Feminist Knowledge | Review Essay

October 1995, Revised April 1996 (Originally published by CODESRIA as one of their Green Books)

Women's Studies and Studies of Women in Africa during the 1990's

By

Amina Mama,

Contents

I Introduction

II State and Politics

i) Precolonial States i) Colonial States iii) National Liberation v) Post-independence states

III Culture

i) Religion ii) Sexuality iii) Herstories

IV Work and Economy

i) conceptual debates: domestic labour ii) historical studies:precolonial colonial labour iii) contemporary: formal sector
informal sector

V Conclusions: Whither African Women's Studies?

I INTRODUCTION
The almost worldwide emergence of women's studies as a field of research, teaching and study is generally viewed as resulting from the impact of the international women's movement on the academic establishment. Over ten years after the United Nations Decade for Women held its final conference in Nairobi, the year after the Beijing Conference seems an appropriate time for us to review the contribution of women's studies and gender research to social science in and on Africa.
Publications in these areas have proliferated, testifying to the opening up of a vast new field of research, and to the generation of a new body of knowledge that is not only about women but largely carried out by women scholars motivated by their expressed commitment to furthering the interests of women. While there have always been studies of women, the women's studies that emerged in the 1970's and 1980's was almost exclusively research and teaching conducted 'on women, by women, for women' that is to say, it was firmly identified with the feminist agenda of liberating women. Such studies can justifiably be referred to as 'feminist studies', because they had a progressive political agenda as well as an intellectual one. During the seventies feminists in the West, determined not only to gain a place in these bastions of male privilege and power, but also to transform them into places where women too could study, develop and empower themselves, mounted a full scale assault on the academies. This assault took various forms.
Some women sought the establishment of certified courses in women's studies in the existing institutions of formal education. Others concentrated their efforts on fighting for better representation of women in the mainstream of the academic establishment, arguing that more women should be appointed to the staff and administration of higher education institutions, and that more female students should be admitted to hitherto male-dominated areas of study. Yet other women, disaffected with the existing systems, pioneered the establishment of independent study groups, libraries, research and documentation centres for women. Feminists engaged in science have made major theoretical and methodological contributions to world scholarship across the disciplines. It is now possible to speak confidently of 'feminist science' and debate the ideas of feminist philosophy and epistemology, feminist theory, feminism and methodology, and feminist research, and to refer to the extensive literature on all of these. It is also possible to identify feminist contributions to any of the major disciplines. Women's studies, defined by their historical link to the women's movement, have thus given birth to a range of changes within the scientific academies, and within scientific production. The extent to which the link between the women's movement and academic feminism has become a subject for debate, both in the
West and in the rest of the world. The term 'feminist studies' has been coined by those concerned to emphasise the link between activism and intellectualism in the service of women's liberation. Others have opted to use the more neutral and inclusive term 'gender studies' instead of either women's or feminist studies, ostensibly to convey greater neutrality, and to turn the lens on the oppressors (men) as well as on the oppressed (women), or 'gender analysis' to denote social analysis that is premised on recognition of gender inequality.
The extent to which these paradigmatic changes have penetrated African studies and African social science is difficult to ascertain. Some light can be cast on the question by considering that rather diffuse area of research that can only loosely be gathered under the heading of African women's studies. For pragmatic reasons, in what follows, women's studies is defined inclusively as studies of women, studies by women, and studies for women, rather than limiting consideration to studies which are all three of these. This is because the link to the women's movement is no longer as clear and concise as historians of women's studies in Europe and North America would have us believe it used to be. More pertinently, there are good grounds for supposing that African women's studies are undergoing their own distinct evolution.
Women's studies have emerged more recently in Africa, with significant numbers of African women really only developing interest in the 1980's. Prior to this virtually all studies of African women were carried out by Western researchers on anthropological forays into the largely untouched territory of African women's lives (see eg. Paulme 1963, Rosaldo and Lamphere 1974, reviews by Wipper 1972, 1988). The earliest anthropological studies were purportedly apolitical rather than feminist, but since the 1970's, anthropological studies of African women have increasingly been influenced by the concerns of the women's movement. Nonetheless, it is clear that they have remained studies of 'Other' women, if not studies of 'woman as Other' (de Beauvoir 1972).
In the middle of the 1990's it is important to assess the extent to which this legacy has persisted: to what extent can we say that women's studies carried out in Africa now emanate from African as opposed to external and Western interests and forces? If we decide that the answer to this question varies from one place to another, then we might put the question another way and ask: what are the conditions which bestow distinctive characteristics on African women's studies? We can no longer disregard the local and regional influences on this growing body of thought. Today growing numbers of African scholars of both sexes are involved in studies of women and gender relations, carrying out work far removed from the early feminist anthropological studies of women . However, the extent to which these are merely studies of women, as opposed to feminist studies deserves some attention.
