From Boy Wives and Female Husbands: Studies in African Homosexualities

by Stephen O. Murray and Will Roskam

© 1998 by Stephen O. Murray,, and Will Roskam. This document may not be reproduced in any format without express permission of the authors. The text is an early manuscript portion of the book, which is due out from St. Martin's Press, NYC, in October 1998.

Homosexuality in “Traditional” Sub-Saharan Africa and Contemporary South Africa*

Super-Saharan Africa provides a well-documented site of age-defined homosexuality. Despite the recent excesses of a Shi’a “Islamic Revolution” in one country with pre-Islamic proscription of homosexuality, the social structure of North Africa almost guarantees male and female homosexualities. Segregation of the sexes is nearly total, and, historically, women have been monopolized in harems, where, rarely visited, albeit closely watched by eunuchs, they almost certainly turned to each other. Being neither literate nor of much interest to those who were, there are few traces of what harem life was like for women.

Although I do not think any serious scholar has attempted to deny pederastic practices in North African cultures, recurrent attempts have been made in English to deny any indigenous homosexuality in sub-Saharan Africa. For instance, Hrdy (1987:1113) categorically asserted, “Homosexuality is not part of traditional societies in Africa” and, after asking some chiefs and headmen about it, Gelfand (1979) wrote that “the traditional Shona [of Zimbabwe] have none of the problems associated with homosexuality [so] obviously they must have a valuable method of bringing up children, especially with regards to normal sex relations, thus avoiding this anomaly so frequent in Western society.”

Dynes (1983) traced the myth of exclusive heterosexuality in indigenous Africa to the 94th chapter of Edward Gibbon’s Decline and Fall of the Roman Empire (1781). Gibbon wrote, “I believe and hope that the negroes in their own country were exempt from this moral pestilence.” Obviously, Gibbon’s hope was not based on even casual travel or inquiry. Sir Richard Burton, who a century later “inadvertently helped to reinforce the myth of African sexual exceptionalism by drawing the boundaries of his ‘sodatic Zone’ where homosexuality was widely practiced and accepted to exclude sub-Saharan Africa” (Dynes 1983: 20), was personally familiar with male homosexuality in Islamic societies within his zone, but had not researched the topic in central or southern Africa — where there were “primitive” hunter/gatherer societies and quite complex state formations before European conquest.

Especially where Western influences (notably Christian and Marxist) have been pervasive, there is now a belief that homosexuality is a decadent, bourgeois Western innovation forced upon colonial Africa by white men, or, alternately by Islamic slave-traders. Around the world, people view homosexuality as a vice of some other people. Thus, the recurrent British claim Norman conquerors introduced homosexuality to the British Isles. Various French accounts view homosexuality as Italian, Bulgarian, or North African. Italians accept only the latter two homelands. Bulgarians attribute greater popularity and/or the origins of homosexuality to Albanians, and Albanians in turn to Turks. Similarly eastern Bantu claimed that pederasty was imported by the Nubians (Schneider 1885:295-6), Sudanese blame Turkish marauders (Weine 1848:120), etc. Such views tell us something about perceived ethnic boundaries, but nothing about the origins or the historical transmission of cultural traits. The belief by many Africans that homosexuality is exogenous to the history of their people is a belief with genuine social consequences — in particular stigmatization for those of their people engaged in homosexual behavior or grappling with gay identities. These beliefs are not, however, based on serious inquiry, historical or otherwise.

Without indigenous writing systems prior to the end of the 19th century, there are no analyses of the social structures of African societies prior to alien contact. What is inscribed of “traditional” African cultures was written by some of the Northerners who disrupted not the timeless stasis of this or that “traditional” African culture, but, rather, who disrupted little-known-because-unwritten histories of African cultures (see Wolf 1982, J. Burton 1988). In many cases the observers inscribing “traditional African culture” did not understand that their presence as observers was in-itself a product of history and domination, as those now increasingly denied a privileged role of observer are keenly aware. As Talal Asad (1973:17) put it, “The colonial power structure made the object of anthropological study accessible and safe — because of it sustained physical proximity between the observing Europeans and the living non-Europeans became a practical possibility." Granting that the observing Europeans were not so invisible and without effects on what they observed as they thought or pretended, they are the only source of data on homosexuality in Africa before the most recent few decades. Although memories and oral history may still be salvaged in some places, most of what can be read about African societies was inscribed in the last decade of the 19th century or later.

Within this literature there are reports of native conceptions and native practices of male homosexuality in many societies. The itinerary through a very uneven corpus of reports will proceed from Ethiopia through the Sudan to the Kenyan coast, then fly to the West Coast. Finally leaving Islam behind, this account will move west to east through central Africa, and then resume the southward course. With fewer stops the journey is repeated in the literature on female homosexuality. Again, there are extended stops in the Sudan, Mombasa, and South Africa. A discussion of woman-woman marriage on “the Slave Coast” (between the Benin and Volta rivers) forms a later chapter in this volume. With so many men working elsewhere, it would be unusual for central African women not to turn to each other, as south African women in the “homelands” do, but no one seems to have looked at or for such a parallel. Very little has been published about female homosexuality in the labor reserves of central Africa. Indeed, very little has been published about female or male homosexuality in post-colonial Africa. Absence of evidence, particularly an absence proclaimed in official ideology, should not be taken as evidence of absence, but rather as absence of research in contemporary, urban Africa.

The literature that can be reviewed is out-of-date, but still of interest in showing that various homosexualities existed at the time of European colonization.

Table One relates the tribal names mentioned below to language families. The language families are mapped in Figure One.

Ethiopia, Eritrea, and the Sudan

In Ethiopia Bieber (1909) “encountered Uranism” among the Semitic Harari people and noted that “sodomy is not foreign to the Harari. Albeit not as commonly, it also occurs among the Galla and Somal.” (nearby peoples speaking Cushitic languages). He also noted mutual masturbation by both sexes and all ages for all three peoples, and specified that among the Harari, Uranism was practiced as often between men as between men and boys.

More recently, also in Ethiopia, homosexual relationships among Qemant (Kemant) shepherd boys were reported by Gamst (1969: 106). Among Amhara peasants, Messing (1957:550) found (better-accepted) male transvestites, who they view as “god’s mistakes.” Wändarwäräd (“literally male-female”) with visible male sexual characteristics, but whose structure is popularly believed to be defective,

live as individuals, not forming a society of their own, for they are tolerated. Only their kinfolk are ashamed of them, so they to to live in another province. Women tolerate a transvestite “like a brother”; men are not jealous of hium even when he spends all his time with the womenfolk. Often the transvestite is an unusually sensitive person, quick to anger, but intense in his personal likings, sensitive to cultural diffusions from the outside world, especially those carried by Arab traders; and Muslim Arab traders are often the only male contacts he tolerates (p. 551).

In his fieldwork during the mid-1960s in the walled towns of the Konso in the highlands at the southern edge of Ethiopia, C. R. Hallpike found a complex of beliefs about the dangerousness to men of being weakened by contact with vaginas along with restrictions of when and how often marital intercourse are permissable that matches the most extreme Melanesian prototypes of sexual antagonism [cf. Herdt 1984). Konso men believe that "women have an emotionally as well as physically deleterious influence on men" (1972:153) and one told Hallpike "Some girls’ vaginas are so strong that they can snap off a man’s penis" (p. 152).

The Konso have "two words each for penis, vagina, and sexual intercourse, but no less than four for ‘effeminate man’" (p. 150). Sagoda, the one Hallpike elaborated upon, includes men who never marry, weak men, and men who wear skirts.

Men who actually wear skirts are very few, and thse who do are clearly incapable of acting as men. I knew one in Gaho, who earned his liviing curing skins, a female occupation. He was very effeminate in voice and manner. . . . I was told that sagoda liked to play the passive role in sodomy, and the description I was given of the manner in which a sagoda would induce a man to perform this upon him in the night was so detailed that it could not have been invented. The question is whether normal men only practice sodomy with sagoda or among themselves. I am strongly inclined to think it is not confined to relations with sagoda. (p. 151)

Konso men “were generally very reluctant to talk about sexual matters with” Hallpike, but he heard “coarse remarks on occasion.” These included jokes about taking a man reputed to be something of a sagoda out into the fields to rape him. Hallpike concluded, “This sort of occasion, the conduct of transvestites, and the sexual strains put on men by society, lead one to suppose that they seek relief among themselves on occasion. But this is not to say it is approved of” (1972:151).

Among the Maale of southern Ethiopia, “a small minority [of men] crossed over to feminine roles. Called ashtime, these (biological) males dressed like women, performed female tasks, cared for their own houses, and apparently had sexual relations with men,” according to Donald Donham (1990:92), who wrote, “In 1975 I was aware of only one ashtime in Maaleland, although informants asserted that more had existed in the 19th century. Indeed, part of the Maale kin’s traditional installation had consisted of a ritual ordination of an ashtime.” Nonetheless, by 1975, the Maale considered ashtime “abnormal.” The one whom Donham knew gave a clear statement of third gender conception: “The Divinity created me wobo, crooked. If I had been a man, I could have taken a wife and begotten children. If I had been a woman, I could have married and borne children. But I am wobo; I can do neither.”

Donham suggested that there was a continuous gradation of maleness from the ritual kings to subchiefs on down, rather than absolutely discrete gender categories. The ritual king “was the male principle incarnate” (p. 112). No woman of childbearing age could enter the king’s compound, so domestic labor generally done by women was done by ashtime who traditionally were gathered and protected by the kings. Also, on the night before any royal ritual the king could not have sexual relations with women, whereas, “lying with an ashtime was not interdicted.” Thus, ashtime constituted “part of the generativity of maleness in Maale” (p. 113).

Paolo Ambrogetti (1900:16) wrote of relations in Eritrea with diavoletti (“little devils,” not an indigenous label) that were openly pursued . Regarded as being no more than a mild fault, involvement in such relations were tolerated as a source of income by the boys’ fathers. After puberty, diavoletti generally undertook intercourse with women, but those especially attached to their patrons continued as late as age twenty. An unusual case was a chief aged 25 and married, who continued to pursue receptive intercourse with men “senza lucro” (not for payment), while many apparently effeminate male adults were not “seuxal inverts” (Ibid.). Ambrogetti was one of a number of late-19th-century and early-20th-century writers who tried to argue away same-sex sex among “nature peoples” as situational (in this instance, lucrative) with few (when any) “ inclined” that way (i.e., to long-term commitment to homosexual eros or to cogenital inversion).

Reports from indigenous African societies of age-stratified homosexuality are sparse. Clearly, gender-defined homosexuality existed from Nubia to Zululand on the East Coast of Africa (and offshore on Zanzibar and Madagascar, as well), often related to possession cults in which women have prominent roles and male participants tend to transvestitic homosexuality.

Cross-gender homosexuality not tied to possession cults has been reported in a number of East African societies. Needham (1973:109-27) described a religious leadership role called mugawe among the Meru of Kenya which includes wearing women’s clothes and hairstyle. Mugawe are frequently homosexual, and sometimes are married to a man. Bryk (1939[1928]:151, 1964:228) reported active (i.e., insertive) Kikuyu pederasts called onek. and also mentioned “homoerotic bachelors” among the pastoralist Nandi and Maragoli.

Among the Bagishu Bryk (1939[1928]:151, 1964:228) claimed that hermaphrodites

are quite numerous and are called inzili; among the Maragoli kiziri. A seventeen year old boy told me, without being at all embarrassed, the he had such a mzili in the posterior. The passive fellow called him and gave him ten shillings for this. While he was with him, the pederast had his flabby penis tied to his stomach.

The boy told him that he had rejected many other offers of mzili posteriors. Among the Mandari of the Sudan, Buxton (1963: 103) reported,

Homosexuality is viewed as a ludicrous and non-productive act. Thus while all perverts are not necessarily thought witches, since the latter know the aberrant or harmful nature of those habits, they exploit them designedly, in accordance with their wish to obstruct normal development (also see Buxton 1973:209).

Folk fear of witches is widespread in Islamic cultures, although a link between witchcraft and pederasty is unusual in existing ethnographic reports of Islamic cultures. Nadel (1955) did not mention any such link in contrasting two other Sudanese peoples: the Heiban in which there is “no expected corollary of homosexual acts” (i.e., no homosexual role, and particularly no connection to occupational specialization as healers), and the Otoro who recognize(d) a special transvestitic role in which men dress and live as women, but also are not especially likely to be healers (p. 677). Nadel (1947: 242) also mentioned transvestitic homosexuality among the Moro, Nyima and Tira, and reported marriages of Korongo londo and Mesakin tubele for the brideprice of one goat (1947: 285). The generally young husband in these marriages might also obtain women wives. Nadel did not specify whether londo and tubele were older than their husbands although marking the latter as “young” implies this. In these tribes with “widespread homosexuality and transvestiticism” Nadel (1947: 300) reported a fear of heterosexual intercourse as sapping virility and a common reluctance to abandon the pleasures of all-male camp life for the fetters of permanent settlement: “I have even met men of forty and fifty who spent most of their nights with the young folk in the cattle camps instead of at home in the village.” It should be noted that in these pervasively homosocial societies, the men who were wives were left at home with the women, i.e., were not in the all-male camps.

In traditional, monarchical Zande culture, before the pacification which made it safe for ethnographers like Nadel and Evans-Pritchard to work in the Sudan and central Africa,

men used to have sexual relations with boys as they did with wives. A man paid compensation to another if he had relations with his boy. People asked for the hand of a boy with a spear, just as they asked for the hand of a maiden of her parents. All those young warriors who were at court, all had boys (Kuagbiaru, quoted by Evans-Pritchard 1970:1430)

homosexuality is indigenous. Azande do not regard it as at all improper, indeed as very sensible for a man to sleep with boys when women are not available or are taboo. . . In the past this was a regular practice at court. Some princes may even have preferred boys to women, when both were available. This is not a question I can enter into further here beyond saying I was told that some princes sleep with boys before consulting poison oracles, women being then taboo, and also that they sometimes do so on other occasions, just because they like them (Evans-Pritchard 1971: 183).

