Gender

The claim that will organise this file is Oyèrónkẹ́ Oyěwùmí's: that gender was not a primary organising category in Yoruba social thought before the colonial encounter, that seniority did the work that gender does in European social organisation, and that the category "woman" as it is used about Yoruba society is largely an import that arrived with colonial administration, missionary translation and Western scholarship. This is one of the most influential arguments in African gender studies and one of the most contested. It has been criticised in detail by other African feminist scholars, most sharply by Bibi Bakare-Yusuf, and the criticisms are serious.

This file sets out the argument, then the criticisms, then the documentary record of what Yoruba women actually held and did, which is substantial and which neither side of the theoretical dispute denies. Confidence is marked contested because the central interpretive question is genuinely open, not because the underlying facts are unknown.

Oyěwùmí's argument

The Invention of Women: Making an African Sense of Western Gender Discourses (University of Minnesota Press, 1997) argues that gender is not a universal organising principle and that the Ọ̀yọ́-Yoruba case shows a society that did not use it as one. The argument has several linked parts.

The critique of the visual. Oyěwùmí attributes the biologising of social difference in European thought to the primacy of vision in that intellectual tradition. Privileging the visual makes appearance and visible markers into the basis of categorisation, so that the Western episteme constructs its categories and hierarchies as binary and visible: male and female, white and black. The physical body is therefore always linked to the social body. African cultures, on her account, have not historically been ordered by vision but through other senses, which is why she proposes "world sense" rather than "worldview" as the appropriate term for an African way of knowing 1.

The linguistic claim. Yoruba does not mark gender grammatically, while it does mark seniority. There is no gendered third-person pronoun. Names are not gender-specific. There are no gender-specific words for son, daughter, brother or sister. The words ọkọ and aya, conventionally translated husband and wife, are not gender-specific in the way the English terms are. What is marked is relative seniority, in the third-person forms and throughout the address system 1.

Her account of ọkùnrin and obìnrin is central and worth quoting as Bakare-Yusuf reproduces it:

the word obinrin, erroneously translated as 'female/woman', [...] does not derive etymologically from okunrin, as 'wo-man' does from 'man'. Rin, the common suffix of okunrin and obinrin, suggests a common humanity; the prefixes obin and okun specify which variety of anatomy. There is no conception here of an original human type against which the other variety had to be measured. Eniyan is the non-gender-specific word for humans 2.

The force of this is precise. In English, "woman" is morphologically derived from "man", so the female is linguistically marked as a variant of a male norm. In Yoruba, ọkùnrin and obìnrin are parallel constructions on a shared element, and the unmarked category for a human being is ènìyàn, which is neutral. So Yoruba does not encode a male norm with a female deviation. Oyěwùmí's conclusion is that obìnrin is therefore not symbolically or normatively inferior to ọkùnrin, and that the "woman" theorised in Western feminist discourse in terms of negation and limitation has no Yoruba equivalent 2. Because there is no gendered noun of this kind, she uses "anamale" and "anafemale" to name anatomical categories without importing the social freight of "man" and "woman".

The institutional claim. Yoruba social institutions did not divide by anatomy. The compound, the agbo ilé, was the site of legitimacy, authority and power, and its ordering principle was the insider-outsider relation. Every member of the lineage, anamale or anafemale, is both ọmọ ilé and ọkọ, and stands as senior to anyone marrying in. Oyěwùmí's own formulation, which is the most-quoted passage of the book:

Although ana-females who joined the lineage as aya were at a disadvantage, other ana-females who were members of the lineage by birth suffered no such disadvantage. It would be incorrect to say, then, that anatomic females within the lineage were subordinate because they were anatomic females. Only the in-marrying aya were seen as outsiders, and they were subordinate to oko as insiders 2.

She extends this to the division of labour, arguing that the assumption in West African studies that men farm and women trade is an imposition of an alien model that distorts reality, since both ọkùnrin and obìnrin were represented in trade and farming, and an anafemale could be aya in one lineage and ọkọ in another as well as warrior, diviner, hunter and farmer 2.