Although we can link the genesis of women's studies in Africa to the rise of the women's movement internationally, it would be fallacious to insist that African women's studies are generated exclusively by regional or national women's movements. Other factors are clearly at play. Amongst these are the influence of the development industry, national and subregional political conditions, the crisis in African education and the emergence of state feminism. All of these make it pertinent to ask whether African women's studies can be said to have emerged in the service of African feminism in the same way that say North American or European women's studies are said to have emerged as the intellectual wing of a more widespread women's movement . So, what kind of linkage can be traced between African feminism and African women's studies? Very often it seems that this potentially transformational field of study runs the risk of being reduced to an expedience deployed by undemocratic regimes seeking legitimacy, or a donor-driven phenomenon irrelevant to the democratisation process. This in turn leads us to pose a strategic question: what can be done to strengthen the links that do exist between African women studies and the African women's movement, and so to ensure that African women's studies emanate from the collective concerns and interests of African women?
The reticence of the African middle class towards the idea of women's liberation has meant that only an intellectual minority of this class have overtly embraced feminism, articulating the oppressed situation of African women in political terms. For the most part the African intelligentsia has preferred to dismiss feminism as something alien and Western, to regard the international women's movement as resulting from the activities of a covey of sexually abnormal, man-hating eccentrics far removed from the concerns of 'real' African women. The 'real' African woman of the collective imagination is content with her subordinate position as wife, mother and beast of burden. She is passive in the face of abuse, tolerant of all forms of infidelity; her only real ambition is to retain respectability by labouring for the maintenance of a stable marriage and family and seeing to the satisfaction of her husband's desires (see e.g. Babangida 1988).
Concern over the marginalisation of women in African societies only really took firm root in the consciousness of Africans when the decade for women ushered in a discourse which did not challenge the gross inequalities of prevailing gender relations, under the rubric of 'women in development'. WID, as it became known is able to avoid directly challenging patriarchy and capitalism and demanding the confrontation of women's oppression, instead targetting women as a group to be 'integrated into development'. Under this rubric, governments of whatever political colour are called upon to mobilise women for their vaguely defined development agenda, in the name of the equally vague notion of national interest. This call gave rise to the establishment of regional and national governmental machineries for women all over the region, ranging from women's desks in ministries of social welfare, to departments of women's affairs, to the grandly named ministries for women and development. Indeed many of these high-profile structures were pioneered in Africa, and became a source of regional pride (see Snyder and Tadesse 1995). However, so far there is little evidence that government structures for women have become effective vehicles for the articulation and defense of women's collective concerns and interests.
In contrast to the conservative tone of women in development (WID) discourses, the long involvement of women in the radical politics of the region has given rise to a more subversive tradition of militancy with clearly feminist elements. As early as the 1920's, Egyptian women the nationalist movement had formed an autonomous organisation, the Egyptian Feminist Union, and were producing their own magazine (L'Egyptienne) . However, it was more typical for women to throw their support behind male-led struggles than to organise autonomously to fight for more equal gender relations. During the 1950's Ghanaian women rallied around Nkrumah's calls for mass mobilisation, and Kenyan women took to the bush as fighters and played key roles in support of the Land and Freedom Army. During the 1960's Algerian women joined the National Liberation Front (FLN) to smuggle weapons and explosives and carry out intelligence work against the French, Josina Machel formed the first women's unit in the FRELIMO guerilla force and Angolan women responded to Augustino Neto's call to resist the Portugese, to name only a few of the examples enriching African history. Women's involvement in African liberation struggles became a favoured subject for feminist scholars during the 1970's and 1980's, hence the documentation of numerous historical examples of female militancy across the region, not so much amongst the elite as amongst ordinary women in both rural and urban communities (see eg. Amadiume 1987, Urdang 1989, Walker 1991, Manuh 1991a).
During the 1970's, the domination of international fora by Western women with at best charitable attitudes to women from the capitalist periphery increasingly came under challenge. A small number of highly educated African women entered the fray, protesting against their being marginalised, and demanding that their voice be heard. At the intellectual level, it was made clear that African women were no longer content to merely be the objects of study, whose situation was used to test and verify theories conceptualised elsewhere, by Western women scholars whose concerns and preoccupations often differed from their own.
With the establishment of the Association of African Women for Research and Development (AAWORD) in 1977, African feminist intellectuals sought to institutionalise their presence, and so to articulate the agenda of African feminism by facilitating research and activism by African women scholars. Early in its existence AAWORD facilitated workshops on methodology, women and rural development, reproduction, the mass media and development assistance. Despite these promising beginnings, and for reasons that cannot be detailed here, AAWORD has so far not been very successful in acheiving its goals, or in bringing together the growing numbers of African women scholars now involved in women's studies and gender research within and beyond the region.
In the absence of a viable regional forum however, a series of nationally-based initiatives have given autonomous voice to African women scholars, amongst these the Women's Research and Documentation Centre at Ibadan University, The Development and Women's Studies group at the University of Ghana, the Women Research and Documentation Project at the University of Dar-Es-Salaam, the gender unit at Eduardo Mondlane University in Mozambique, and several others in the process of formation. A number of primarily activist projects have also generated new understandings of women's realities, among these the Zimbabwe Women's Action Group, Women in Nigeria, the Tanzanian Media Women's Association, and Zambian women working on woman abuse. There has also been a proliferation of regional and subregional projects and networks working on gender issues, of which the Women and Law in Southern Africa project has perhaps been the most dynamic to date. A number of university courses have also been established over the years, but a great many of these have remained under-staffed, under-resourced and marginal, often depending on the efforts of individual faculty members and a high degree of voluntarism (eg those at Ahmadu Bello University in Zaria, Nigeria, Makerere in Uganda, and at the Universities of Addis Ababa, Zimbabwe and most recently, Malawi).