Kuagbiaru told Evans-Pritchard that when a boy (between the ages of twelve and twenty) appealed to a prince, the prince would summon the boy as a page. Later, the prince “provided bridewealth for his pages when they grew up,” although “when a prince dies they do not let his pages escape; they kill them after the prince is dead, for they have eaten the prince’s oil. People call them ‘the prince’s old barkcloth’, for, because he used to summon them all the time, they are like his old barkcloth” (p. 185). In addition to pages to princes,

Many of the young warriors married boys, and a commander might have more than one boy-wife. When a warrior married a boy he paid spears [brideprice], though only a few, to the boy’s parents, as he would have done had he married their daughter. The warrior in other ways acted towards the parents as though he had married their daughter. . . He addressed the parents as gbiore and negbiore , ‘my father-in-law’ and ‘my mother-in-law.’ He gave the boy himself pretty ornaments; and he and the boy addressed one another as badiare, ‘my love’ and ‘my lover’. The boy fetched water for his husband, collected firewood and kindled his fire, bore his shield when traveling.... The two slept together at nights, the husband satisfying his desires between the boy’s thighs. When the boy grew up he joined the company and took a boy-wife in his turn. It was the duty of the husband to give his boy-wife a spear and a shield when he became a warrior. He then took a new boy-wife. Thus, Kuagbiaru, a member and later a commander of one of Prince Gangura’s companies, married three boys in succession. . . (Evans-Pritchard 1971:199-200).

Another commander, Ganga, told Evans-Pritchard (1970:1431) that “there were some men who although they had female wives, still married boys. When a war broke out, they took their boys with them,” although they were left in camp, as befitted their wifely status, not their future status as fellow warriors."The warrior paid bridewealth (some five spears or more) to the parents of the boys and performed services for them as he would have done had he married their daughter.... Also, if another man had relations with his boy, he could, I was told, sue him at court for adultery” (p. 1429).

Evans-Pritchard maintained that “it was on account of the difficulties of getting satisfaction in heterosexual relationships that boy marriage was recognized,” because with easier (and earlier) marriage between men and women, “boy-marriage has in post-European times entirely disappeared” along with the cultural contexts. That is, the military companies and the royal court, also disappeared. Even after the custom and much else of the culture had been disrupted, “I have never heard anyone speak of sleeping with a boy with distaste” (p. 1429).

R. P. Graer (1929:362) did not consider any homosexual preferences, instead attributing unnatural vices among young men to the monopoly of women by the rich and powerful. Similarly, French colonial administrator Adolphe Cureau (1904:644-5) attributed warrior pederasty among the Sandeh to the monopolization of women in the vast harems (bodimoh) of Sandeh royalty. Vassals, soldiers, and servants had to make do with what the rulers left. Boy nsanga (servants d’armes) took the place of women. Wearing their hair artfullyparted, with arms and necks loaded with decorations, a woolen skirt around the hips, and their bodies oiled and glistening, the boys were at the disposals of soldiers. These ndongo-techi-la followed the soldiers on their marches, carrying their husbands’ rifle, hammock, and a little bag with pipe, firestick, and some millet. In the camps they cooked and managed household finance.

Among the Mossi in what is now Burkina Faso, soronés (pages), chosen from among the most beautiful boys aged seven to fifteen, were dressed and had the other attributes (including le rôle) of women in relation to chiefs, for whom sexual intercourse with women was denied on Fridays. I do not read Tauxier’s (1912: 569-70) report as indicating sexual unavailability the rest of the week. He did stress that only the chiefs were forbidden women on Fridays and that the soronés who proved their discretion were entrusted with state secrets.

While serving as a soroné, there were annual tests to make sure that the boy had not been sexually intimate with any woman. After the boy reaches maturity he was given a wife by the chief. The first child born to such couples belonged to the chief. A boy would be taken into service as his father had as a soroné, a girl would be given in marriage by the chief (as her mother had).

Among the Bantu-speaking Pahouin slash-and-burn farmers (Bene, Bulu, Fang, Jaunde, Mokuk, Mwele, Ntum, Pangwe), who live in thee rainforest north of the Congo River in present-day Gabon, homosexual intercourse was bian nkû’ma, a medicine for wealth, which could be transmitted from bottom to top in anal intercourse, according to German ethnographer Günther Tessman (1904:23), who also mentioned that

it is frequently “heard of” that young people carry on homosexual relations with each other and even of older people who take boys, who as is well known “have neither understanding nor shame” and readily console them by saying biabo pfia’nga (we are having fun, playing a game)”.... Adults are excused with the corresponding assertion: a bele nnem e bango = he has the heart (that is, the aspirations) of boys” Such men were said to have a heart for boys: bian nku’ma.(1904:131; HRAF translation).

Even more remarkable than Pangwe belief in the medical benefits of anal intercourse is Hulstaert’s (1938a:86-7) report that among the Nkundu (of the then-Belgian Congo) the younger partner penetrated the older one, a pattern quite contrary to the usual pattern of age-graded homosexuality. His conventional Christian rhetoric of unnatural vice (and benighted savages) makes it unlikely Hultsaert had any personal sympathy for homosexuality, e.g., he wrote of “the game of gembankango in which boys, imitating monkeys chase each other through the trees and creepers can — and does — result in reprehensible scenes” (p. 73).

Based on her fieldwork in Khartoum from 1970-2, Fluehr-Lobban (1977:134) stated that “if a man dares to gain some sexual experience before marriage, he must do it in the prostitutes’ quarter or through temporary homosexual liaisons that are tolerated before marriage.” She did not indicate whether involvement in homosexality is temporary for both partners, or what they actually do with each other.

Among the Muslim riverian peoples northern Sudan, Constantinides (1977:63) mentioned male participation in a healing cult called zaar that is mostly the domain of women (in a society with a marked sexual division of labor and sexual segregation): “Some men are regular participants at cult rituals, and a few become cult group leaders. Of this male minority some are overt homosexuals, while others may initially have symptoms, such as bleeding from the anus or penis, which tend symbolically to classify them with women.”

Although “men who attend zaar rituals regularly are suspected by both men and women of being homosexual,” there is also suspicion that some men may “dishonourably [be] gaining access to women by feigning illness”(Constantinides 1977:63). Similar suspicions arise about the ’yan daudu in the bori cult among the Hausa. The bori cult is similarly a refuge from hyper-patriarchal Islam for females and effeminate males, as is discussed below.

Among Swahili-speakers on the Kenya coast, Shepherd (1978a: 133) reported, “In Mombasa, both male and female homosexuality is relatively common among Muslims; involving perhaps one in twenty-five adults.” Shepherd (1987: 240 — with no data nor discussion of the basis for either the earlier estimate or its revision) raised the estimated rate to one in ten. In the first report emphasized, male homosexuality was confined to prostitution:

Mombasa’s mashoga are passive male homosexuals offering their persons for money. They advertise themselves in bright tight male attire in public places, usually, but may, when mingling with women at weddings, don women’s leso cloths, make-up and jasmine posies. Mashoga have all the liberties of men and are also welcome in many contexts otherwise exclusive to women (Shepherd 1978a: 133; emphasis added).

Shepherd (1978b: 644) asserted that “though there are long-lasting relationships between homosexuals in Mombasa, most homosexual acts are fleeting, paid for in cash.” In a more recent analysis, Shepherd (1987:250) explained

The Swahili [word] for a male homosexual is shoga, a word also used between women to mean ‘friend’. Homosexual relations in Mombasa are almost without exception between a younger, poorer partner and an older, richer one, whether their connection is for a brief act of prostitution or a more lengthy relationship. In the former case, there are fixed rates of payment, and in the latter, presents and perhaps full financial support for a while. But financial considerations are always involved and it is generally only the person who is paid who is called shoga. The older partner may have been a shoga himself in his youth, but is very likely to be successfully married to a woman as well as maintaining an interest in boys. Only if he is not married and has an apparently exclusive interest in homosexual contacts will he perhaps still be referred to as a shoga. The paid partner usually takes the passive role during intercourse, but I think it is true to say that his inferiority derives from the fact that he is paid to provide what is asked for, rather than for the [sexual] role he adopts.... The paying partner is usually known as the basha — the Pasha, the local term for the king in packs of playing cards.

The main thesis of Shepherd (1978a, 1987) is that rank is more important than gender in Mombasa, and in Mombasan conception of homosexuality. Indeed, according to Shepherd (1987:255), trading sexual complaisance (specifically, providing a site for penetration) for money (or or help) is only “patron-client relations given a sexual dimension.” The primacy of wealth to sexual behavior is stressed in the passage quoted above. Startlingly, it is followed by a discussion of a folk view recognizing (pre-cognizing) which boys will engage in homosexuality on the basis not just of relative wealth, but also effeminacy (“prettiness”):

People say that they can predict who will be a homosexual, even with boys as young as 5 or 6 years old at times. They seem to base their prediction upon prettiness and family circumstances; boys reared in all-female households by a divorced mother and several sisters are likely to become homosexuals, they say, and the prediction is self-fulfilling since these are the boys whom men are certain to approach . ‘If he’s not a homosexual yet, he will be’, say women of teenage boys from such households (Shepherd 1987:250-51).

To me, this suggests that more than economics is involved, even if Shepherd were right that it is primary in differentiating the basha from the shoga.

Although Shepherd (1978a,b) strenuously objected to Wikan’s (1977) suggestion of a “transsexual” or “third gender” conception, Shepherd (1987: 259-60) acknowledged some gender variance in dress and asserted that shoga “tend to employ the gait and voice which are the international signals of homosexuality. These seem to be imitated from other homosexuals, not from women, and the modest and quietness of ideal Swahili womanhood are quite absent in homosexual behaviour.”

Godfrey Wilson (1957:1) earlier reported that in Lamu, a Swahili town north of Mombasa, boys dressed as women, performed a striptease and then paired off with older men from the audience (Shepherd 1987:269 note 9). The first Swahili-English dictionary (Krapf 1882:891) included hánithi (clearly cognate to what Wikan romanized as xanith) for “catamite,” as well as mumémke (mume=man, mke=woman). Moreover, as Bleys (1995:168) noted, Baumann (1899) attested Swahili labels for sexually receptive men—mkesimune and mzebe, not just the Arabic hanisi.

Shepherd argued convincingly that shoga are not classified as “women,” the “second sex.” For instance, they are not given women’s tasks to do, but are rather used as junior male kin are” (1987:253). There is no evidence in any of Shepherd’s publications, however, which bear on whether shoga departures from the expectations of masculine deportment, dress, and financial independence are considered a third sex or gender.

Also on the Kenyan coast, Duran (1993:186) notes that “the picturesque townships of Malindi has been turned into a German sex colony frequented by single men and homosexuals” in quest of black phalluses and reports an Islamicist backlash against what is viewed as foreign corruption of impecunious young males.

Interpreting Absence of Evidence and Assertions of Absence

On the peripheries of Islam, as in those further south, any confidence that lack of mention of homosexuality in the ethnographic literature for other tribes indicates the absence of the phenomenon should be undermined by Evans-Pritchard’s (1970, 1971) deathbed reports that the Zande warriors about whom he had written extensively decades before paid brideprice for the young boys who served them as temporary wives. The denial of homosexual behavior in a culture with a (gender-defined) role (kitesha), discussed below, or the contrast below between Edel (1957) and La Fontaine (1959) similarly should caution against acceptance of claims that there is no homosexuality in a culture on the basis of lack of mention in any particular ethnography. Even native denial of the possibility of homosexuality must be treated skeptically, as MacDermot’s experiences among the pastoralist Sudanese Nuer showed. Sexual behavior between men did not occur, he was told (MacDermot 1972:99). Then, one day, he observed “a crazy old man... accepted by everyone in the village... [who] either tended the cattle or at other times helped the women harvesting corn or carrying burdens.” Seeing the man heading off to work with the women prompted Doereding, MacDermot’s primary Nuer informant,

to tell me a story which completely contradicted all I had thought and learnt so far about Nuer homosexual relations. It had always been stressed by the tribesmen that homosexuality between men was impossible, for if discovered amongst them it could be punishable by death. Doereding now told me about a crazy man he had once known who lived near Nasir in the Sudan and who frequently dressed as a woman. This was different Doereding explained, because “the man had actually become a woman”; the prophet of Deng had been consulted and had agreed to his change of status. The prophet had decided to call on the spirits and after consultation had declared that indeed the man was a woman. Therefore, he could dress in women’s clothes and behave as a woman. From that time onward it was agreed that ‘he’ should be called ‘she’, and ‘she’ was allowed to marry a husband. All very confusing, and so totally against what the Nuer had been telling me, that I questioned Doereding carefully, but he failed to produce further explanation (MacDermot 1972:119).

That is, in the Nuer conception sex was not occurring between two men. One was socially recognized as a woman, a not uncommon way of regendering those accidentally born with male sexual organs.

An early, much praised ethnography (Driberg 1923:210) uncritically passed on the dubious folk explanation that impotence is the basis for assuming the gender-crossing role among the Lango of Uganda (cf. Herskovits 1937: 117). This is remniscent of the frequent confusion about the biological basis of the “berdache” role in native North America. Under the protection of the folk explanation, the mudoko dako [transformed man] is treated as a woman, and may live as a legitimate wife to a man. According to Driberg, such persons were rare among the Lango (50 out of 17,000 people), but Lango informants told him that such conduct was very common among neighboring groups to the east, specifically the Iteso and Karamojan (Karimojong).

One early-noted exception to “the negro race [being] mostly untainted by sodomy or tribadism” in Burton’s “Terminal Essay” is that “Joan dos Sanctos found in Cacongo of West Africa ‘Chubidu, which are men attyred like women and behave themselves womanly, ashamed to be called men; are also married to men and esteem that vnnaturale damnation an honor’” (Burton 1930[1886]:87-88 citing “Aethiopia Orientalis, Purchas ii 1558”).

Spiritual Vocations in Western Africa

Among the Dagara of southern Burkina Faso, Malidoma Somé explained,

gender has very little to do with anatomy.... The Earth is looked at, from my tribal perspective, as a very, very delicate machine or consciousness, with high vibrational points, which certain people must be guardians of in order for the tribe to keep its continuity with the gods and with the spirits that dwell there—spirits of this world and spirits of the other world. Any person who is this linke between this world and the other world experiences a state of vibrational consciousness which is far higher, and far different, from the one that a normal person would experience. This is what make a gay person gay. This kind of function is not one that society votes for certain people to fulfill. It is one that people are said to decide on prior to being born. You decide that you will be a gatekeeper before you are born. And it is that decision that provides you with the equpiment that you bring into this world. So when you arrive here, you begin to vibrate in a way that Elders can detect as meaning that you are connected with a gateway somewhere.

Somé’s 1994 book Of Water and Spirits relates some of his own experiences during (a belated) initiation after running away from a French missionary school. Later, sent to study in the West (earning doctorates from the Sorbonne and from Brandeis), Somé (1993:7) wondered about Dagara

who feel the way that certain people feel in this culture that has led to them being referred to as ‘gay.’... When I asked one of them, who had taken me to the threshold of the Otherworld [as one of the elders supervising his initiation cohort], whether he feels sexual attraction towards another man, he jumped back and said, “How do you know that?” He said, “This is our business as gatekeepers.”