The colonial claim. If the category was not indigenous, it had to arrive. Oyěwùmí's account is that British colonial rule, missionary translation and subsequent scholarship imposed a gender binary on a society that had organised by seniority, lineage and community, and that ọkùnrin and obìnrin acquired their symbolic gender-based layers of meaning through the colonial project. Colonial administration then compounded the imposition institutionally by recognising only male authority, which converted a theoretical distortion into a material one.

That last point is not contested. It is corroborated by the general historical record of indirect rule across Yorubaland, which dealt with male chiefs and dismantled the parallel women's authority structures. That is treated in History: Christianity, Islam and Colonial Conquest.

The criticisms

Bakare-Yusuf's "'Yorubas Don't Do Gender'" (2004) is the most detailed critique and it grants a good deal before attacking. She agrees that gender should not be universalised as a first-order principle across all societies and times, that uncritically adopted foreign categories distort local dynamics, and that the questions a researcher brings determine the answers found: if a gendered question is posed in a society where seniority is dominant, a gendered response will nonetheless result 3. Her objections are to method, not to the general project of decolonising the categories.

Language does not equal social reality. This is her central charge. Oyěwùmí, she argues, "completely flattens the relationship between language and reality by assuming that there is total isomorphism between the two," and fails to distinguish de jure from de facto description, that is, between what happens at the level of language, discourse and juridical norm and what happens in everyday lived experience 4. Her illustration is post-apartheid South Africa, where nothing at the discursive or juridical level limits a black South African under one of the world's most progressive constitutions, and yet real inhibiting factors operate at the level of lived experience 4. The absence of gender marking in Yoruba grammar therefore does not establish the absence of gender inequality in Yoruba life.

The etymological method is unsound. Oyěwùmí's argument requires that ọkùnrin and obìnrin had an original, pre-colonial meaning recoverable beneath colonial mistranslation. Bakare-Yusuf's reply is that we cannot guarantee that a word carried a given meaning at a given point in history, that this is especially difficult in an oral culture without a dictionary history, and that the method assumes a culture remained pure across time without discontinuities 5. She compares the move to Heidegger's etymological recovery of "original" German and Greek meanings in service of a theory of German historical destiny, which is a pointed and deliberate comparison 5. Against it she offers Wittgenstein's rope: words consist of a multiplicity of interrelated usages, and meanings are mediated through different practices, discourses and institutions that invest concepts with multiple layers, so that "women" already includes multiple and contradictory meanings and is already open to shifts 5.

Power is not monolithic. Even granting that seniority is the dominant language of power, Bakare-Yusuf argues it cannot be the only one, and that no mode of power, gender, seniority, race or class, has the same value from context to context or exists in isolation from the others 6. Because Oyěwùmí wants seniority to stand alone, she cannot see how the ideology of seniority masks other power relations, or how its vocabulary is used to couch sexual abuse, familial violence (especially against the aya in a lineage) and symbolic violence. Bakare-Yusuf's verdict on this point is unusually blunt: "her theory of seniority becomes disturbingly naive and politically dangerous" 6 Her example is the abuse of seniority in the Nigerian student-teacher relationship, which goes unchallenged because victims are unwilling to be seen disrespecting a senior, which she reads as another form of power-over disguised as respect for the elder 6.

The statistical objection. This is the most concrete. In a patrifocal society, Bakare-Yusuf notes, it is anafemales as aya, not anamales, who have to move out to another lineage and become subordinate outsiders in the spouse's lineage; insofar as heterosexual coupling and motherhood are privileged, it is anafemales who marry out; and it is women who have primary responsibility for food preparation and childcare, giving them the dual burden of nurturer and provider 7. Oyěwùmí meets this by questioning the value of statistics and by pointing to counter-examples, and Bakare-Yusuf holds neither response adequate: that probabilities can be calculated at all indicates a normative structure at work, and denying the validity of statistics is a weak way of denying standardisation itself. Powerful women monarchs and hunters can be found in even the most unreconstructed patriarchal cultures, and such counter-examples are "the 'miraculous exception' within what is in reality a hegemonic framework," after Bourdieu 7. Her conclusion on this point is that the experience of princesses and privileged women cannot be used to evaluate the position of all women or to deny collective subordination, and should instead be used to imagine new possibilities and fight inequality 7.

