In my advanced undergraduate course on colonial Ghana, I set students the following essay question: “Did cash-cropping and colonialism conspire to reduce the autonomy and status of Akan women?” In order to answer this question, students must develop a working understanding of: 1. the spread of cash-cropping in the Gold Coast and Asante, and the particular inputs of land and labor that are required for the production of cocoa; 2. Akan social and political organisation, including the concepts of matrilinearity and rights-in-persons, and the role of stools and lineages in determining access to usufruct on different types of land; and 3. the nature of indirect rule and the potential for male chiefs and elders to increase the control of senior men over women and junior men. Those students who can see the connections between this essay and other segments of the course (particularly those which address missionary Christianity and formal schooling), tend to produce more sophisticated answers. Across the range of marks, however, students almost invariably conclude that the answer to the question is “yes,” and this is consistent with the position of most of the authors who feature on my reading list (e.g. Jean Allman, Victoria Tashjian, Gwendolyn Mikell, Beverly Grier, Takyiwaa Manuh, and, to some extent, Gareth Austin).

In this article, however, Stefano Boni argues that the thesis of “retrograde steps” for Akan women in the twentieth century “rests on an assumed, unproven past,”^{Footnote 1} and that continuity is more important than change in understanding gender relations. Using evidence from colonial ethnography and from courts in Sefwi, Boni counters two core elements of the wider “retrograde steps” thesis: firstly, that the onset of cocoa cash-cropping resulted in a much more intensive exploitation of women’s labor and therefore narrowed their scope for independent economic activity; and secondly that fathers gradually asserted greater rights over their children at the expense of mothers and the matrilineage.

On the first point, Boni argues that marriage had long entailed the right of the husband to benefit from the labor of his wife, and that wives (or their matrikin) had long recognised “fluid” unions, in which they refused to accept monetary payment from the husband, as a means of mitigating his demands and expectations. Boni concludes that male efforts to assert their rights over their wives’ labor predated the impact of cocoa cash-cropping in Sefwi, and that evidence from elsewhere in the Akan region points to a similar pattern: “[M]en used their wives’ unpaid labor, alongside unfree labor, in precolonial profit-making activities (…) . Husbands’ demands for wives’ work on cocoa farms was an exploitative adaptation of existing marital norms.”^{Footnote 2}

Secondly, although the matrilineal “model” would have rights over children residing in their mother’s brother, Boni points to examples from early twentieth-century Sefwi in which biological fathers made successful claims to the custody and labor of their children and played a role in negotiating the marriages of their daughters. He concludes that fathers’ rights were already well-established before the effect of either cocoa or colonial rule was fully felt, and that “Rights over youngsters were rather negotiated dynamically between different agents, the father certainly being one of them. Whoever catered for the child (…) was often recognized as having privileged rights over the youngster.”^{Footnote 3} Men were therefore “willing, at least officially, to comply with their paternal duties”^{Footnote 4} because this was a key justification for their subordination of women and children. Conflict lay not in the ideal of household head as provider, but in whether the degree to which a man fulfilled his duties to his dependants justified the extent of his demands upon them.

I have chosen this article not because I think that Boni is “right,” for there are specific elements of the “retrograde steps” thesis (particularly the impact of migration and formal schooling) that are not fully addressed in this article, probably due to limitations of space. I have chosen the article because it raises for students key questions about access to and interpretation of evidence, and points to two major challenges in the field of African history: periodization and inter-disciplinarity.

Whereas inter-disciplinarity is usually heralded as one of the most exciting features of African history, Boni suggests that the overwhelming support for the “retrograde steps” hypothesis is an unwelcome consequence of the close relationship between gender history and the social sciences. Scholars of Ghana have been influenced by an older literature which theorized gender inequality as the product of specific historical formations, and therefore denied the ubiquity and inevitability of female subordination by seeking evidence of more complementary and reciprocal relations in earlier historical periods. Indeed, the term “retrograde steps” was coined by social scientists and not by historians.^{Footnote 5} Boni suggests that the influence of the social sciences has diminished the quality of specifically historical research by producing “the expectation towards change and degradation,” and by leading historians “to ‘read’ written and oral sources with the primary aim of searching a confirmation of such statements.”^{Footnote 6}

In order write better history, he argues, the gender historian must overcome the great challenge of periodization: rather than identifying the subordination of women at the turn of the twenty-first century and seeking out the twentieth-century determinants, we should rather concentrate on developing our understanding of the more distant past in order to give adequate consideration to underlying continuities over time. Fundamentally, Boni suggests that female subordination is not the outcome of specific historical formations (in this case, the transformation of social and labor relations with the entry en masse of rural producers to the world capitalist economy) but that, over the longue durée, it is maintained primarily by ideology. Thus, if Akan women were more oppressed by the end of the twentieth century than they had been in the beginning, this was the result of the collision of new forms of production and exchange with a much older ideological framework which justified hierarchical and exploitative relations between the sexes and within households. Here, Boni anticipates more recent statements, such as that by Richard Reid,^{Footnote 7} on the critical need for sustained research on pre-colonial Africa, and research of a kind that allows us to access not only events, but also ideology and self-understanding.^{Footnote 8}

Such arguments about the state of the field certainly impact upon my students, not least in determining the types of courses that faculty design for them and the scholarly literature that is available for them to read. Of more immediate relevance for their essays in my course, however, is the question of how historians can access and evaluate evidence of change or continuity over time. Boni makes a valid point about the need for gender historians to research across the great colonial/pre-colonial divide, and to exercise scepticism in the use of colonial-era ethnography as a source for pre-colonial history. He gives surprisingly little credit to those scholars who have tried to achieve this through a substantial investment in oral history.

The court records on which Boni relies in this article provide snapshots into the lives of individuals – although of course if one has enough snapshots spread over a long enough period of time, one can legitimately seek out wider patterns. But Boni does not explicitly concede here that life history interviews provide opportunities to track change over an individual’s lifetime, and to recover from elderly women the experiences of their mothers and grandmothers. Boni’s court cases exemplify very well why gender historians must take account of the ways in which age, reproductive status and acquisition of resources interact with sex in the constitution of relations of power within the household and family. In Boni’s examples, a woman who is about to become a wife shifts “from being a dependant of the household of the father/mother/mother’s brother to the one of the husband.”^{Footnote 9} For him, the important point is that the woman in question is still subordinate and dependant. But was the same woman equally subordinate and dependant a quarter of a century later? Did her more advanced age, her experience of motherhood, or the resources accumulated and investments made over half a lifetime have any impact on the nature of her subordination or the forms in which it might be perpetuated in a future generation? Boni’s own analysis would suggest that it did, for if maintaining a child allowed for greater claims to be made upon his/her future labor, or for greater influence to be exercised over his/her marriage contract, the woman who was able to fill the gaps in a husband’s provision was in a better position from the one who was not, and, if the former woman had a daughter (or indeed a son), her/his life chances might look rather different from that of the child of the same man by a less able wife.

The life history interview, then, allows gender historians to peer across lifetimes and generations, and to understand the experiences of different females within the same lineage or household. Students who grasp this are better placed to problematize the category “woman” in their essays on my course, whilst those who are thinking ahead to dissertation research might make good use of the transcripts that Allman and Tashjian have deposited in libraries at the University of Ghana and at Northwestern University.