He noted that “the gay person is very well integrated into the community” and that this particular man had a wife and children. Dagara, including Somé (1993:8), believe that the survival of the cosmos (not just of the earth) depends on such gatekeepers:

Unless there is somebody who constantly monitors the mechanism that opens the door from this world to the Otherworld, what happens is that something can happen to one of the doors and it closes up. When all the doors are closed, this Earth runs out of its own orbit and the solar system collapses into itself. And because this sustem is linked to other systems, they too start to fall into a whirlpool.

He also noted that among the Dogon, ”a tribe that knows astrology like no other tribe that I have encountered, the great astrologers of the Dogon are gay” and generalized that outside Christendom, ”everywhere else in the world gay people are a blessing” (p. 8). While claiming religious justification for homosexual desire (not exclusive of heterosexual functioning), Somé does not claim that any sexual relations with the gatekeepers conveys spiritual benefits.

Reverting from an all-too-rare native representation to typically hostile alien/Christian notices as this survey wends southeastward, we find that Talbot (1926,III:766) alluded to ”unnatural rites performed during the ceremonies and sympathetic magic practiced in order to secure fruitfulness for human being, animals and fishes” in a secret society variously called Obukele, Obukere or Oweu among the Ibo and Abuan peoples of Nigeria, adding that ”the chief celebration takes places when the Niger and the adjoing rivers are full” (p. 767), but provided no details of the rites or the roles within the cult (there or in Talbot 1927).

Hausa Roles

Fremont Besmer (1983) discussed in greater detail a possession “cult” among the (generally Islamic and urban) Hausa that is strikingly similar to New World possession cults among those of West Africa descent, and is ”generally regarded as the displaced religious tradition of the pre-Islamic Hausa.”. As in Haitian voudou(n), the metaphor for those possessed by spirits is horses ridden” by the spirit. Homosexual transvestites in the Hausa bori cult are called ‘Yan Daudu, son of Daudu. Daudu is a praise name for any Galadima (a ranked title), but specifically refers to the bori spirit Dan Galadima (literally, son of Galadima; the Prince),who is] said to be ” a handsome young man, popular with women, a spendthrift, and a gambler” (Besmer 1983:30n4).

Joseph Greenberg (1941:56) noted that ”the group of Hausa spirits known as ‘Yan Dawa, ‘children of the forest’ have their counterpart in the Dahomey aziza, the Bambara (of Mali) kokolo, and the Yoruba divinity Arnoi.” According to Besmer (1983:18) ’Yan Daudu are not possessed by Dan Galadima, and are not possessed by other spirits when he is present. Instead, they make and sell ”luxury snacks” — i.e., more expensive, more prestigious food such as fried chicken,” (Pittin 1983:297). ‘Yan Daudu also operate as intermediaries between (female) prostitutes and prospective clients.

In his ethnography of Hausa possession religion, FremontBesmer (1983:18) wrote: ”Women provide the bulk of membership for the cult and are stereotyped as prostitutes” (also see Smith 1954:64, Hill 1967:233). Pittin reported

The economic enterprises of the ‘Yan Daudu are centered on three related activities: procuring, cooking, and prostitution. Procuring, the mobilisation of women for illicit sexual purposes, clearly demands close ties between the procurer and the women. The dan daudu [sing.], in his combination of male and female roles, can and does mediate between men and women in this context. (1983:296)

Living among women in the strangers’ quarters of Hausa towns provides “a cover for men seeking homosexual services. The dan daudu and his sexual partners can carry out their assignations with greater discretion than would be possible if the yan daudu lived together, or on their own outside the gidan mata,” where visitors would be marked as seeking sex with she-males (Pittin 1983:297).

In patriarchal Hausa society, the bori cult provides a niche for various sorts of low status persons:

Women in general and prostitutes in particular.... Jurally-deprived categories of men, including both deviants (homosexuals) and despised or lowly-ranked categories (butchers, night-soil workers, poor farmers, and musicians) constitute the central group of possessed or participating males” plus “an element of psychologically disturbed individuals which cuts across social distinctions.” (p. 19).

Besmer’s account leaves problematic how Hausa individuals come to be defined by themselves or by others as homosexual or transvestitic. In the terms of labeling theory, joining the bori cult is “secondary deviance.”

One whose status identity is somewhat ambiguous, arising from some personal characteristic, specific social condition, or regularly recurring condition associated with the life cycle can seek either to have his social identity changed or his social status regularized and defined through participation in bori rituals. Marked by ‘abnormality’ and accepted as a candidate for membership in the cult through an identification of iskoki as the cause of the problem, a person’s behavior becomes explainable, and simultaneously earns a degree of acceptability, after the completion of the bori initiation. Symbolic transformation from a suffering outsider — outside both the society and the cult — to one whose status includes the description horse of the gods is achieved. (pp. 122-3)

Although the bori cult provides a niche with an alternative [to the larger society’s] prestige hierarchy, the cult itself remains marginal. As Besmer noted:

Status ambiguity is not completely eliminated through involvement in the bori cult. While an initiated individual achieves a specific, formal status within the cult since possession is institutionalized, it is not possible for him to escape the general social assessment of his behavior as deviant. (p. 21).

Nonetheless, there are at least indications of “tertiary deviance,” which is to say rejection of the stigma of the cult and its adepts (p. 18), although Besmer did not attempt to discover the degree to which adepts accept the disvaluation of their “kinds of people” in the dominant Islamic Hausa culture.

From Bayero University (in Kano, Nigeria), Kleis and Abdullahi (1983; building on Salisu Abdullahi’s interviews of 140 ’yan daudu) presented a functionalist analysis of a wholly secular ’dan daudu role. In their view, prostitutes provide a “safety valve” in a Hausa society in which female seclusion has been increasing since the early 19th century. Rather than being prostitutes (like the Omani khanith described by Wikan 1977), the ’dan daudu procures females for males, recruiting runaway women and “socializing them in the seductive arts” and on a routine basis, “soliciting suitors, arranging contacts, extolling and advertising her charms, and managing relations with the authorities” for these female prostitutes, who live and work literally ouside the walled city core (birni). The ’dan daudu is paid for his services both by customers and prostitutes, and these commissions “constitutes the bulk of his daily income” (Kleis and Abdullahi 1983:45). Kleis and Abdullahi argue that the economic rewards are “sufficient to account for recruitment without assuming a personality predisoposition” (p. 52) to dress and behave like a woman. In explaining why they undertook the ’dan daudu role, 56 percent attributed economic reasons. Only seven percent explained being a ’dan daudu as due to “nature.” Kleis and Abdullahi consider the role as an economic niche for poor emigrants (not all of whom are Hausa) from the countryside, not especially a refuge for males seeking sex with males: They note that “many ’yan daudu are assumed also to be homosexuals, although this does not seem to be the major feature of their social status, which hinges more on their self-identification as females” (p. 44). They suggest that “a male with masculine gender identity and pronounced heterosexual interests would be less suitable as a broker because he might well find it difficult to separate his personal and professional involvements with the prostitute and would risk becoming a rival of her customers” (p. 46). They do not consider that a “homosexual” ’dan daudu might become a rival for female prostitutes’ customers, and do not appear to have inquired on who are the sexual partners of ’yan daudu.

They note that there is no Hausa or Sohari role for women acting like men, and stress that both Omani khanith and the Hausa ’dan daudu are conceived as kinds of males. Indeed, “they occupy these positions precisely because they are anatomical males.” That is, ’dan daudu is a role some males take—and sometimes abandon. Like the khanith, the ’dan daudu “can temporarily, alternately, or permanently switch back and take up conventional male roles—a course definitely unavailable to anatomical females” (p. 49).

Like Kleis and Abdullahi, Gaudio discounts the connection between possession cults and ’dan daudu in Hausa society. He focuses more than they did on (homo)sexual relations involving ’yan daudu and other Hausa men.. Talking to members of what he calls “Kano’s gay male community” about a 1994 Muslim Brotherhood-affiliated Hausa newspaper Gaudio observed

challenges to the newspaper’s characterization of homosexual marriages as a Western practice alien to Hausa Muslim culture. Few of the self-described Hausa homosexuals I met were even aware of Western gay life.... My conversations with Hausa gay men suggest that they perceive homosexuality and homosexual marriage as practices that are indigenous to Hausa Muslim culture [even] as they are marginal within it. (1996a abstract)

Gaudio (1996b:123) argues that “gay” is appropriate in that these Hausa men “are conscious of themselves as men who have sex with men and who consider themselves to be socially distinct from men who do not have this kind of sex.” One term for those who do not is mahaho, “blind man” (p. 127). Those Hausa men whom Gaudio calls gay “refer to themselves as masu barka ‘those who do the business,’ often abbreviated to masu yi, ‘those who do (it)’” (p. 135n7) and do not see homosexuality as

incompatible with heterosexuality, marriage or parenthood, which constitute strong normative values in Muslim Hausa society: at some point in their lives most of the men I am calling ‘gay’—including those who identify as womanlike—marry women and have children, even as they maintain their covert identity as men who have sex with men. . . . Most Hausa people do not see marriage as a choice but rather as a moral and social obligation [to sire children]. . . . My Hausa acquaintances did not see a necessary connection between marriage and heterosexual desire. (pp. 125-6)

While they seem to regard their homosexual desires as real and as intrinsic to their nature, they also regard their generative status as real, and ultimately more important (in contrast to wasa, “play” (p. 129).

In addition to the gendered idiom for the homosexuality of effeminate ’yan daudu and their patrons, is another for men who are not notably effeminate in self-presentation who have sex with younger men. The older, wealthier man is call k’wazo (a masculine noun meaning hard-working in other contexts), the younger man who is sexually penetrated and receives presents like female sexual partners do is called baja, a feminine noun meaning “merchandise” (p. 130). Grammatical gender here maintains a social gender contrast between same-sex sexual partners, although Gaudio notes that there are also instances of “successful pursuit of a younger man by a ’dan daudu” (p. 131), so that age, wealth, and masculinity are not always neatly bundled. Feminine-identified males sometimes have sex with each other, calling it kifi: “lesbianism” (p. 132), a term also extended to sex between two masculine-identified males—who may ti canji “do an exchange.” Gaudio comments that labeling non-role-bound sex between males as “‘lesbian’ and therefore ‘feminine,’ reveals the normative strength of the idea that a valid sexual encounter involving a ‘real’ man must entail a distinct power asymmetry” and notes that “many gay Hausa men—including ’yan daudu —deride or condemn ‘lesbian’ sexual relations between ’yan daudu as absurd or immoral” (p. 132), although sex between two masculine-identified men is not condemned (p. 133). Apparently, real sex requires at least one masculine person, but need not include a feminine one.

Yoruba Exceptionalism: Antagonistic Acculturation?

Among the neighboring (to the southwest in Nigeria and Dahomey), Yoruba, non-Islamic savanana agriculturalists who have long been intensely hostile-to-Hausa, a psychiatric team in the early 1960s surveyed villagers around Abeokuta, Nigeria about symptoms thought by the Yoruba to indicate or constitute mental and emotional illness. “Homosexuality was not mentioned; and when we asked about it, we were told it was extremely rare, probably found only in changing parts of urban centers” (Leighton et al. 1963:111). The failure to mention homosexuality when questioned about illness is not the firmest basis for drawing conclusions about incidence. It might instead suggest that rural Yoruba did not consider homosexuality an “illness,” as Western psychiatrists of that time did.

As among many other African peoples, among the Yoruba spirit possession (òrìsà gìgún), is primarily associated with women. Most Sango priests are female, and those who are not dress in women’s clothing, cosmetics, jewelry and sport women’s coiffures when they are going to be possessed (Matory 1994:6-7, also see 183-215; Baumann 1955:33-5; Matory 1986:51). About the cross-dressed (cross-coiffed) priests (elegun, literally “mounts”) possessed by Sàngó, Matory wrote, “They regularly have multiple wives and children, and no one seems to wonder if they engage in sex with other men” (1994:208). The Yoruba verb for possession gun (“to mount”)

often implies suddenness, violence, and utter loss of self-control.... Sango will “mount” the initiand bride, an act whose sexual implications are clear. Not only do gods “mount” priests, but male animals “mount” females in the sex act....It does not normally apply to human beings. Applied to human beings it suggests rape. Hence, its application suggests metaphorically the violence and absolute domination implicit in Sango’s command. (pp. 175, 198, 270n16),

And initiates of both sexes are called a “bride” (iyawo) of the god (p. 179)

For the Yoruba—in marked contrast to their religious and political antagonists, the Hausa, and to thewould-be Yoruba emulators, Bahian possession cultists, and to many other peoples—the priests who are mounted by gods are not regarded as being mounted by men. Of the elegun Matory wrote, “They regularly have multiple wives and children, and no one seems to wonder if they engage in sex with other men” (1994:208). Matory assured me that he had broached the subject, and that in this instance absence of evidence is not a function of absence of inquiry. In Afro-Brazilian possession religion which is also has studied,

there is an extensive public discourse suggesting that males who are possessed by spritits tend also to be the type of men (i.e., bichas) who habitually get penetrated sexually by other men. By contrast, my point about the Nigerian possession religions is that they and their [non-believing] neighbors present no public discourse suggesting this link. To me the absence of such a discourse does not by itself mean that no Sango priest is penetrated by other men. It just means that few people, if any at all, in this socio-cultural context recognize penetrated men as a type of social personality and that sexual penetration is not the aspect of the sacred marriage that Oyo-Yoruba royalists highlight in e construction of god-priest relations or, incredible as it may sound, husband-wife relations.... I do not mean to infer that these men never have sex with other men [although] I doubt that they do, based upon my limited but deliberate investigation of the possibility. (personal communication, 12 April 1996)

