applies to the written and visual ethnographic oeuvre of the native commissioner cum self-styled ethnographic authority C. H.L. (‘Cocky’) Hahn.[20]

The colonial era ethnographies are typified by their stasis in space and time. Most published sources only depict the eastern Owambo communities of Ouk – wanyama and Ondonga, in spite of their authors’ claims that they had for their subject ‘the Owambo’ in general (e. g. Hahn 1928). There is not sufficient space here to deconstruct the representation of a uniform Owambo culture, although this has a definite bearing on current representations of initiation.[21] Contrary to the assumption of cultural uniformity in the colonial ethnography of Owambo, women’s initiation rites were and are subject to enormous variation in detail and duration. There is thus a need to cautiously depict what happened in different places and at different times as alterations in ceremonial style and practice affect songs and costumes, ritual roles and ritual events, and, in effect, cultural meaning and understanding. None of this is reflected in the ethnographic representations that, therefore, need to be read with caution.

Owambo women’s initiation was, first of all, a rite of social maturation. It pub­licly transformed the initiates, denominated ‘brides’ (aafuko in oshiNdonga, ovafu – ko in oshiKwanyama), into adults, completing them with the essential attributes of their new status. In the past these attributes included the attire and hairstyle of an adult woman. What is most important, after having passed through the ceremony a young woman could give birth legitimately, because she was regarded as a married woman (oshiNonga: omukulukadhi; oshiKwanyama: omukadi) even if she did not have a husband immediately thereafter. Ethnographic sources stress that a young woman who fell pregnant before she had undergone the initiation, and was thus regarded as a ‘girl’ (omukadhond), was burnt to death in most precolonial Owambo communities (Louw 1967:31,42 on Ongandjera; Tuupainen 1970:47 on Ondonga; Loeb 1962:240 on Oukwanyama). The efundula, ohango or olufuko ceremonies were thus the cornerstones of legitimising female sexuality and reproduction.

Подпись: 40

It appears that, secondly, the rites were meant to enhance fertility (Williams 1991:110; Tuupainen 1970:50—51). This aspect was especially highlighted by the Powell-Cottons, as Gwyneth Davies has emphasised in her thesis based on the sisters’ oeuvre (Davies 1987:102—103). Davies further argues that the efundula pro­vided a context for teaching and instruction. The rites and stages of the initiation expressed a reiteration of social knowledge on appropriate gender roles. Through participating in the efundula, the initiates experienced the society’s views towards such knowledge (Davies 1987:85—86).

Подпись: Arnfred Page 41 Wednesday, March 3, 2004 2:38 PM
Efundula: Women’s Initiation, Gender and Sexual Identities in Northern Namibia

While most authors described the rites as initiation ceremonies, which precede the actual marriage (cf. Tuupainen 1970:53), the Kwanyama efundula has also been dubbed a ‘group-marriage ceremony’ (Loeb 1962:34). Nubility, if not the actual nuptials, is central to the meaning Owambo attach to it today, when commonly translating efundula as ‘traditional wedding’. It is possible that the meaning of the ceremony encompassed the notion of marriage in Oukwanyama, where no elab­orate marriage ceremonies took place after the efundula, and the young wife rather unceremoniously went to live with her husband soon after the ceremony (Tonjes 1911:143; Tuupainen 1970:57). Certain parts of the Kwanyama efundula, moreo­ver, involved the initiate’s future husband, who ritually ‘proposed’ during the cer­emony, which has been described as ’symbolic engagement’ (Bruwer 1959:119). On the other hand, the ohango in Ondonga was possibly not understood as ‘mar­riage’ proper, as the initiation was followed by specific individual marriage cere­monies conducted afterwards (Hahn 1928:32; Tuupainen 1970:56—57). It thus appears that the meanings attached to women’s initiation may have varied over time and space.

Different forms of political and ritual power, and their gendered embodiment, require a closer examination. It appears that generally the kings decided the date of commencement of a cycle of initiation rites. The actual impact of the ruler re­mains unclear, however; Native Commissioner Hahn greatly stressed the role of the king (Hahn 1928:29); he may have overemphasised this for reasons of his own, however, as I discuss below (see: Lining up bare-breasted maidens). Missionaries usually accorded the kings a far less prominent part in the ceremony (cf. Tonjes 1911:135—36; Estermann 1976:70—72). The rites were directed by a ritual leader (namunganga in oshiNdonga, omupitifi in oshiKwanyama). Most of the older eth­nographies depict the ritual leadership as a male prerogative. Hahn (1928:29) and Tuupainen (1970:46) noted that in Ondonga the ceremony involved one male and one female ritual leader; however, they described the male namunganga as the ‘master of ceremonies’, whereas the female’s part was depicted as that of an as­sistant. For Oukwanyama, Tonjes (1911:139), Bruwer (1959:118) and Loeb (1962:246) all emphasised that ovapitifi were invariably old, circumcised men. Since male circumcision in Oukwanyama as with most other Owambo polities had ceased during the 19th century, ovapitifi were invited from those areas where male circumcision was still practised, and by the early twentieth century those were all in Angola. In contrast to the hegemonic depiction of exclusively male rit­ual leadership, Gwyneth Davies, drawing on the Powell-Cottons’ 1930s research, noted that a woman might also direct an efundula if she was the daughter of a cir­cumcised man (Davies 1987:109).

The ethnographic emphasis on male ritual leadership is indeed curious. Present-day informants in Oukwanyama have pressed home the point that the position of omupitifi, often combined with that of a healer (onganga) has been passed on ‘for generations’, as some stressed, from either a mother or a father to either a son or daughter. Gender appears to be largely irrelevant, if this is indeed the case. Instead, the selection of a ritual leader seems to have been based on cer-

Arnfred Page 42 Wednesday, March 3, 2004 2:38 PM