In keeping with this situation, it is true to say that most of the material published within the region has not been generated within any autonomous African feminist forum. Instead, most of it has been either undertaken by scholars visiting from more affluent places, or by individual African scholars, often operating under extremely difficult conditions. Some work under the auspices of national universities, sometimes assisted by the support of international donors. Others are only able to carry out research in unequal collaboration with Western scholars who bring in the funds. A great many other would-be-researchers remain unresourced, available to service international agencies as assistants. The most foolhardy embark on unsponsored field work and produce research reports that never make it into publication, but lie gathering dust on office shelves.
It seems fair to say that although African women have long been favoured as objects of study within women's studies, perhaps a result of many of the early feminist scholars having background in anthropology, it has taken somewhat longer for African women's studies - studies by African women, for African women - to develop. Combined with the dismal decline of African universities over the last two decades, it has been difficult to sustain the material base of the research community and its intellectual autonomy, let alone open up new areas of study.
CODESRIA's 1991 seminar on Gender Analysis was the first regional gathering to include both men, who treated the issues with varying degrees of seriousness, and women (see Imam, Mama and Sow forthcoming). Heated debates culminated in participants acheiving some consensus that gender research was a legitimate enterprise to engage in, and one which had already yielded a body of knowledge and methodological innovations of value to African social science in general.
It is still true to say that a disproportionate amount of the internationally available research on African women and gender relations is that carried out by Western scholars, often guided by philosophical, theoretical and methodological concerns that emanate from Western rather than African feminism. This is particularly evident when we consider the publication lists of the major international publishing houses. It is significantly less evident from the listings of in-house research and publications produced in local research institutions and universities for here we see that there is a burgeoning interest in women's studies that is being spearheaded by African women scholars. This is remarkable in view of the impoverished and declining condition of so many African academic institutions.
The following essay considers women's studies and gender research conducted in and on Africa during the last six years, regardless of the race or origin of the individual scholars. Needless to say one cannot help being cognizant of the fact that unequal power relations exist between Western and African scholars, as much within gender research and women's studies as elsewhere. The aim of this review is to stimulate and inspire the production of more and higher quality work in the field of African women's studies and gender research. In so doing I assume that encouraging more African scholars to undertake and engage in research and publication in this area is the best way of doing this. Not only can indigenous scholars bring local understanding of the subtleties and nuances of our diverse and rapidly changing realities to bear in theorising and knowledge production, but the possibilities of conducting academic work with intrinsic value to the communities under study are far greater if such scholarship is indigenously grounded. No doubt anthropology has generated a great many insights into so-called primitive societies; even so, the acheivements of anthropologists are seldom compared to those of Marx or Freud or Foucault - all of whom have developed their theory in reference to a detailed understanding of the societies they knew best. When Africans study Africa, not only are the conclusions they draw about African societies often different from those reached by others, as Diop's historiography or Hountoundji's philosophy indicate, but their theoretical and philosophical relevance to the region and to world scholarship is likely to be more profound (e.g. Diop 1974 & 1978 , Hountoundji 1983).
In discussing the rather wide-ranging selection of material gathered for this review, I concentrate on drawing out the main preoccupations and theoretical advances in each area of interest. In the final section I examine the extent to which gender research and women's studies in Africa successfully reflect the African feminist agenda of transforming gender relations in the direction of greater equity.
My initial literature searches revealed a vast number of publications released over the last five years. A significant proportion of these are reports produced by international agencies and aid organisations, notably the International Bank for Reconstruction and Development (IBRD) better known as the World Bank, the various United Nations agencies and other multilateral and bilateral agencies. These I have not included in this review. Instead I have limited my consideration to publications from universities, research organisations and those nongovernmental agencies involved primarily in research (like the Centre for Basic Research in Uganda, CODESRIA and AAWORD). My coverage is thematic rather than comprehensive, and is confined to materials published in the English-speaking world. I leave to others the task of carrying out similar studies of French, Portugese and Arabic publications in this wide and changing field.
To make discussion coherent, I have found it expedient to group the material into a number of areas. This does not mean that these areas are distinct. For example it is apparent that many publications concerned with sexuality could be grouped readily under the heading of health and reproduction or under cultural studies. Sex work could come under sexuality, but I chose to place it under work since the most insightful research on the subject looked at it as an economic activity. Where the discussion demands, items are referred to under more than one heading. The three major divisions I finally decided to use are as follows:
I.Women and the State, to include governance, politics, nationalism, liberation movements and structures for women.
II. Culture: religion, sexuality, identity and life history studies.
III. Work and economy: urban and rural, formal and informal sectors, domestic labour and sex work.