I accept that there is no Yoruba role in which homosexuality is an expected feature. It seems to me that Yoruba (Christian and “traditional”) “natural sex” is contrasted to the corruption of Islamicized Hausa in the same way highland (mostly Christian) Kenyans Kikuyu view their sexuality as evidence of their moral superiority to coastal (Islamicized) acceptance of transvestitic homosexuality, and that this difference has over time increasingly become a “marker” in ethnic-religious conflict in colonial and post-colonial multiethnic states (in invidious contrast to English as well as against Muslim rivals). While contrasting most Hausas’ ready acknowledgment that homosexuality (including gossip about the involvement of “big men”) exists in Hausaland with many Yorubas’ denial that there is any Yoruba conception or practice, Rudolf Gaudio

categorically reject[s] the idea that there is no homosexual self-identity in contemporary Yoruba communities. I met at least two Yoruba self-identified ‘gay’ men in Kano, neither of whom had ever lived abroad, who told me about the many other ‘gays’ they knew in such cities as Ilorin, Ibadan, and, of course, Lagos where there is a ‘Gentleman’s Alliance’ with pan-Southern menbership. My Kano Yoruba contacts told me that GA members have private parties at each other’s homes, and that there is a division of Yoruba gay male social circles into ‘kings’ and ‘queens. . . When I asked one of these Yoruba ‘queens’ whether there was any Yoruba equivalent to the Hausa ’yan daudu, he said that no, Yoruba queens had more ‘respect’ than the ’yan daudu, insofar as Yoruba queens keep their outrageous, feminine behaviors a secret from other people. (personal communication, 13 March 1997)

Tade Akin Aina, a sociologist at the University of Lagos, wrote about male concubines and male prostitutes (many of whom double as pimps for female prostitutes) in contemporary cities such as Lagos and Kano. Patrons often "operate plural realtionships with a retinue of young men," who may use the resources they derive from their homosexual liaisons to pursue heterosexual sex: "They are often identifiable as big spenders and playboy socialites" (1991:88). Male prostitutes he interviewed in Kano and Lagos

still believe that therea re magical and witchcraft effects associated with male homosexual intercourse. They also believe that if the dominant partner is a businessman, such associations confer spiritual benefits to his business. This, they state, affects the price they place on their services. Also it is felt that homosexuality conveys some unique advantages on its practitioners; for instance, they feel that homosexuals tend to be rich and successful men.... {They] feel that they are at risk of becoming impotent (the "eunuch effect") or permanently incapable of conventional heterosexual relations once any of their clients exploit the relationsh for ritual or witchcraft purposes.... Among the Nigerian prostitutes spoken to modern risks such as AIDS or sexually transmitted diseases did not carry weight as sources of fear. (p. 88)

A Sometimes Lifelong “Adolescent Phase”

Herskovits (1938) reported the view of the Fon, the predominant people in Dahomey (now Benin), that homosexuality was an adolescent phase:

[When] the games between boys and girls are stopped, the boys no longer have the opportunity for companionship with the girls, and the sex drive finds satisfaction in close friendship between boys in the same group.... A boy may take the other ‘as a woman’ this being called gaglgo, homosexuality. Sometimes an affair of this sort persists during the entire life of the pair.

The last statement shows the insufficiency either of the native model or of Herskovits’s understanding of it (or ignorance of an additional model of another role). That the native model is prescriptive rather than descriptive is demonstrated by the need to carefully conceal homosexual relationships which continue beyond adolescence (p. 289). (Herskovits adds, “Homosexuality is found among women as well as men; by some it is claimed that it exists among women to a greater extent” (p. 289). Suspicion of what women are up to among themselves and a strong need for the shelter from male competition/hostility/suspicion in an emotionally-intense best friend relationship are especially strong in West Africa.

Much earlier, Norris (1789:422) mentioned seeing castrated men in his journeying through Dahomey. Captain Sir Richard F. Burton (1924[1865]:123) mentioned that “it is difficult to obtain information in Dahome concerning eunuchs, who are special slaves of the kind, and bear the dignified title of royal wives”: akho’si (fromaAkoshu=king, ‘si=wife) (His observations of troops of the king’s women soldiers are discussed below.)

Langle (1876:243) described a privileged effeminized role at the court in Ouidah. The lagredis of the Dahomey rule were chosen from among the sons of the country’s best families. In their youth they were forced to drink potions that stifled their passions. They had unlimited powers, and their headman played an important part at court and in the king’s council. Two lagredi accompanied any emissary of the king’s, monitored his negotiations, and reported what they observed directly back to the king.

Bastian (1879, vol 3:305) suggested that the king’s monopolization of women made pederasty inevitable. Gorer (1962 [1935]: 141) also opined that women were so monopolized by Dahomean royalty that “sexual perversion and neurotic curiosity were developed to an almost European extent.”

Among the nearby Fanti (of Ghana) there are gender-crossing roles for men and for women (Christensen 1954: 92-3, 143). Those with heavy souls (sunsum), whatever their biological sex will desire women, while those with light souls will desire men, in this worldview. David Greenberg (1989:87) reported interviewing Eva Meyerowitz about her observations among Ashanti and other Akan peoples in what was the British Gold Coast colonybetween the 1920s and 1940s:

At that time men who dressed as women and engaged in homosexual relations with other men were not stigmatized, but accepted. There were good reasons for Akan men to become women, she commented — the status of women among the matrilineal Akan was exceptionally high. The situation may have changed later, she thought, as a result of missionary activity.

Hutschinson (1861:129-30) mentioned male slaves, who were treated as lovers, wore pearl necklaces with gold pendants, and who were killed when their Ashanti masters died, suggesting time depth for a gender-mixing role among the Ashanti.

Signorini (1971) reported that, in contemporary Ghana, Nzema men (who traditionally are forest cultivators around the mouth of the Tano River) marry attractive young men for “social, not sexual consumption.” They speak frankly about attraction to the physical beauty, character, and oratorical skills of their partners. They “fall in love,” contract relationships, and share beds, but persistently denied (to Signorini) that they committed “sodomy.” The husband pays bridewealth to the male wife’s parents, as he would for a female bride, and celebrates with a wedding banquet, regardless of the sex of the wife. The same incest taboos that restrict availability of female wives apply for contracting male-male marriages. Divorce requirements are also the same for male-male couples as for male-female ones. The male wives are younger than the husband, though the disparity seems to be less than trans-generational (e.g., a thirty-year-old husband marrying a twenty-year-old male) and no gender anomalies are involved. Grottanelli (1988:210) also noted age disparities in Nzema “Agyale or ‘friendship marriage,’ usually between a man and a male teenager, more rarely between two women, in which preliminaries similar to those of real marriage are performed, partners cohabit for short periods, exchange presents, and share the same bed or mat.”

Gender-Stratified Homosexuality in and around Senegal

In another West African culture, Senegal (then Saint Louis), Corre (1894:80n1) encountered black men of feminine dress and demeanor, who, he was told, made their living from prostitution. In Boké (Guinea), he saw a prince’s dancer miming his sexually receptive role. Some decades later, Gorer (1962 [1935]: 36) reported that, in Senegal, the matrilineal Wolof

pathics are a common sight. They are called in Wolof men-women, gor-digen, and do their best to deserve the epithet by their mannerisms, their dress and their make-up; some even dress their hair like women. They do not suffer in any way socially, though the Mohammedans refuse them religious burial.

This report was echoed two and a half decades later by Crowder, who saw young males waiting to be picked up in the Place Prôtet, the main square in Dakar (renamed Place de l’Indépendence) and recorded a Wolof word for homosexual: “Gor-Digen. The elders and faithful Muslims condemn men for this, but it is typical of African tolerance that they are left very much alone by the rest of the people” (1959: 68). Gamble (1957: 80) added his opinion that homosexuality was rare in the countryside, but well established in large towns. Davidson (1988[1970]:167-8) recounted the acceptance a transvestite male prostitute received in a Dakar bar in 1958 and (in a section reprinted in this volume) his visit to boy brothels outside Dakar that were patronized entirely by Senegalese. while 17.6 percent of the males and 44.4 percent of the females in an opportunistic sample by Schenkel (1971:343) reported having engaged in homosexuality.

In the exceptionally vague-about-geography generalizations of their experiences at some unspecified time in Senegal and Burkina Faso, Potman and Ruijgrok refer to “militant transvestites and other uninhibited types in bigger cities” and noted

One of us took part in a ceremony where two male couples married in front of their friends. Each of the friends delivered a speech and rings were exchanged. The married couples, however, did not have the means to live together in the same house. Only a few rich urban men have the opportunity to live together. (1993:169)

They also noted that there are both men who “will never consider or identify themselves as homosexuals, even if they have sex with other men regularly” and others who “clearly prefer homosexual contacts” but “will never label themselves publicly as homosexual because of the consequences. They therefore hide both their identity and activities” (p. 169). Within their relationships, “traditional male/female casting plays an important part. Partners characterize themselves as ‘the man’ or ‘the woman’; this choice of social role is reflected also in sexual intercourse, with an active and a passive partner.”

This strong role dichotomization has continued, as Teunis’s observations of a Dakar bar in 1990 show. The men he met there refer to themselves as homoseuèles in French (the language of conversations with him) and gordjiguène in Wolof. The latter is obviously the same word as in the earlier reports. Teunis (1996:161) glosses it as “manwoman,” although it encompasses both insertors (yauss) and insertees (oubi, which literally means “open”), according to an informant. “When two men have sex, the one who acts as inserter is the one who pays money or gives gifts to the other” (p. 161), as men give the women whom they penetrate more than their semen and the stimulation attendant to its ejaculation. However, the sexually receptive males “were able to support themselves and were often wealthier. . . [and therefore] not economically dependent on the yauss” (p. 161).

Teunis also reports that one oubi may “play” sexually with another oubi. The Senegalese who hung out in this bar distinguished such play from “sex” (i.e., anal penetration), but (not least by refusing to participate in it) Teunis did not learn how roles or sexual acts between oubi are negotiated.

A particularly interesting aspect of his article is his claim that physical appearance is the sole basis for assigning a male to the role of yauss or that of oubi. The evidence for his categorical claim, however, is slender and equivocal. He knew one big, strong yauss, whom he calls Babacar, who told Teunis of “his desire for lightskinned (not white) men with big chests and huge muscles, like Rambo” (p. 164). It is less obvious to me than it was to Teunis that this must mean wanting to be topped by a light brown Rambo rather than topping one. Babacar’s sexual partners were boys aged around seventeen who were not part of the bar subculture and who were paid to be sexually receptive by Babacar. Thus, his visible gender and sexual behavior were consistent, while his role in sex with his fantasy object was unspecified.

Teunis’s other basis for concluding that role assignment is immediate and unshakable is that without any indication of his desired role he (Teunis) was typed as an oubi. Even while maintaining his chastity as he did, he might have tried to claim that he was a yauss (or that he was a European gay man and thereby neither or both) to test the unshakability of initial judgement of his womanliness/penetrability that was made by the hustler/pimp who led him to the bar and by its denizens. In that some of Babacar’s boys offered to be anally pentrated by him (p. 164) he was seen by some Senegalese insertees as a possible insertor. I have argued elsewhere that intercultural sexual relations provide an uncertain basis for reaching conclusions about what “natives” generally do with each other, and am not criticizing Teunis for failing to take up invitations for either “sex” or “play,” but he does not seem to have inquired about departures from the neat cultural scenario either.

Leaving Dakar and the present, there are some other reports of feminized boys being sexually used by adult male West Africans. What the Anyi (of the Ivory Coast) do and think are only dimly visible through the nearly opaque a priori Freudian theory of Parin et al. (1980), but they asserted that “in every village there are some men who, for neurotic reasons, do not have sexual relations with women. A number of them are known to practice occasional reciprocal masturbation with boys.” As in Dahomey, “bisexual development may be even more common in women” (p. 204). In one case study (pp. 289-95) the approaches of a gift-giving sorcerer to the male analysand is ambivalently recounted to/by the psychoanalysts. Hammer (1909:198) reported mutual masturbation among young male Dualla in the Cameroons, Torday and Joyce (1906:48) of the Ba-Yaka farmers of the middle Congo River. Wilhelm Hammer noted that homosexual relations were not rare among Liberian Kru youths (1909:198), whose domestic services to men of other tribes included being sodomized, according to Hellwald (1886,4: 591).

Tessman (1921:124-8) described generally accepted, male-male sex among adolesdent Bafia (in the Cameroons), involving penetration of younger boys by older boys and reciprocal anal intercourse (jigele ketön) which he reported was common among friends judged too young to have intercourse with females. They were expected to give up same-sex sex when they reached the ntu age-grade and could have sex with females, and certainly to have outgrown it by the time they became fathers (the mbang age-grade). Tessman believed that not all did, but did not know of any specific instances.

The Kitesha Role Among the Bala

Among the Bala (sometimes referred to as the Basangye in older literature) in Kasai Oriental Province of the Congo, Merriam (1971: 93-4) reported denial of any homosexual behavior in one sentence and in the next reported that Balan men believe the kitesha “is a homosexual.” Clearly, kitesha is a Bala role, a role at variance with the conventional male role in that culture (particularly patterns of dress and of subsistence activity), and comes with concomitant expectations of unconventional sexual behavior. But while Merriam’s three informants (one kitesha, two not) agreed that bitesha (the plural form of kitesha) do not like to work and, improperly for men, go about bare-chested (p. 97), there was no accord in their views of what deviant sexual behavior went with the role: homosexuality, exhibitionistic heterosexuality or asexuality.

The one self-acknowledged kitesha in the village Merriam studied “is married and says that he likes his wife and likes to have intercourse with her,” and was quoted as saying, “I like women; I don’t want to be without them” (p.95). One non-kitesha informant, however, opined,

Bitesha like to expose their genitals in public, but they do not have homosexual relationships. They are not interested in women, and as for masturbation, who can tell? In a desperate situation, a kitesha may have public intercourse with a female kitesha. This is terribly shameful: no one would ever look at such a thing on purpose, but people cannot help seeing. Then they know he is a kitesha and they leave him alone. He does not have to work in the fields (pp. 94-5).

While this statement indicates social tolerance for dropping out of the male role (as something some must do) it does not much clarify kitesha involvement in heterosexual coitus. It also makes one wonder what female kitesha do other than have intercourse with male kitesha.

Merriam’s other informant was reportedly “at no loss” to speak of “males who act like women,” reeling off this list of traits:

He doesn’t want to work. He doesn’t want to be with other men. He doesn’t even have a concubine. He eats everywhere except at his own house. He doesn’t do the things other men do. He never keeps a job. He has good luck. He acts like a woman, that is, rushes about hither and yon and wiggles his hips when he walks. He wears women’s clothing, but not kerchiefs. (p. 94).

The one already-quoted kitesha “says stoutly that it is not shameful for him to be a kitesha but that he knows other men think it is shameful” (p. 97) , while denying both that bitesha commit homosexual acts and that anyone else thinks they do. Although Merriam did not observe the man’s sexual activities, he noted that “others in the village swear to having been eyewitnesses” of Mulenda having intercourse with a female kitesha“ (pp. 96-7).

