aug. 1993 to May 1995 and June to Sept. 1999. I thank the UK Economic and Social Research Council, the US National Institute of Health, the University of London, the Royal Anthropological Institute and the London School of Economics for funding different portions of this research; and the Tanzania Commission for Science and Technology (COSTECH) for granting me research clearances. Comments by Albert Schrauwers and the anonymous reviewers helped to sharpen the argument. As is always the case, I alone am responsible for any shortcomings that remain. Finally the men and women of Ihanzu deserve untold appreciation for their unfailing hospitality and their indefatigable interest in my ongoing project. I can only hope I have “told it right”, as they always insisted I must.
2 Writings on African rainmaking are too numerous to detail. See bibliographies in Petermann (1985), Sanders (1997) and Zimon (1974). For some early accounts of various African rain rites, most of which offer interesting but decontextualised accounts of various rain rites, see Cooke & Beaton (1939), Dornan (1928), Driberg (1919), Feddema (1966), Hartnoll (1932), Ludger (1954), Ntudu (1939) and Wright (1946).
3 There are some noteworthy exceptions. See Feierman (1990), Jacobson-Widding (1985, 1990) Kaare (1999), Kaspin (1996), Middleton (1971, 1978) and Packard (1981).
4 Elsewhere, when examining a Tswana rain song, Schapera (1971: 100) says: “I cannot explain the allusions in the second song to ‘whores’…”
5 Other scholars relegate into footnotes their informants’ remarks concerning gender. Hans Cory (1951: 51n) tells us that: “The kibiga is considered a house in which the shigemero represents the male and the four pots represent the female element. Rain is generated by these two elements to the accompaniment of lightning and thunder.” To this he adds no further explanation.
6 The term “complementarity”, I am well aware, does not logically demand equality. Different social classes or people in a caste system, for example, might well be considered “complementary” in that they are interdependent but they are also highly unequal. How complementary elements fit together is ultimately an empirical not theoretical question. I use the term as it relates to Ihanzu understandings of transformative processes.
7 This is not to argue, however, that “transformative processes invoke the human model as the measure of all things” (Herbert 1993: 5); that the human body-with its universal physiological attributes and processes-necessarily provides a shared “primordial psychobiological experience” (Turner 1967: 90) that subsequently allows for higher order modes of social classification. Nor is it to suggest that society invariably provides a template for making the body (Durkheim & Mauss 1963; Mauss 1973). More likely, it would seem, is that both positions are correct: that “the opposition between masculinity and femininity [. . .] constitutes the fundamental principle of division of the social and the symbolic world” (Bourdieu 1990: 78) and that neither bodies nor society can be given logical priority.
8 It would be inaccurate to see any one of these gender models as exclusively held by either men or women, for both genders subscribe to all three perspectives at different times, and under different circumstances.
9 Practices sometimes belie these ideals. For example, sheep and goats are often herded by young girls and women. And a woman has disposal rights over the milk of her husband’s livestock. Further, some women, though not many, inherit and own livestock themselves.
10 This is not just an ideal either. On several occasions I spoke with men who ruefully related to me instances where they were personally denied access to their grainstores by their wives. And without exception, rather surprisingly perhaps, these men accepted (if sometimes reluctantly) their wives’ decisions. None of this is to suggest, of course, that women’s control over grain translates directly into unlimited options or opportunities for them. Women are still bound by cultural constraints of appropriateness. A woman could not, for example, choose to sell all her grain and take a bus trip around Tanzania without suffering ostracism, or possibly worse.
11 As elsewhere in Africa, it is only certain categories of persons whose souls may become spirits. On death, in theory, everyone’s soul (nkolo) becomes a spirit (mũlũngũ). Yet in practice it is generally the spirits of the elderly-both male and female-that afflict people and are thus remembered, regardless of what they have or have not done during their living years.
12 For a description of a very similar ancestral offering to heal cattle (ipolyo la ndwala), see Sanders (1999).
13 Traditionally sacrificial animals were suffocated to death (Obst 1912: 116). Slitting their throats is a relatively recent ritual innovation which probably began either in the late 1930s under Chief Sagilũ, the first Muslim Ihanzu chief, or under Chief Gunda in the 1940s who was also a Muslim.
14 The original address was as follows: “ũewe Mũnyankalĩ nũpũmile kũ mũtala wako nũkũlũ n’ũinzũ kũ mũtala wako nũnino, wakĩla mũnũ m’Ihanzu wahanga kũkete ipolyo, ipolyo la m’ikulungu. Kũkete ipolyo la mbula. Kũipoelia mazĩ n’ĩnkolo naitugilwe ũtikũ. ũko n’ũninzũ ũtwale ninza aya ni mabĩ ũmagũmĩle mũ lũzĩ mũ Nyanza”.
15 There are a number of caves in Ihanzu and surrounding areas where giant drums are found (Hunter 1953). The Ihanzu do not know who made them, but claim that they were already there when they migrated into the area long ago (Kohl-Larsen 1943: 168). It is the sacred caves, not the drums in them, that people find most significant in these offerings.
16 Associations between the sun and masculinity are common among the Ihanzu’s neighbours. This is so amongst the Turu (Jellicoe et al. 1967: 28), Sandawe (Ten Raa 1969: 28), Sukuma (Tanner 1956: 51-52) and Iramba (Pender-Cudlip c.1974: 14). The Iraqw, on the other hand, who live to the east of Ihanzu, associate the sun (Looaa) with femininity and the earth spirits (neetlaamee) with masculinity (Snyder 1999: 227-228).
17 I should like to thank an anonymous reviewer for drawing my attention to this latter point.
Haut de page

Pour citer cet article

Référence papier

Todd Sanders, « Reflections on Two Sticks »,  Cahiers d’études africaines, 166 | 2002, 285-314.

Référence électronique

Todd Sanders, « Reflections on Two Sticks »,  Cahiers d’études africaines [En ligne], 166 | 2002, mis en ligne le 10 juin 2005, consulté le 31 octobre 2020. URL :  ; DOI :

Haut de page

Droits d'auteur

© Cahiers d’Études africaines

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search