Each section comprises a discussion of the literature reviewed in the light of the concerns raised by way of introduction. This is intended both to indicate the state of the art and to identify areas for further work, particularly by scholars based within the African region.
I conclude with a brief consideration of the future of African women's studies and gender research in the region and the implications for African feminism.

II.WOMEN, POLITICS, AND THE STATE
Precolonial states
Very few historical studies of the role of African women in precolonial African states appear to have been conducted over the last five years. This has helped to perpetuate the general paucity of information and understanding of gender relations and women's lives in that period. Even prior to this, information has been often been as much mythological as factual. Nontheless the cultural and psychological importance of well-known historical figures like 16th century Amina of Zazzau and the 18th century Nzinga of Angola, is evident in the evolving feminist consciousness of African women. Awareness of this must lie behind the publication of accessible collections of profiles of eminent women in history for school or university use (eg Sweetman 19 , and more recently Awe 1992).
A number of earlier studies emphasized the complementarity of sex roles in what they describe as 'dual-sex systems' of government. This thesis put forth the argument that the European colonial state excluded African women from playing hitherto important and valued roles in political life (eg Okonjo 1983, Arhin 1983, Amadiume 1987), a thesis which must be viewed more critically today in the light of the evidence that in precolonial systems too, most women were subordinate to most men (e.g. Manuh 1991a).
Musisi (1991) provides us with a rare study of the relationship between women, polygyny and the state in precolonial Buganda. She confronts the methodological challenge of relying on limited information, undertaking a reinterpretation of the available accounts, both ethnographic and biographical, written by men. these she supplements with interviews with 'the last wife of nineteenth century royalty'. Her study reveals how political structures both shape and are shaped by the marital and gender relations of the ruling class, particularly by the 'grand polygyny' practiced by the political elite. State formation coincided with increased social inequality, and a deterioration in women's power was one aspect of this. Nontheless, her research demonstrates the involvement of Baganda women in state formation, not only as mothers of kings and king-makers, but also as providers of political balance and cementers of alliances. While very few women wielded much real power (notably princesses, who were also exempted from marriage), a great many functioned as objects of exchange in relationships between men and groups of men. This study indicates not only how much can be learned from studies of how precolonial gender relations affected and were affected by politics and statecraft, but that the method of reanalysing existing data from a gendered perspective is a viable one which can open up this neglected area of historiography (see also Mikell's 1989 paper on the role of women in the Akan kingdoms).
Even the most recent publications on ancient and precolonial Africa do supply little detailed information on the roles of non-royal women in politics and statecraft. Because of the focus on the few exceptionally powerful women who have invariably been members of the feudal aristocracies, one might be forgiven for gaining the impression that ordinary women in general were politically unimportant in many African societies. However, earlier historical studies suggest that no such conclusion is warranted in the absence of further empirical research on the subject (Leith-Ross 1939, Okonjo 1983).