In his 1974 book, Merriam repeated the discordant social view of the role and Mulenda’s self-report of liking women and adds Mulenda’s assertion that

female bitesha do not like men, prefer the company of women, but most of all wish to be with male bitesha. They like to have sexual relations with men, particularly male bitesha, and they do not like to wear men’s clothing. They do not like to cook or to do other female tasks save to gather wood “because that is the easiest work (p. 321).

Merriam did not encounter any female bitesha, so this statement about female bitesha may be another assertion by Mulenda of his own heterosexuality against the general view of bitesha, acknowledging only the component of laziness.

Although it seems that kitesha is a gender-crossing role (especially in rejecting the work allocated to one’s sex in Basongye society), rather than a primarily homosexual one, a possible reconciliation of the seemingly contradictory views that there is not homosexual behavior among Bala men and that bitesha are homosexuals is that the Bala do not consider bitesha to be men, so that the Bala afford another example of heterogender relations that is not conceived as involving two men but, rather one who is socially male and another who is socially female.

Other Central African Reports

In an earlier report on another Kongo tribe, the Bangala, Weeks (1909: 448-9) reported that mutual masturbation was common and that sodomy was“very common, and is regarded with little or no shame. It generally takes place when men are visiting strange towns or during the time they are fishing at camps away from their women.” Soyaux (1879, vol. 2:59) also attributedboy-love to the lengthy business trips made by Bangala men, unaccompanied by their wives. As Karsch-Haack (1911) noted, this explanation begs the question of why the Bangala travelers do not strike up acquaintances with women as well as with boy, and inferred that although many Mabara make good business profits from their marriages, their real passion is for boys. Although pressing considerably beyond the circumscribed ethnographic record and eager to find such alternative valuations, Karsch-Haack’s inference is not without plausibility.

From north of the Congo River, Karsch-Haack also elicited from Günther Tessman the assertion that the Loanga were known to other tribes as major poisoners and pederasts. Tessman had observed male favorites being free of burdens when even the chief carried loads.

In a much earlier report, Labat (1732: vol. 2, 195-9) wrote of a venerated office of first sacrificer in the Nquiti (Kongo) Kingdom. In Labat’s view the Ganga-Ya-Chibanda was “a bare-faced, insolent, obscene, extremely villanous, disreputable scoundrel.” Usually wearing women’s clothes, he was referred to as “Great Mother.” Able to “perpetrate every villainy without fear of being punished with death, the Ganga-Ya-Chibanda committed the foulest crimes.” Funeral rites were so indecent and shameful “that the paper dirtied with its description would blush.” The condemnations are obviously Labat's, not the Nquiti’s. That the Ganga-Ya-Chibanda had sacred functions made such “deviltry” worse in the pious Christian view.

Trevisan (1986:55) excavated the following interpretation of ways of advertising availability back where the Bahian slaves originated from Denunciations of Bahia, 1591-1593:

Francisco Manicongo, a cobbler’s apprentice known among the slaves as a sodomite for “performing the duty [being penetrated] of a female” and for “refusing to wear the men’s clothes which the master gave him.” Francisco’s accuser added that “in Angola and the Congo in which he had wandered much and of which he had much experience, it is customary among the pagan negros to wear a loincloth with the ends in front which leaves an opening in the rear. . . this custom being adopted by those sodomitic negros who serve as passive women in the abominable sin. These passives are called jimbandaa in the language of Angola and the Congo, which means passive sodomite. The accuser claimed to have seen Francisco Manicongo “wearing a loincloth such as passive sodomites wear in his land of the Congo and immediately rebuked him.” (ellipsis in Trevisan)

Although the closeted anthropological popularizer Colin Turnbull (1965:122; 1986:118) wrote that in Mbuti (pygmie) hunting bands, who live in the Ituri forest at the heart of the Congro basinmale homosexuality is rare and regarded derisively, he wrote to David Greenberg (1989:87) “that when men sleep huddled together, sometimes one ejaculates, but he then ridicules himself for this ‘accident.’ He added that he did not see the Mbuti as rejecting homosexuality so much as favoring procreation very strongly.”

As mentioned earlier (Driberg 1923:210) passed on Lango assertions that homosexual conduct was very common among neighboring groups, specifically the Iteso and Karamojan. Similarly, further south (in southwestern Uganda) Mushanga (1973:181) reports that Nkole informants told him that the Bahima (but not any Nkole) practice homosexuality.

In the old kingdom of Rwanda, Maquet (1961: 77-8) reported that male homosexuality was common among Hutu and Tutsi youth, especially among young Tutsi being trained at court, i.e., they were made sexually available to guests at court. Johnson (1986:29) wrote

Mutabaruka, a 19-year-old [Tutsi] college student told me that, traditionally, in his tribe there was an extended period during which boys lived apart from the rest of the village while they are training to be warriors, during which very emotional, and often sexual, relationships were struck up... “Sometimes these relationships lasted beyond adolescence into adulthood,” he told me. Watusi still have a reputation for bisexuality in the cities of East Africa.

Umuswezi and umukonotsi, two Kirundi (Burundi) words for “sodomite” and at least five Urundi words for male-male sexuality (kuswerana nk’imbwa, kunonoka, kwitomba, kuranana inyuma, ku’nyo) in Burgt’s dictionary belie sodomy being unknown by them. Karsch-Haack (1911) and Bleys (1995:169) also interpreted Burgt (1904:20, 107) as unknowingly revealing Mirundi gender-crossing priests ikihindu and ikimaze (mislabeled by the missionary “hermaphrodite,” a very imprecise label at that time).

An early, highly-regarded ethnography (Driberg 1923:210) uncritically passed on the dubious folk explanation that impotence is the basis for assuming the gender-crossing role among the central African (now Ugandan) Lango (cf. Herskovits 1937: 117). This is remniscent of the frequent confusion about the biological basis of the “berdache” role in native North America. Under the protection of the folk explanation, the mudoko dako [transformed man] is treated as a woman, and may live as a legitimate wife to a man. According to Driberg, such persons were rare among the Lango (50 out of 17,000 people), but they told him that such conduct was very common among neighboring groups, specifically the Iteso and Karamojan. Similarly, further south (in southwestern Uganda) Mushanga (1973:181) reports that the Nkole told him that the Bahima (but not any Nkole) practice homosexuality.

Faupel (1962: 9-10, 68, 82-3) reported that Uganda’s King Mwanga’s 1886 persecution of Christian pages was largely motivated by their rejection of his sexual advances. He found it increasingly difficult to staff his harem of pageboys and is supposed to have been enraged when Mwafu, his favorite, refused any longer to submit to anal penetration by the king.

Half a century later, Bryk (1939:151, 1964:229) reported that

in Uganda I saw two boys, a Mgisho and a Baganda, lying in bed together, whereupon another boy sneered at them with the words, “They love each other like husband and wife.” When one of the embarrassed boys objected, the boys deriding them answered quite rightly, “A man does not sleep with another boy in broad daylight.”

More recently, Southwold (1973:170) reported that Baganda knew that sodomy was practiced, attributed its introduction to Arabs, and “tend to regard it primarily as foolishg: why fool around with a man when women are freely available?” Bryk (1933:150) attested the Arabic hanisi as being used by Baganda for natal males “whose penises had died.”

Laurance (1957: 107) asserted that among the Iteso, people of hermaphroditic instincts are very numerous.... The men are imporent and have the insticts of women and become women to all intents and purposes; their voices are feminine and their manner of walking and of speech is feminine. They shave their heads like a woman and wear women’s ornaments and clothing. They do women’s work and take women’s names,” adding that “I myself know of no cases in which they live with men as a ‘wife,’” but noting that in Serere prison one was kept with the women because “the male prisoners would assault him were he imporisoned in the men’s cell.” He also included a song “The fellow who pinches a hermaphrodite is a fool” (pp. 160-`). Karp and Karp (1973:392) noted group masturbation among young Iteso males in Uganda and Kenya, but had no information on its frequency. Baumman (1899:668) noted an effeminate Wganda “totally given to passive pederasty.” La Fontaine (1959:60-1) reported that the Gisu of Uganda “scorned but did not regard with revulsion transvestite [transsexual in that they denied being men] buyazi.” He added that the three men and one women he encountered cross-dressing in the early 1950s were “not mentally disordered” and were not believed to have any special magical powers. “It is said that today [1950s] transvestism is associated with homosexuality, whereas formerly it was not.” In contrast, in another Uganda Bantu tribe, the Nyoro, Needham (1973:316) suggested a relationship between “sexual reversal (by homosexuality, feminine accoutrements, putative child-bearing, or be other less dramatic means” and the power of diviners [embandwa with a the root nda being a term for womb): “At the ceremony of initiation into the Cwezi cult the novice is given to believe that he must demonstrate his genuine possession by the spirits by becoming a woman.”

In a chapter based on observation of a Ugandan prison Tanner (1969:301) wrote that the “majority could understand but not tolerate homosexuality in others, and they constantly referred to the practice in admonitory terms” (in earshot of colonialists). Tanner (1969:302) elaborated a typology nearly as opaque as the understand/tolerate dichotomy:

A few made quasi-normal adjustments by adapting homosexual practices while in prison in order to get material advantages. A few entered the prison with homosexual traits already in existence and provided service for the previous group [?]; they were mainly Arabs or Somali. They were not usually identifiable by ways of dressing or mannerisms. The majority of all fights were over what prisoners call their ‘wives.’

Southern Africa


Much earlier, the Portugese priests Gaspar Azevereduc and Antonius Sequerius, in Luanda, the capital of Angola met met many men called chibados, who should have but were not ashamed to dress, sit, and speak like women, and who married men just as if they were women in order “to unite in wrongful male lust with them.” Even more shocking to them was that such marriage were honored and even prized (Jarric 1616:482).

The Dongo queen Anna Xinga maintained fifty-to-sixty cross-dressed men, while herself dressing as a man and calling herself (and her cross-dressed courtiers) male, according to Dapper (1670:617).

Speaking generally, Cardonegawrote:

Sodomy is rampant among the people of Angola. They pursue their impudent and filthy practices dressed as women. Their own [language's] name for those who have carnal relations with each other [of the same sex] is quimbanda. Some quimbandas are powerful wizards, who are much esteemed by most Angolans. (1680:86, my translation)

Also in Angola, Hambly’s (1937:426; 1934a: 181) main Mbunda informant Ngonga told of transvestitic medicine men, and related the case of a boy who persisted in dressing in women’s clothes and beating corn (women’s work), despite punishment. Livingstone (1857:452) also mentioned (without hazarding any opinion of their sexual predelictions) many Temba “dandies” whose bodies were adorned all over with decorations and whose hair was so heavily greased that their shoulders were covered with it.

The Kalahari Desert

Kolb (1719:362) mentioned Khoi-Khoin males, called koetsire, who were sexually receptive to other males. Long-time resident of German South-West Africa (what is now Namibia) Kurt Falk (1920) reported homosexuality among the Wawihé, Ovivangella and Nginé of Angola. Falk (1925) maintained that homosexuality among speakers of Khoisan languages

is fairly common among both men and women, and especially among young married women. There exists among the Naman a practice whereby two individuals, either of the same or of opposite sex, will enter into a specially intimate bond of association, soregus, with each other. This is initiated by one of the parties drinking from a bowl of water (or nowadays often coffee), and then handing the rest of the liquid to the other to drink.... As a rule the relationship thus entered upon primarily implies deep friendship and mutual assistance, especially in economic matters. But, according to Falk, it is also used as a means of establishing a homosexual relationship, especially by boys, who jealously watch over each other. The customary form of homosexual practice is mutual masturbation, among both men and women; pedication (coitus in anum) between men, and the use of an artificial penis between women, are also found, but more rarely. Falk mentions also three cases known to him of men who were confirmed sexual inverts. Unfortunately, he gives no concrete data as to the public attitude either towards such men or towards homosexuality in general, but as far as can be gathered from his short account, homosexual practices are regarded somewhat as a matter of custom. (Schapera 1930:242-3; also see Falk in this volume; Schultze 1907:319).

In an extremely hostile characterization, the missionary Johann Irle (1906:58-59) depicted the the nomadic pastoralist Herero people (of what is now Namibia) as liars and thieves who devote themselves to sensual pursuits from the age of ten and, when challenged, say that they have grown from childhood in their own natural way. Rather than speak of the unspeakable, Irle invoked St. Paul’s antihomosexual Romans I:18-31 (forsaking the natural use of women and burning with desire for each other) as the only indication of what these sensual pursuits were. Falk (1925:205-6), however, specified that Herero special friengships (omapanga) included anal sex (okutunduka vanena), as well as mutual masturbation.

The Finnish missionary M. Rautanen (1903: 333-4) wrote that both pederasts and men presenting themselves as women (eshenga) existed and were detested by the Ondonga. Most of the latter, he wrote, were shaman (oonganga).

In the Oshiwambo language spoken by about 600,000 Ovambos in northern Namibia, anally receptive males are called eshengi (plural ovashengi), “he who is approached from behind.” According to University of Namibia historian Wolfram Hartmann, “the ovashengi of the Unkwambi, a subgroup of the Ovambo, are respected as healers, or onganga. Among another Ovambo subgroup, the Oukwanyama, the ovashengi are not treated as well; however, they are the only Oukwanyama members entitled to play the ekola, a spcial music[al] instrument” (in Jones 1996:41).

Among the contemporary Ambo (in southwestern Angola) Estermann (1976:197) noted that the highest order (called ovatikili) of diviner (kimbanda) who “recruited exclusively among men [who] are few and feared and their activity is surrounded by profound mystery.” Although Estermann did not say that all ovatkili are possessed by female spirits, he wrote that a large number of the owners of omakola, a musical instrument translated as “big gourd” were passive homosexuals (omasenge), who dressesd like women, did women’s work, and ‘contracted marriage’ with other men” who already had female wives. He explained,

In a general way this aberration is to be interpreted by the spiritism or spiritist belief of these people. An esenge [singular of omasenge] is essentially a man who has been possessed since childhood by a spirit of female sex, which has been drawing out of him, little by little, the taste for everything that is masculine and virile.

Also in Angola, Hambly’s (1937:426; 1934a: 181) main Mbunda informant Ngonga told of transvestitic medicine men, and related the case of a boy who persisted in dressing in women’s clothes and beating corn (women’s work), despite punishment. Livingstone (1857:452) also mentioned (without hazarding any opinion of their sexual predelictions) many Temba “dandies” whose bodies were adorned all over with decorations and whose hair was so heavily greased that their shoulders were covered with it.