The authenticity objection. Oyěwùmí's focus specifically on the Ọ̀yọ́-Yoruba, as against every other Yoruba subgroup, is read by Bakare-Yusuf as evidence that the account represses internal difference in favour of an ideologically driven authenticity, and she invokes Nancy Fraser on identity politics discouraging intra-group scrutiny of patriarchal strands as inauthentic 8. Her own alternative is a "polytheistic" reading of Yoruba culture, by which she means a structural disposition toward plurality, negotiation and accommodation rather than a single ordering principle, and she cites Hountondji's argument that pluralism is inherent in every society rather than arriving from outside, so that the decisive encounter is not Africa against Europe but "the continuing encounter between Africa and itself" 8

Where else the criticism comes from. Oyěwùmí's argument has also drawn objections on the ground that the linguistic evidence is overstated (Yoruba does mark sex lexically where it matters, in ọkùnrin and obìnrin themselves and in numerous compounds), that Ọ̀yọ́ is not representative of Yoruba societies generally, and that the pre-colonial baseline she reconstructs is itself reconstructed from sources shaped by the colonial period she is trying to see behind. Ifi Amadiume's Male Daughters, Female Husbands (1987) on the Nnobi Igbo is often read alongside Oyěwùmí as the other major statement of flexible African gender, and Amadiume's own position, that gender categories existed but were not rigidly tied to biological sex, is a materially different claim from Oyěwùmí's, that gender was not a social category at all.

Where the argument has settled, so far as it has

The most useful current position is one that takes the strongest part of Oyěwùmí's case and drops the strongest form of the claim. Insa Nolte's work on eighteenth-century Yoruba marriage and belonging arrives at a formulation worth quoting: reflecting the importance of patrilineal kinship as a template for social and political organisation, marriage associated husbandly roles with power and belonging by birth and wifely positions with support and origin from the outside, so that while women were excluded from some forms of authority, "a primarily relational understanding of gender meant that husbandly roles were frequently open to women" 9

That is the reconciliation available on the evidence. Gender in Yoruba society was relational rather than fixed to anatomy: it named a position in a set of relations, and a person of either anatomy could occupy the position of ọkọ. This is precisely why ọkọ cannot be translated as "husband". But the positions were not randomly distributed. Because the system was patrilineal and virilocal, it was systematically women who moved out and entered as outsiders, and systematically men who stayed and remained insiders. Both things are true, and Oyěwùmí is right about the first and understates the second while Bakare-Yusuf holds them together.

A reader coming to this fresh should also note what neither scholar disputes: that colonial administration and missionary institutions imposed a gender ideology far more rigid than anything indigenous, and that they did substantial and documented damage to women's institutional authority. The theoretical dispute is about the precolonial baseline, not about the colonial effect.

What Yoruba women actually held

The documentary record here is strong and largely uncontroversial. It stands independently of the theoretical argument above.

Ìyálóde. The Ìyálóde is a chieftaincy title held by a woman, rendered as mother in charge of external or public affairs. The holder represented women in the town's political process, sat in the council, had oversight of women's trade guilds and the marketplaces, adjudicated disputes among traders, set market rules and carried women's interests to the ọba and chiefs 10. Bọ́lańlé Awé's study of the office in the traditional Yoruba political system is the standard scholarly treatment 11. The office was elected on merit and standing rather than inherited, which distinguishes it sharply from most hereditary titles.

Its holders were not figureheads. Efúnṣetán Aníwúrà, the second Ìyálóde of Ìbàdàn from 1867 to 1874, built one of the largest fortunes in nineteenth-century Yorubaland through long-distance trade and was among the pre-eminent slave traders of the city, with a substantial armed following of her own 10. That last fact needs stating plainly rather than being edited out: women of standing participated in the slave trade as principals, and the record of women's authority is not a record of moral superiority.