Colonial States
Studies of the colonial period are far more numerous. Much of this work initially set out to counter the imperialist claim that colonialism improved the condition of African women, hitherto living as slaves and beasts of burden at the mercy of virulently patriarchal traditional cultures. What has been revealed is a far more complex scenario, in which transformations in gender relations have not only been mediated by class, ethnic and cultural factors, but have also varied widely across Africa and between different forms and stages of colonialism. Colonial states appear not to have ever resolved the administrative, legal and social problems posed by the fact that whereas the state required certain things of men and set about exploiting and extracting these, African societies have always comprised both men and women. The resulting confusion is reflected in the contradictory official attitudes towards the presence of women, particularly apparent in the administration of urban areas. Here we see that periods of tolerance, during which women settled in the towns performing various tasks which usefully provided all manner of social services to men recruited to live and work in labour compounds, have been interrupted by the harsh imposition of decrees and edicts designed to remove women from the towns (White 1990, Schmidt 1991, Barnes 1992). Clean-up campaigns sought to selectively remove women viewed as disreputable, explicitly those engaged in liquor brewing and selling and the provision of sexual services. Yet, the virtual absence of social and welfare provisions meant that removing even targetted groups of women also undermined the informal systems that sustained the cheap labour pool on which the colonial state relied. Writing on Zimbabwe (then Southern Rhodesia) Schmidt (1991) uses detailed evidence to argue that there was deliberate connivance between African patriarchs who felt that women were 'getting out of control' and colonial authorities for whom the control of women's and children's labour by African men was necessary for both the establishment and consolidation of colonial rule. What is clear is that African and European men shared the idea that African women were inferior beings who should be kept under male control. Consequently, both were disturbed by the growing tendency of women to exercise greater mobility and to flee the tyranny and drudgery of life in the homesteads, something that is clearly reflected in the numerous recorded complaints to the authorities between 1910 and the 1930's. In her book, Schmidt details various ways in which inflexible customary laws were developed so as to coerce women into staying with their husbands, highlighting the way in which

...indigenous and European structures of patriarchal control reinforced and transformed one another, evolving into new structures and forms of domination (1991:734)