Southeastern Africa

In Zambia, Poewe (1982:172) mentioned a cross-gender role without any indication of sexual components of the role: “Dual sexuality, if such a designation is appropriate, refers to the fact that a person who is anatomically male, can play a female role. For example, a person’s maternal uncle, specifically “his/her maternal uncle”, is nalume. Na means mother of, lume means masculine gender.”Hilarity” is her characterization of the people she studied in Luapula “on the rare occasions” that they mention forms of sexuality other than heterosexuality (p. 147).

In his extensive description of Mukande initiation rites in what is now northwestern Zambia, Victor Turner noted that initiates mimed copulation with a senior male’s penis. The Mukande considered this a way to strengthening the novice’s own penises. Earlier, among the matrilineal Balovale savana-farming peoples (Chokwe, Luchazi, Lucho, Lunda, and Luvale) in the same savanna region, during the first phase of initiation while the initiates are recovering from their cirumcisions, they wore no clothes. During that phase, novices played with the penises of the vilombola (keeper of the initiation lodge) and tulombolachika (initiated assistants of the vilombola). White (1953:49) reported that “this is considered to hasten healing; the novices also hope that by so doing, their own penes will grow large and strong. The same is done to visitors to the lodge.”

In what is now southwestern Zimbabwe, Livingstone (1865:284) asserted that elderly chiefs’ monopolization of women was responsible for “immorality” of younger men. Smith and Dale (1920, II: 74) mentioned one Ila-speaking man who dressed as a woman, did women’s work, lived and slept among, but not with, women. The Ila label mwaami they translated as “prophet.” They also mentioned that “pederasty was not rare,”but was considered dangerous because of the risk that the boy will become pregnant (p. 436),.

Although the current official history of (male) homosexuality in Zimbabwe is that it diffused from South Africa during the 1950s, Epprecht’s (1996) review of 250 court cases from 1892 to 1923 found cases from the beginnings of the records. The five 1892 cases all involved black Africans. A defense offered was that “sodomy” was part of local “custom.” In one case a chief was summoned to testify about customary penalties and reported that the penalty was a fine of one cow, which was less than the penalty for adultery. Over the entire period, Epprecht found the balance of black and white defendants proportional to that in the population. Of course, as he notes, what came to the attention of the courts was far from a random sample of sexual behavior in the colony, most consensual relations in private not provoking notice by the courts. Some cases were brought by partners who had been dropped or who had not received promised compensation from their former sexual partner. And although the norm was for the younger male to lie supine and not show any enjoyment, let alone expect any sexual mutuality, Epprecht found a case in which a pair of black males had stopped their sexual relationship out of fear of pregnancy, but one wanted to resume mounting and loving each other, i.e., taking turns anally penetrating each other.

Monica Wilson’s 1951 book (based on fieldwork during the mid-1930s) of Bantu-speaking Nykakyusa northwest of Lake Nyasa (on both sides of the Tanzania/Zimbabwe border) is a classic account of extreme age-grading. The inhabitants of each village contained only one generation (spanning five to eight years; p. 32) of males. For unmarried male contemporaries to dwell together for awhile in temporary villages or in mens’ houses (especially during initiation) was common in “traditional” (tribal) Africa.

The peculiarity of the Nyakyusa consists in the fact that contemporaries live together permanently through life, not merely as bachelors.... The Nyakusa themselves associate living in age-villages with decency inm sex life—the separaation of the sex activities of successive generations, and the avoidance of incest.... The emphasis on the separation of parents and children is matched by the value laid on good fellowship (ukwangala) between contemporaries... The value of good fellowship with equals is constantly talked about by the Nyakyusa, and it is dinned into boys from childhood that enjoyment and morality alike consist in eating and drinking, in talking and larning, in the company of contemporaries. (p. 159, 162-3)

Nykakyusa derided heterosociality, and feared sharing of food and beer between fathers and sons. “To force a fellow to have homosexual intercourse against his will was a serious offence, comparable to witchcraft” (p. 88) or is witchcraft (p. 196). An “exceptionally reliable informant” told Wilson that a man “never dreams of making love to another man” and that “not many cases of grown men having intercourse together come to light, but only of boys together or of a man and a boy.” He elaborated:

When a boy sleeps with his friend they sleep together; it is nor forbidden. Everyone thinks it all right. Sometimes when boys sleep together each may have an emission on the other (bitundanila).. If they are great friends there is no wrong done.... Boys sometimes agree to dance together (ukukina) and work their evil together and that also is no wrong.... Boys do this when they are out herding; then they begin to dance together and to have intercourse together.... To force a fellow this us witchcraft (bo bulosi); he is not a woman. But when they have agreed and dance together, then even if people find them, they say it is adolescence (lukulilo), all children are like that. And they say that sleeping together and dancing is also adolescence. (p. 196)

He reported that interfemoral intercourse is “what boys mostly do” and also reported anal and oral sex. He was aware of and disapproving of “some, during intercourse, work[ing] in the mouth of their friend, and hav[ing] an orgasm.... That of the mouth people do very rarely when they dance together” (p. 196).

Egalitarian/reciprocal sex between adolescent friends seems to have been most common, although some age-stratification also occurred:

When out herding, some of the older boys do evil [see note 43] with the young ones, the older persuade the little one to lie down with them and to do what is forbidden with them between the legs. Sometimes two older boyrs who are friends do it together, one gets on top of his fellow, then he gets off and the other one mounts. (pp. 196-7).

Two other informants agreed that homosexuality occurred frequently in boys’ villages: “A boy has intercourse with his fellow, but a grown man? No, never, we’ve never heard of it it. They always want women; only when a man cannot get a women he does this, only in youth. A few men do not marry but they are half-wits who have no kind of intercourse at all” (p. 197).

In addition to gender-crossing Zulu possessed diviners, there appears to have been at least occasional substitution of boys from women in establishing the potency of men seeking recognition as adult warriors (Krige 1965:276-7; Morris 1965:36,52), and men/women not engaged in spirit mediumship: Vilkazi (1962) attested that such men were called inkosi ygbatfazi, chief of the women.

Morris (1965:46) Robert (1975:86-7) suggested that the great warrior Shaka, who forged a Zulu kingdom in southern Africa prior to European incursions, was homosexual. He had no wives, sired no children and preferred the company of the uFasimba, a regiment of the youngest bachelors (Morris 1965:36). His soldiers “must not be enervated by matrimony and softened by family ties,” until “the advance of years which, while unfitting him for soldiering, won him permission to marry” (Ferguson 1918:206). Journalist John Gunther (1955:523) contrasted “tribes like the Zulus [who are] sternly upright and moral for the most part and tribes greatly addicted to homosexuality, like the Pondos,” whatever this may mean.

Wives of the South African Mines

The 1890s were a time of violent dislocation of black South Africans. One notable group of rebels/bandits south of Johannesburg, called Umkhosi Wezintaba, the Regiment of the Hills, by its Zulu refugee Inkoos Nkulu (King), ‘Nongoloza’ Mathebula, emulated Shaka’s armies. The Zulu leader who took the name Jan Note and was called by whites the “King of Nineveh,” ordered his (mostly non-Zulu) troops to abstain from all physical contact with females: “Instead, the older men of marriageable status within the regiment—the ikhela—were to take younger male initiates in the gang—the abafana—and keep them as izinkotshane, ‘boy wives’” (van Onselen 1984:15). In 1900 Nongoloza was captured, but his organization extended from townships to mining camps to prisons, in all of which the sex ratio was very skewed and men’s concern was great about veneral disease among such few women as there were. The homosexual relations were not , however, a result of the prison environment. Nongoloza testified that the practice [hlabonga] “has always existed. Even when we were free on the hills south of Johannesburg some of us had women and others had young men for sexual purposes” (Director of Prisons Report, quoted by Achmat 1993:99). In castigating social historians (van Onselen in particular) for elliding local understandings (this self-report by the leader of the Ninevitesin particular), Achmat, who as a political prisoner in 1978 had personal experience of the enduring Ninevite allocation of young prisoners for sex with Ninevite leaders who then undertook their protection, stresses that “Nongoloza did not apologise for the fact that some of the Nineveites ‘had young boys for sexual purposes.’ He did not try to justify its existence by referring to verneal disease or tradition. Instead, he justified it in terms of sexual desire” (Achmat 1993:100)

The Swiss Presbyterian missionary Henri Junod vacillated between attributing elaborately organized homosexuality among Tsonga migrant laborers in South African mines to the unavailability of women in labor camps or to a pre-existing, indigenous homosexual preference. The nkhonsthana, boy-wife, “used to satisfy the lust” of the nima, husband, received a wedding feast, and his elder brother received brideprice. Junod mentioned that some of the “boys” were older than 20, and also described a transvestitic dance, tinkhonsthana, in which the nkhontshana donned wooden or cloth breasts, which they would only remove when paid to do so by their nima (1927: 492-3, 294).

An aged Tsonga named Philemon recalled that wives of the mine (tinkonkana) were expected to perform domestic chores for their “husbands”:

Each of these xibonda [room representatives] would propose a boy for himself, not only for the sake of washing his dishes, because in the evening the boy would have to go and join the xiobonda on his bed. In that way he had become a wife. The husband. . . would make love with him. The husband would penetrate his manhood between the boy’s thighs (Sibuyi 1993:53).

The grateful husbands bought presents for these wives, including clothes, blankets, and even bicycles. Moodie (1988:231) explained that intercrural intercourse

is typical of a form of sexual play amongst adolescent Nguni boys and girls called metsha among Xhosa-speakers and hlobongo by the Zulu. These young ‘boys’ of the miners are not merely sexual partners, but are also ‘wives’ in other ways, providing domestic services for their ‘husbands’ in exchange for substantial remuneration,” adding that these homosexual dyads occurred “almost exclusively between senior men (men with power in the mine structure) and young boys. There is in fact an entire set of rules, an mteto, governing these types of relationships, whose parameters are well-known and enforced by black mine authorities.

Fidelity was expected and jealousy on occasion led to violence. Philemon was very explicit that “some men enoyed ‘penetrating the thighs’ more than they did the real thing [i.e, penetrating vaginas] (p. 54). Moreover, agency was not always a monopoly of the established elder, since he mentioned the consequences “when a boy decided to fall in love with a man. . .” (p. 54) and that male couples “would quarrel just as husbands and wives do. Some quarrels would also lead to divorce” (P. 58). When Sibuyi asked Philemon whether the boys wishes to become someone’s wife, he replied, “Yes: for the sake of security, for the acquisition of property and for the fun itself” (p. 62)

The Taberer Report in 1907 noted that

it appears to have become a well-recognized custom among the mine natives recruited from the East Coast to select from the youths and younger men what are termed amankotshane or izinkotshane. An inkotshane may be described as a fag and is utilised for satisfying the passions. Any objections on the part of the youth to becoming an inkotshane are apparently without very much difficulty overcome by lavishing money and presents upon him. . . An inkotshane’s duty appears to be to fetch water, cook food and do any odd work or run messages for his master and at night time to be available as bedfellow. In return for these services the inkotshane is well fed and paid; presents and luxuries are lavished upon him (p. 2, quoted by Moodie 1988:234).

An old Mpondo added that tinkonkana “were boys who looked like women — fat and attractive” (p. 232).

Cross-dressing does not appear to have been a requisite of the wife role, although there are archival reports of parties in which some boy wives donned artificial breasts and impersonated women. Proper wifely behavior did not include ejaculation by the youth, or any kind of sexual reciprocity. Another

The boys would never make the mistake of ‘breathing out’ [ejaculating] into the hubby. It was taboo. Only the hubby could ‘breathe out’ into the boy’s legs. . . . [Another] thing that a nkonkana had to do was either to cover his beard with cloth, or cut it completely off. He was now so-and-so’s wife. How would it sound if a couple looked identical? There had to be differences, and for a nkonkana to stay clean-shaven was one of them. Once the nkonkana became a ‘grown-up,’ he could then keep his beard to indicate his maturity, which would be demonstrated by him acquiring a boy. . . . When the boy thought he was old enough he would tell the husband that he also wished to get himself a wife, and that would be the end. Therefore the husband would have to get himself another boy. (Philemon, in Sibuyi 1993:58, 57, 55).

As in ancient Greece, a beard indicated that a boy had become a man and was no longer a sex object for other men, but was now a competitor for boys.

Although (like Gill Shepherd) preoccupied with economic aspects of relationships and unwilling to take seriously statements such as “We loved our boys better [than our wives],” Moodie (1988) interpreted the institution of mine marriages as partly a resistance to the proletarianization of those laboring in the mines but committed to traditional life back home, e.g., describing a young man in 1940 who consented to be a tinkonkana, because he wanted to accumulate money to buy cattle for brideprice. Thus, in some cases “men became ‘wives’ on the mines in order to become husbands and therefore full ‘men’ more rapidly at home” (p. 240). Those playing the wife role could accumulate money (bridewealth and gifts) — while the husbands not only received domestic and sexual service, but spent less than they would have in dance halls with women prostitutes.

Although these relationships for the Mpondo seldom extended beyond one contract and were never brought home, and although men preferred to conceal these liaisons from their home fellows, everyone knew that such affairs existed and joked with each other about them.... According to Philemon, among the Tsonga ‘mine marriages’ were accepted, indeed taken for granted by women (including wives) and elders at home and relationships might extend beyond a single contract. (Moodie 1988:233).

Philemon mentioned that when one partner finished his contrast before the other the husband might go to their boy-wives’ homes or boy-wives’ might go to their husband’s homes. They would be “warmly welcomed” (p. 57), everyone knowing “that once a man was on the mines, he had a boy or was turned into a wife himself” (56).

At least some of the older men who continued to return to the mines may have preferred young male “wives,” though Moodie (1988) and van Onselen (1976) did not consider this. In recent years the “wife of the mine” role has declined, if not disappeared. Migrant labor has become less common. With the breakdown of rural society, wives accompany or follow their husbands and live as squatters near the work sites. Moodie (1988:255) concluded,

It is precisely because mine marriages were isomorphic with marriages at home that they are breaking down as the home system collapses.... The old arrangements represented accommodations to migration and at the same time resistance to proletarianisation.... The contemporary turn to ‘town women’ and squatter families represents accommodation to the exigencies of stable wage-earning.