Madam Tinúbú, exiled from Lagos by the British in 1856, relocated to Abẹ́òkúta, supplied arms during the war with Dahomey and was made Ìyálóde of all Ẹ̀gbáland 10. Her career is the clearest single case of a woman exercising military-political power at the level of a state.

Ìyálọ́jà. Mother of the market: authority over a particular market and its traders, with jurisdiction over the allocation of stalls, the settlement of trading disputes and the regulation of the market's internal affairs. The office survives, and the Ìyálọ́jà General of Lagos remains a substantial political position in contemporary Nigeria.

Market power as political power. The connection is not incidental. Because women dominated market trade and because markets were the circulatory system of the economy, an organised body of market women could halt the economy of a town. The most famous demonstration is the Abẹ́òkúta women's revolt of 1946 to 1949, led by Fúnmiláyọ̀ Ransome-Kuti, which forced the abdication of the Aláké and the abolition of the tax on women, and which is treated in History: The Twentieth Century.

Female rulers and regents. Ọ̀rọ̀mpọ̀tọ̀ is recorded in the Ọ̀yọ́ king list as an Aláàfin who was anatomically female, the sister of her predecessor Ẹgúngúnọ̀jọ̀ 12. The traditions surrounding her, including the account that she concealed or transcended her anatomy in order to reign, are exactly the kind of material that both sides of the theoretical argument read differently, and the historicity of the details is uncertain. Regencies held by women, and female occupants of specific stools in various Yoruba towns, are documented across the region. What is clear is that female occupancy of the highest offices was rare rather than normal, which supports Bakare-Yusuf's "miraculous exception" point, while the fact that it was possible at all supports Oyěwùmí's point that anatomy was not an absolute bar.

Palace and title offices. The Erelú, the Ìyá Ọba (king's mother, an office rather than a biological relation in some towns), the Ìyá Kere and other palace titles gave women formal positions inside the machinery of kingship. Onadeko records that the Ògbóni executive body of six, the Ìwàrèfà, included the Ìyá Abiye among the Olúwo, Lísa, Aro, Ọdọ́fin and Apènà 13. A woman sitting on the six-member executive of the judicial and constitutional check on the ọba is a concrete institutional fact and not a symbolic gesture.

Òrìṣà priesthood. Women hold priesthood at every level of òrìṣà practice, including headship of major cults, and in several traditions the priesthood of specific òrìṣà is held predominantly or exclusively by women. This is treated in section 06.

Àwọn ìyá wa. "Our mothers", also Ìyá mi and the ẹlẹ́yẹ, owners of birds, names a collective spiritual power attributed to elder women and understood as capable of both protection and destruction 14. It is treated at length in the òrìṣà and cosmology sections because it is a religious category, but its social significance belongs here: the acknowledged existence of an autonomous female spiritual power that men must propitiate rather than command is a structural counterweight, and the Gèlèdé masquerade, held to honour and placate the mothers, is its public form. The category is a double-edged one. The same conception that makes elder women powerful makes them the ones accused of àjẹ́, and the witchcraft accusation was, as noted above, the specific charge that could take a woman out of family jurisdiction and into the Ọba's court 15.

How to hold all this

Three propositions can be held at once without contradiction, and a reader is better served by holding them than by picking a side.

Yoruba social organisation ran on seniority and lineage insiderness to a degree that has no European parallel, and translating its terms into English gender vocabulary genuinely destroys information. That is Oyěwùmí's real contribution and it stands.

Anatomy nevertheless correlated systematically with social position, because a patrilineal virilocal system moves women out and keeps men in, and no amount of terminological care makes that asymmetry disappear. That is Bakare-Yusuf's real contribution and it also stands.

And the colonial period did impose a harder gender order than the one it found, dismantled the women's side of a dual-sex political structure, and produced much of the scholarship through which the precolonial period is now read. That is documented history and is not in dispute between them.