Barnes 1992 takes a slightly different position, emphasising the ambivalence of a colonial state which provided few services for the indigenous workforce, and therefore ultimately relied on the services provided by African women in the towns and mine compounds. Although she is reluctant to support Schmidt's (op cit) thesis of an 'unholy alliance' between patriarchies, Barnes (op cit) too draws our attention to the disapproval of women's mobility and the 'immorality' that both African and European men associated with their presence. The contradictory policies that were deployed in colonial efforts to allow some categories of women and yet remove others from urban areas reflected a deeper struggle over the appropriation of women's labour.
Literature on this period highlights the way in which the colonial era, accompanied by the introduction and consolidation of capitalism, saw a gradual commodification of gender relations. The tradition of lobola for example, changed from being symbolic to becoming an important source of cash which they could rely upon to pay taxes and other expenses. This was the case until their daughters started moving to the towns and selling their services directly to men, keeping the money themselves. Documenting the shifts in policy over time, Barnes (op cit) concludes that in this contradictory policy climate, the constant resourcefulness of women produced an administratively insoluble situation. In the end, a great many women retained their mobility, thereby gaining some control over the fruits of their labour under hostile circumstances.
Studies of the actual role of women in colonial government have until recently been rare. Denzer (1989) provides us with a study of women's employment in colonial government service, in Nigeria between 1862-1945. Here women were largely deprived of access to education, something she attributes to the fact that 'Nigerian and British attitudes concerning female roles had much in common' (1989:6). Once women did start formal schooling, there was a slow incorporation of women into very restricted areas of government service - women could only be employed in the positions of nurse, lady physician, school mistress, education superintendent or confidential secretary. Even in these select areas, women were denied access to any position requiring them to exercise authority over men, a policy justified on the basis that the idea would be too alien to Africans. So began the institutionalised gender segregation that has affected the employment of women in government - the largest formal employer - to date. Even the nursing profession was dominated by Nigerian men until 1949, when a policy to replace them with more women was introduced. Denzer catalogues the pioneering careers of the individual women who did work in government during the colonial period.
Both Manuh (1991) and Perbi (1992) reveal a similar pattern of gender segregation in colonial Ghana, with even the most highly qualified women being refused entry to the administrative class of civil servants, and those employed being expected to resign once they married or got pregnant.
Similar studies pulling out the available data in a range of African countries would be of great benefit to our analyses and understanding of the gender dynamics pervading contemporary African states, and highlighting some of the longstanding obstacles to women's equal participation in governance today.

National Liberation
Studies of women's participation in national liberation and independence movements have been more numerous than those of women in colonial states. One wonders if this reflects the identification of feminism with other liberation movements? If early studies were optimistic about the potential for the transformation of gender relations through women's involvement in national liberation struggles (Urdang 1970, Mies and Reddock 1982), research conducted in the 1980's has more often reflected disappointment that the fruits of liberation appear not to have been equally shared between the genders, Algeria and Zimbabwe being the most obvious examples.
In the last six years these have been only slightly tempered by more mature assessments of how far the exigencies of war can be expected to translate into women's equality once the war is over, and other forms of subversion take over (eg. Urdang 1989, Lazreg 1990, 1994). Incisive analyses of the gender politics of women's participation in either the army or the armed struggle dispense with the notion that picture-poster images of women carrying guns herald women's liberation (Cock 1992). History has taught us that participation in armed struggle does not guarantee gender equality in peacetime, with commentators on Zimbabwe drawing our attention to the harassment and humiliations experienced by former freedom fighters disparaged on the basis they are said not to make 'good wives'.
Urdang's (1989) study of post-independence Mozambique catalogues the way in which post-war reconstruction and development were being hampered by South African-backed sabotage. Here, economic catastrophe has followed the war and undermined the struggle for change. These generally unfavourable factors nothwithstanding, she also discusses the loss of momentum with regard to women's liberation, specifically noting a number of opportunities missed by the official Organisation of Mozambican Women (OMM). Never an autonomous movement, the OMM appears to have offered rearguard support for the ruling party and concentrated its efforts on mobilising women for political support and economic production, that is, in rather less than revolutionary areas. From her account, even the famous Frelimo Women's Detachment founded by Josina Machel in the 1960's appears to have been more of a ploy to shame men into fighting than anything else, given that women who joined up were deployed not as combat troops, but rather as porters, caterers and providers of other support services (Urdang 1989). With apartheid South Africa supporting RENAMO during the period of her research, actual transformation of gender relations was clearly being subsumed in the struggle to survive, with the heady promises and commitments to the liberation of women which were made in the earlier period remaining unfulfilled.
Lazreg's analysis of gender and politics in colonial Algeria (1990) notes that the French had an obsessive preoccupation with Islam. In popular reactions against this, Islamic conservativism became a key aspect of Algerian self-identity. Similarly, their repeated attempts to symbolically appropriate Algerian women (e.g by unveiling them on public stages) alongside appropriation of the Algerian nation, has had persistent undermining effects on Algerian gender politics, effects only partly mitigated by the heroism that women demonstrated during the war. Even then the FLN's definition of women's tasks was based on the conventional sexual division of labour, allocating to women the tasks of lending support to combatants (male), information, liaison and supply work, sheltering militants from the police and assisting the families of guerillas and prisoneers (Lazreg 1990, 1994 Tlemcani 1992). Nontheless, as these writers point out, even this somewhat limited participation in the war marked such a radical break with tradition, that it has not been without influence in postindependence politics.