Although the “wife of the mine” role developed under conditions of migration to capitalist enterprises owned and operated by white Europeans, there is no evidence of it being imposed or suggested by white Europeans. As an adaptation to the conditions of prolonged sex segregation it drew on conceptions of what a Tsonga wife should be and on behavior (viz. intercrural intercourse) within the existing repertoire of adolescents in the rural society. Sources do not discuss whether the same asymmetry of roles occurred among boys involved in sex play with each other in the home villages, or whether certain individuals specialized in sexual receptivity and were regarded as effeminate. Shangaan workers from Mozambique were reputed to be the most frequent and enthusiastic participants in bukhontxana (Harries 1990:327), and some attempts to claim native purity contended that homosexuality was introduced by these “foreigners.”

The elderly Pondo ex-miner Themba, whom Ndatshe interviewed in 1982, told her that “most of the miners agreed to be ‘girls of the mines. Some wanted to pay lobola once they had returned to their homes. . . . He also mentioned that boys always said, ‘Why should we worry since we can’t get pregnant” (Ndatshe 1993:51) and stressed the domestic duties of cooking, washing, and cleaning that went with the role.

During the 1950s in “Mkumbani,” a black settlement adjacent to Durban, there were, according to Louw (1996) male-male weddings at the rate of about one a month. Each celebration lasted a whole weekend. Some “brides” wore Zulu dress, some wore Western bridal white (with bridesmaids in attendance). The pair might live together, with the “wife” doing domestic work, or live spearately (like many male-female couples). Some of the males involved in these relationships already had female wives (who treated the new one as a junior wife, one who would not complicate life by producing children rivalling her own). Louw suggested that the term for the female-gendered homosexual men, skesana might be cognate to zenana. Their masculine partners were called iqgenge.

In other indigenous cultures of black, sub-Saharan Africa I have reviewed instances both of reciprocal adolescent male homosexuality and of roles involving sexual receptivity to men along with other expected aspects of female roles. Discussion of black African homosexuality may be a taboo among Africans sensitized by missionaries and/or Western education to the Christian stigmatization of homosexuality, anxious about the ill repute of black people’s hyper-sexuality, and/or resentful of sexual exploitation of boys in mission boarding schools. This taboo, these anxieties and resentments are products of European domination. Rather than homosexuality having been introduced to Africa by (“corrupt”) Europeans, the colonialists carried and transmitted intolerance for homosexuality. There were labels — often derisive labels — for gender-crossing roles. As elsewhere in the world, an innovation of European penetration was pressure to desist from such immorality, or at least to hide and deny there was any.

Contemporary Black South Africa and Zimbabwe

In the black townships of contemporary South Africa, Miller (1992:4,8) reported a party sponsored by Gays and Lesbians of the Witwatersran (GLOW) in which

the 40 or so guests were mostly men, all quite young and very campy. They referred to their boyfriends as wives and barked orders at them. The boyfriends were deferential... GLOW attracted younger black men who, traditionally deprived of many social outlets, saw the organization as a place to have fun.

Similarly, Isaacs and McKendrick (1992:110) mention

a discotheque in Salt River, adjacent to the [Cape Town] suburb of Woodstock, on occasions hosts drag competitions for so-called ‘couloured’ gays.... Of note [in a September 1988 observation] was the number of families of gay participants who offered not only their moral support, but who applied make-up and who helped to costume their sons.

Although writing about the emergence of white South African gay identity and community, they also noted that

Soweto reputedly has 15 small shebeens and music halls exclusively for African gays. In Cape Town, a new group of African gays has recently emerged under the title of the African Gay Association (AGA), and has an active membersjip of up to 70 persons. A particular club in the centre of Cape Town provides a venue for predominantly ‘coloured’ gay men and women (p. 100).

A 28-year-old (soon thereafter dead) black gay leader from Soweto called Linda Ngcobo, the founder of GLOW, described township gay culture to Miller (1992:14, 17) as centered on cross-dressing and role-playing:

No one, including gay men, seemed to be quite sure what “gay” meant — were gay men really women? men? or something in between? This confusion was compounded by the fact that sex education was extremely limited in the townships; moreover, among some tirbes, talking about sex at all was taboo. “Moffie,” the derogratory Afrikaans expression for male homosexual, widely used in South Africa, literally means [and abbreviates] “hermaphrodite.”... Some, like Linda, were treated as women by their families [his parents expected him to do “women’s work, including washing, ironing, and baking] and seemed to believe that deep down they actually were females.

Linda also recounted the belief of an early boyfriend and his family that he could bear a child.

I wouldn’t have lent much credence to this story except for an article about township gay life I had read in the Weekly Mail newspaper. In the article, the reporter asked a member of a gay male couple if he practiced safe sex. The man said he was “scared” of condoms, because, “if I used them I won’d be able to have a baby. I am throwing my sperm away.” The reporter pressed him: By sleeping with a man, wasn’t he throwing his sperm away as it was? “No,” the man replied. “One day my ‘wife’ and I hope to have children like any normal couple” (p. 15).

Male pregnancy is not the only “exotic” belief about homosexuality. A 32-year-old man named Robert who had left his 20-year-old male second wife along with a female wife and child in the diamond-mining center of Kimberley told Miller (1992:54) that if his parents find out about his homosexuality they “will think I am bewitched and will take me to a witch doctor, “ as Simon Nkoli’s did.

Nkoli recounted that in Soweto

townships, people use the Sesotho wor sitabane to describe gay people. It means a person who has got two organs—sexual organs—a vagina and a penis. . . . We never really saw people with two organs. But if you are a feminine man, many will say you are a sitabane. And if you happen to be over twenty years old and you don't have a girlfriend, many will presume you are gay. . . [even though] in the black community, people don't know what gay is. (1993:22)

Nkoli's stepfather was certain that Nkoli must be normal, because he did not have a female organ (along with his male one), and his siblings were asked, "Is it true that your brother has got two sexual organs?" and fellow students would ask to examine his genitalia (pp. 23-4). A lesbian born in Soweto in 1974, seventeen years later than Nkoli encountered the same belief that "gays have two genitals" and for a time though, "I can't be gay because I have to have two things" (Mamaki 1996:3). The general Southern African derogation moffie derives from (condenses) hermaphrodite, as already mentioned.

Whether the indigenous cultures conceive of two or three sexs/genders, husband-wife roles continue to be the idiom for same-sex relationships in the homelands. The relationships also seem to be marked by at least some disparities in age, compounding the patriarchal authority, as also in black African male-female relationships.

A public black gay subculture has been identifiable in Cape Town since at least 1950 (Gevisser 1995:72) reported on in national publications aimed at black audiences. Chetty (1995) tracked down three of the “moffie” performers photographed and discussed in the 1950s. All three were hairdressers and all three reported early and ongoing family support for both homosexuality and cross-dressing from an early age.

In the largest cities, an interracial gay liberation movement is mobilizing, as Bull (1990) and Miller (1992), and Gevisser and Cameron (1995) document. On 13 October 1990 in Johannesburg 800 (30% of whom were black) people participated in the first Gay Pride Parade in Africa (Bay Area Reporter, 1 Nov. 1990).

The interim constitution’s bill of rights (section 8, part 2) that followed the end of apartheid and the coming to power of the African National Congress (ANC) in 1994 included a sexual orientation clause. The major opposition party in South Africa, the Intkatha Freedom Party, also included sexual orientation in its proposed constitution (Gevisser and Cameron 1992:96). The ANC provision was incorporated in the new constitution (as section 9, part 3) that was adopted on 8 May 1996, making South Africa the first nation in the world with protection of persons from discrimination by sexual orientation included in its constitution. Patron (1995:22) suggests that “perhaps it is because the South African liberation struggle lasted so long that the liberation movement was able to achieve a level of maturity that recognized the necessity of full and genuine inclusion of all minorities in society”— in explicit contrast to neighboring Zimbabwe.

Lurink (1995) reported:

Until July this year, gays and lesbians in Zimbabwe lived relatively undisturbed lives, even though the number of gay bars in Harare was limited eventually to one. And to use the term "homosexuality" in the media was not done. Maybe the gay movement’s "coming out" by applying for a stall at the book fair, and its first publications appearing in Shona and English, were going just a step too far. Maybe it was seen as a provocation by those in power. Maybe the economic misery and the approaching presidential elections next year created a need for a new scapegoat. Maybe the South African winds of change shook nerves in Harare’s presidential offices. But Evelyn’s explanation is more psychological. She suspects that Mugabe himself, when he underwent a 12-year prison sentence under Ian Smith, may have been sexually abused. "It happens all the time in our prisons. Moreover, his white adversaries once smeared him as a `moffie.’"

Evelyn also supplied Lurink with a Shona word for "gay": ngochane. Coutinho (1993:63) noted that homosexual behavior was not sanctioned in traditional Shona and Ndebele law and reported that "older gay men tell tales of dodging the police while cruising Salisbury’s Cecil Square (now Africa Unity Square in Harare) as far back as the early 1950s. Older black gay men recall tales of romance and hardship on the other side of town in the single-sex hostels for migrant labors" and also in prison. An organization called Gays and Lesbians of Zimbabwe was formed in 1990. "With its limited and apolitical agenda, GALZ is treading a cautious path designed to break down isolation without drawing unwanted attention from the police" (p. 65), and in the townships "small networks of gay men are emerging" (p. 64).

Relations Between Women


Among the much-discussed Zande of the Sudan, the Seligmans (1932: 515) reported that sisters who are married/retained by brothers "have a reputation for lesbian practices." Evans-Pritchard (1970:1429, 1431-2) wrote

All Azande I have known well enough to discuss this matter have asserted the female homosexuality was practiced in polygamous homes in the past and still (1930) is sometimes.... One of the many wives of a prince or of an important commoner in the past might not have shared her husband’s bed for a month or two, whereas some of the dozens, even hundreds, of wives of a king must have been almost totally deprived of the sec life normal in smaller homes. Adulterous intercourse was very difficult for a wife in such large polygamous families, for the wives were kept in seclusion and carefully watched.... Wives would cut a sweet potato or manioc root in the shape of the male organ, or use a banana for the purpose. Two of them would shut themselves in a hut and one would lie on the bed and play the female role while the other, with the artificial organ tied around her stomach, played the male role. They then reversed roles.

In that late article, he reproduced texts from two male informants about special friendship relations (mimicking male blood brotherhood) between women. beginning by break a cob of blood-red corn (kaima) and uttering a spell over it.

After this they should not call each other by their proper names, but they call each other bagburu. The one who is the wife cooks porridge and a fowl and brings them to the one who is the husband. They do this between them many times. They have sexual intercourse between them with sweet potatoes carved in the shape of a circumcised penis, with carved manioc also, and also with bananas. (p. 1432)

This relationship was not approved by Zande men. As Evans-Pritchard observed, "It is a further indication of male dominance that what was encouraged among males was condemned among females. Zande men, princes especially, have a horror of lesbianism, and they regard it as highly dangerous" (Ibid.).

Closer to the time of his field-work he (1937:56) had written that the Zande refer to sex between women with adandara, a species of wild cat which is considered female and unlucky: "both are female actions which may cause the death of any man who witnesses them." With something short of an anthropological attempt to understand behavior as it is understood by the women involved, he wrote,

Zande women, especially in the homesteads of princes, indulge in homosexual relations by means of a phallus fashioned from roots. It is said that in the past a prince did not hesitate to execute a wife whose homosexual activities were discovered, and even today I have known a prince to expel wives form his household for the same reason. Among lesser folk, if a man discovers that his wife has Lesbian relations with other women he flogs her and there is a scandal. The husband’s anger is due to his fear of the unlucky consequences that may ensue from such practices. Azande therefore speak of them as evil in the same way as they speak of witchcraft and cats as evil, and they say, moreover, that homosexual women are the sort who may well give birth to cats and be witches also. In giving birth to cats and in Lesbianism the evil is associated with the sexual functions of women, and it is to be noted that any unusual action of the female genitalia is considered unlucky.

In this passage it is not clear to me whether Evans-Pritchard was arguing that it is not the sexual behavior that is culturally problematic or whether he is rationalizing male fear of women. Lagae (1926:24) did not elaborate on the cosmology of the recurrent sexual relations between wives.

Bieber (1909) mentionsed "tribadie" being as frequent as pederasty among among East Horn tribes (Harari, Galla, Somal). Among Amhara peasants, Messing (1957:550) found "mannish women" (wändawände) suspected of attempting to abrogate male privileges.

Roux (1905) reported sex between women on the Comoro islands (near Madagascar).


In arguing against a third sex or gender conception in Oman or Mombasa, Shepherd (1978a: 133) wrote, "Lesbians [in Mombasa] are known as wasaga (grinders).... The dominant partner... is not seen as a man." She had earlier claimed, on the basis of an unmentioned sample of wasagas (Shepherd 1978a:134), that "there is almost always a dominant and subordinate economic relationship between them." Shepherd (1987:254) elaborated,

The word in Swahili glossed as ‘lesbian’ is msagaji (plural wasagaji) - ‘a grinder.’ The verb kusaga (to grind) is commonly used for the grinding of grain between two millstones. . . The upper and lower millstones are known as mwana and mama respectively: child and mother.

Clear status distinctions characterize msagaji relationships, with the dominant woman usually being older as well as wealthier. Whether these distinctions affect sexual behavior inside the relationship, Shepherd did not report.

Against any claim of a third-gender role Shepherd (1987:259-60) argued,

Lesbians dress entirely as women. Their wealth enables them to dress in a rich and feminine way, and though dominant lesbians are more assertive in manner and conversation than most other women, they make no attempt to look like men. When they go out they wear the veil (buibui ) like all other coastal women. . . Dependent lesbian women are expected to behave like ordinary women. Dominant lesbian women display energetic personalities very similar to those of active, intelligent non-lesbian women.

However, Shepherd (1987:260) also reported, "Women who are dominant lesbians do not obey strict seclusion rules. As household heads they welcome male visitors to the house and sit with them in the reception room, and they frequently go out of the house." Her analyses of homosexuality among women, as among men, are contradicted by the behavior she reported. Msagaji seem to be bidding for some male privilege beyond that of having sex with submissive women. That is, they do not (entirely) conform to Mombasa conceptions of how women should look and act.

As for the shoga, Porter (1995:144) stressed gender variance as more salient than sexual behavior for labeling women misago: Especially women resisting marriage and interested in education and careers are labeled "misago regardless of the erotic preferences of the women. They are being condemned for behaving in ways that are inconsistent with being a woman and for challenging the gender/status system." The word misago itself is derived "from the same verb root as msagaliwa: kusaga, a Bantu word meaning to grind. In this instance, however, it refers to the sexual activities of women, as ‘grinders,’ rather than to grinding wood" (p. 142). The etymological primacy of sexuality does not prove that sexuality is more important than gender in the use or native understanding of the word, but more systematic data collection than either Shepherd’s or Porter’s would be necessary to settle the question of primacy of gender or sexuality.


Some Kaguru men (in central Tanzania) mentioned to Beidelman (1973:266) that

some Kaguru women practise lesbian activities during female initiation, women taking both the roles of men and of women in demonstrating sexual congress to initiates. Women were unwilling to discuss this in detail with me, but conceded that women did demonstrate with one another how to have proper sexual congress.

West Africa

According to Eva Meyerowitz’s fieldwork in the Gold Coast (now Ghana) during the 1940s, "lesbian affairs were virtually universal among unmarried Akan women, sometimes continuing after marriage. Whenever possible, the women purchased extra-large beds to accommodate group sex sessions involving perhaps half-a-dozen women" (D. Greenberg 1989:66; he considered this an instance of egalitarian homosexuality, perhaps from discussion about the relative status of the partners with Meyerowitz, although he published no indication of relative age or status).

In his contribution below, Carrier and I attempt to sort out the literature on woman/woman marriage and purported chastity within the institution in East and West African cultures. An Efik-Ibibio (Nigerian) woman who grew up during the nineteenth-century (and whose husband had eleven other wives) recalled: "I had a woman friend to whom I revealed my secrets. She was very fond of keeping secrets to herself. We acted as husband and wife. We always moved hand in glove and my husband and hers knew about our relationship. The village nicknamed us twin sister" (in Andreski 1970:131).

Nadel (1942:152, 1954:179) made passing mention of Nupe lesbian behavior in Nigeria. Gaudionotes that there are Hausa lesbians (in Kano of the 1990s), some of whom know of and use the terms for male homosexual roles (discussed above) and that gay Hausa males assert that "lesbians engage in the same types of relationships as they themselves do, i.e., involving the exchange of material gifts and the attribution of [dominant/submissive] roles" (1996:136n17).

Mention by Froelich (1949;115-8) of a separate (longer) initiation rite, called kpankpankwondi, for girls in northern Togo who refuse to marry the husbands selected for them is tantalizing.

"Amazon" Troops in 19th-Century Dahomey

In his two-volume account of his 1863-4 mission, Burton (1864:ii, 63-85) devoted a chapter to "the so-called Amazon" troops of King Gelele. In his view "the origin of the somewhat exceptional organisation" of the women troops that he estimated to number about 2500 was "the masculine physique of the women, enabling them to compete with men in enduring toil, hardship and privations" (p. 64), although he also offered a historical explanation — the early 18th century king Agaja’s depleted ranks of male soldiers (p. 65). Noting that Dahomeyan women "endured all the toil and performed all the hard labour" there was to do, Ellis (1965[1890]:183) elaborated on the historical evolution of the Amazon institution:

The female corps, to use the common expression, the Amazons, was raised about the year 1729, when a body of women who had been armed and furnished with banners, merely as a stratagem to make the attacking forces appear larger, behaved with such unexpected gallantry as to lead to a permanent corps of women being embodied [by King Trudo]. Up to the reign of Gezo, who came to the stool in 181, the Amazon force was composed chiefly of criminals, that is criminals in the Dahomi sense of the word. Wives detected in adultery, and termagants and scolds were drafted into its ranks; and the great majority of the women "given to the king" by the provincial chiefs, that is, sent to him as being worthy of death for misdemeanours or crimes, were, instead of being sacrificed at the Annual Custom, made women soldiers. Gezo, who largely made use of the Amazons to keep his own subjects in check and to promote military rivalry, increased and improved the force. He directed every head of a family to send his daughters to Agbomi for inspection; the most suitable were enlisted, and the corps thus placed on a new footing. This course was also followed by Gelele, his successor, who had every girl brought to him before marriage, and enrolled those who pleased him.

Ellis reported nothing and Burton next to nothing about the sexuality of these "Amazons." They were distinguished from the king’s (also numerous) wives (p. 64). "Two-thirds are said to be maidens" (p. 68). In his "Terminal Essay" to his translation of the Arabian Nights, Burton (1930[1886]:88) wrote "At Agbome, capital of Dahome, I found that a troop of women was kept for the use of the ‘Amazons’" referring to p. 73 of the second volume of Burton (1864), but there is nothing there about women reserved for the Amazon troops. There is only an expression of his belief that passion for combat excites rather than substitutes for love: " All the passions are sisters. I believe that bloodshed causes these women to remember, not to forget LOVE."

Commander Frederick Forbes’s journals of his 1849-50 missions to King Gezo of Dahomey (published in 1851) does not deal with sexual behavior of the "Amazon" troops, but is even clearer than Burton about (hyper-)masculine gender identification:

The amazons are not supposed to marry, and, by their own statement, they have changed their sex. "We are men," say they, "not women." All dress alike, diet alike, and male and female emulate each other: what the males do, the amazons will endeavour to surpass. They all take great care of their arms, polish the barrels, and, except when on duty, keep them in covers. There is no duty at the palace except when the king is in public, then a guard of amazons protect the royal person, and, on review, he is guarded by the males.... The amazons are in barracks within the palace enclosure, and under the care of eunuchs and the camboodee or treasurer. d (I:23-4).

Indeed, in a 13 July 1850 parade, amazon troops sang about the effeminacy of male soldiers they had defeated:

We marched against Attahpahms as against men,

We came and found them women.

What we catch in the bush we never divide (II:108).

After a parade of 2400 amazons pledging to conquer Abeahkeutah (a British ally in Sierra Leone) or to die trying, an amazon chief began her speech by asserting gender transformation: "As the blacksmith takes an iron bar and by fire changes its fashion, so we have changed our nature. We are no longer women, we are men" (II:119).

Central Africa

Hulstaert (1938 :95-6) wrote the following about relationships between Nkundó women (in what was then the Belgian Congo, and now is Zaire):

Nkundó girls play at ‘husband and wife’ and even adult married women engage in this vice. According to my informants, the causes are as follows: first, an intense and very intimate love between two women, second and above all, the fact that wives of polygamists find it difficult to satisfy their passions in a natural way. Often they engage in this practice with co-wives of the same man."

He further noted that "in establishments where girls are too securely kept away from the opposite sex, there has been an increase" in sexual relationships between girls (p. 96). "The latter often engage in sex with co-wives. A woman who presses against another woman is called yaikya bonsángo."

La Fontaine (1959:34, 60-1) mentioned a Gisu (Uganda) woman living as a man, but did not specify the sex of this person’s sexual partners (if there were any).

Southern Africa

Hambly’s (1937:426; 1934a: 181) main Ovimbundu (a tribe in Angola) informant, Ngonga, told him, "There are men who want men, and women who want women.... A woman has been known to make an artificial penis for use with another woman." Such practices did not meet with approval, but neither did transvestitic homosexuals of either sex desist (cf. Estermann 1976:197). Indeed, there are early attestations including Andrew Battle, c. 1589, that in Dombe, "They are beastly in their living, for they have men in women’s apparrell, whom they keep among their wives", and a 1606 Jesuit who found "Chibadi, which are Men attyred like Women, and behave themselves womanly, ashamed to be called men; are also married to men and esteeme that unnatural damnation an honor" (Purchas 1625:973,1558).

Further South, Kidd (1904:209) was not specific about what "indecencies", "degradations" and "obscenities" Bantu female initiation involved that, he claimed, demoralized their womanhood forever. Falk (1925:209-10) noted that homosexuality was common, especially among Hottenot (Nama) young married women. And to the east, Schapera (1938: 278) reported that among the Tswana (in addition to homosexuality among the men laboring in the mines together), back home, "lesbian practices are apparently fairly common among the older girls and young women, without being regarded in any way reprehensible." Colson (1958: 139-40) mentioned a possible man-woman and a possible woman-man among the Tsonga of what is now Zimbabwe.

From her research on Nyakyusa age-villages, Monica Wilson (1951:88) reported that "lesbian practices are said to exist, but we have no certain evidenceof this." She went on to speculate that they were "much more likely to be among the older wives of chiefs and other polygynists than among the girls, who have so much attention from young men." This pattern is her logic, not a Nyakyusa report. She noted that "a case was also quoted of a doctor in Tukuyu who ‘is a woman; she has borne children, now her body has grown the sexual organs of a man and her feelings have changed also; but she keeps it very secret, she is spoken of as a woman.’" (p.197), but did not note to what, if any, use the new male organs were put.

Judith Gay (1985) described what she considered a relatively new (i.e., 1950s onward) institutionalized friendship relationship among women left behind in Lesotho by migrant men working in South Africa:

Young girls in the modern schools develop close relationships, called “mummy-baby” with slightly older girls. Sexual intimacy is an important part of these relationships. Mummy-baby relationships not only provide emotional support prior to marriage, but also a network of support for married and unmarried women in new towns or schools, either replacing or accompanying heterosexual bonds (p. 97)

“Relationships are always initiated voluntarily by one girl who takes a liking to another and simply asks her to be her mummy or her baby, depending on their relative age, “Gay 1985:102) reported. “The most frequently given reason for initiating a particular relationship was that one girl felt attracted to the other by her looks, her clothes, or her actions.”

Although eager to deny that these relations are in any way marriage-like, Gay also wanted to interpret them as preparations for “the dynamics of heterosexual relations”—even”explicit opportunities for initiation into heterosexual relations” (p. 109)—or as substitutes for relations in the absence of men who are working in South African mines. She also provides some hints about marriage resistance (at least delay) and contrasted the autonomy of “mummies” with the constraints on Sesotho wives. She also notds that “in many cases the relations established with other girls are transformed, but do not cease altogether” (pp. 107, 109, 110).

A “mummy” may have more than one “baby,” but Gay claimed that the “baby” can only have one “mummy,” although one female may be a “baby” to one female and a “mummy” to another (p. 108). However, all three of the women who provided her “case studies” had had, over time, more than one “mummy” (nine, six, and two).

Although “Sesotho initiation for girls is no longer practiced in most lowland villages, where about half the nation’s population lives,” (p. 99), girls continue to lengthen their own or each other’s labia minora. They believe that having done this or had it done will later enhance sexual pleasure: “The process of lengthening is done alone or in small groups, but is not directly tied to initiation [now]]. The process is said to heighten mocheso (heat) and appears to provide opportunities for auto-eroticims and mutual stimulation between girls” (p. 101). Without alluding to any evidence (or any kind of inquiry), Gay (1985:112) assertts that “the contacts which may be involved in lengthening the labia minora are apparently not regarded as emotionally significant, whereas falling in love with a girl and simply caressing her is” (p.112).

Mueller (1977) also described similar relationships in two Lesotho villages. Blacking (1959) reported fictive kinship relationships among Venda schoolgirls of the Transvaal earlier, and Zulu schoolgirls in 1978. I consider this strong evidence that “age-stratified” is a type of female, not just male, homosexuality. Age differences need not be great, even though the idiom is trans-generational. In the eleven instances in which Gay (1985:114-5) specified the age of both, the mean age differences was 4.8 years, the median age difference was five years (the range was from one to twelve; those playing the “baby” role ranged in age from 8 to 24, “mummies” from 15 to 35).

Deborah Green, an African-American medical student who formed a relationship with a white woman in the Shangaa homeland town of Gazankulu in the northern Transvaal, told Hartman (1992:16).

There are some stories from Venda (a nearby tribe) about lesbian-type people — I mean, women that lived together and raised families together. But the Shangaan had no concept of a lesbian relationship, no preconception about it. They just know that I came along and started living with Tessa, and then Tessa was much happier. So they thought it was a good thing.

Whether the reaction to a Shangaan woman becoming involved with either a black or white woman cannot be extrapolated from this, however, even if the optimistic assessment of what others think about Green’s relationship is accurate.

Among Cape Bantu, “forms of overt homosexual behavior between women are described by female isanuses [chief diviners].” Moreover, homosexuality is widely ascribed to women who are in the process of becoming isanuses (Laubscher 1937: 31). Use of artificial penises was attested among the Ila and Naman tribes of South Africa by Smith and Dale (1934:181) and Schapera 1930:243). Some adult Nandi women (in Kenya) “satisfied each other alternately” using wooden dildoes, according to Bryk (1939:149, 1964:227).

Although I would not wish to deny that there are women-identified-women in “traditional” black Africa, nor that age-stratified relationships may be maternal, the fragmentary data available on homosexuality among women in sub-Saharan Africa suggests that the culturally-marked category is a woman who undertakes a sort of husband role and claims some of the prerogatives of men in these cultures. The major idiom of homosexuality for both men and women is gender. Where there are age disparities, the younger is generally cast in the feminine role. This is true for “boy-wives” and “wives of the mine,” for Lesotho “babies” and the wives of women from Benin to Kenya.

Audre Lorde (1984:50) is surely right that sexual contact “has existed for ages in most of the female compounds across the African continent.” Not talked about by or to men, female-female sexuality has been little recorded in the “traditional” African countryside (in which women were illiterate even where men were not) or in “modern” (post-independence) African cities, although there is a growing body of retrospective accounts by lesbians about growing up invisible to themselves but not to others in South Africa (e.g., Chan Sam 1993, 1995; Mamaki 1993; Radcliffe 1993; Vimbela 1995).

sexual education in the three-year-long seclusion of young women following menarche in the Ulanga district of what is now mainland Tanzania (Geigy and Höltker 1951).

The Greeks named continent Libya and called anyone black Aithiops. Roman province known as Africa and used the same word for the tertia orbia terrarum pars, the third continent

From Herodotus onward, the West’s self representations have always included images of peoples situated outside of its cultural and imaginary frontiers to contrast with thinking subjects. xi

Afer didn’t indicate color in Latin

“Aethiops, the proper name of Vulcan’s son in Greek mythology, is the generic qualification of any dark-skinned person.” 26 heat(calore) color (colore) of people nearer the sun

Aethiopia beyond Egypt, beyond the end of the world

“The declining use of Aethiopia as the name of the continent begins with the European explorations of the 15th century” 27

“The Latin niger corresponds to the Greek melas and, insofar as the color of human beings might be concerned, it strictly translates the Greek Aithops, that is , a face burned by the sun” 27 black c. 1566

Phoenicians working for Pharoh circles in 6th c. B.C; Carthaginian explorations later that century 17-8

(Amazons cast beyond Scythians who were edge of world)

yet Amazons could learn men’s language, men couldn’t learn theirs 87

Diodoros Siculus in 1st c located in Africa

nazism natural conclusion of colonialism

Mudimbe, V. Y. (1994) The Idea of Africa. Bloomington: Indiana University Press.