Post Independence States
I have noted elsewhere that the constitutional and legal status of women and the level of women's participation in governance are often taken as key indicators of the general level of democracy in a society (Mama 1995a). Because of the history of women's involvement in the independence struggle waged across the region, in Africa it is also taken as a measure of the extent to which the promises of nationalism have been fulfilled. Since the UN Decade for Women (1975-1985) and the rise of a highly articulate international women's movement, governments everywhere have found it expedient to display a certain level of committment to the participation of women in development. International donor insistence that attention be given to 'the gender question' has now been incorporated into the post-Cold-War preoccupation with democratisation. It is in this context that African governments have lately come under international pressure to become more democratic. Under the twin rubrics of 'multipartyism' and 'good governance' international bureacracies have now begun to articulate their disillusionment with military dictatorships, one-party states and the other male-dominated forms of authoritarianism which they have for so long financed and supported. Whatever rhetoric is used, women's involvement in politics and governance is rightly seen as integral to democratisation. Furthermore, a decade after hosting the Nairobi conference, African women themselves now expect to play more significant roles in national politics and public life.
In view of the patriarchal character of colonialism itself, and the correspondingly high level of women's participation in independence struggles, it is somewhat ironic that postindependence governments have not made more viable efforts to involve women at decision-making levels of politics, finance and governance. Instead, we have a situation in which wealthy Western nations, now operating as creditors instead of colonial powers, and formerly so disinterested in the well-being of African women, are one of the main forces pushing African governments to address themselves to the question of improving women's level of participation. Indeed, the empirical evidence shows that the vast majority of African women are still working like beasts of burden in under-renumerated tasks, are still almost completely excluded from decision-making levels of government, and play only marginal roles in national politics and public life.
Perhaps it is because women are so marginally involved in government in Africa that scholars have been slow to examine the gender dynamics in statecraft and politics. Yet it is precisely in the areas of government, statecraft and politics that, at least until recently, the bulk of national resources and decision-making power have been concentrated, and from which women have been largely excluded.
The research that has been carried out on the subject of women and the state in Africa has not taken women as actors within the state, or looked at their participation in the public services, administration or policy-making. The influential collection edited by Parpart and Staudt (1989) primarily addresses the effects of the colonial and postcolonial state on different aspects of women's lives, lives which are portrayed as being lived largely outside the auspices of government and the formal economy. It is generally agreed that the state has acted primarily as a vehicle for elite male interests, enhancing and extending men's power over women and offering women little access to ownership of land or means of production, few avenues for participation in the formal economy, and even less political power. Other researchers have drawn attention to the means by which women have struggled to defend and advance their collective and individual interests under changing conditions (Mbilinyi 1989, Munachonga 1989, Tranberg Hansen 1989, Jacobs 1989, Tsikata 1990, Schmidt 1991, Barnes 1992). Nonetheless, Chazan (1989:186) feels able to conclude that African women have played no significant role in statecraft, and have not been able to influence decision-makers in any consistent manner. She feels that this is why state policies towards them have continued to be both discriminatory and coercive. Whether or not one agrees with the grim conclusion that Chazan was able to draw at the end of the 1980's, it is still worth asking whether the same conclusion could be drawn today?
The early emphasis on the effects of the state on women generated little discussion of women's overt or covert influence on national politics and policy-making. This may partly be because the state itself has been conceptualised in fairly monolithic terms until as recently as the late 1980's. The model of the African state is that of a powerful leviathan acting on the various groups within society. With a strongly dominant central state in operation, marginalised social groups experience the state as oppressive and exploitative, rather than beneficial. Since the state does not serve their interests, many choose to operate largely outside the realm of the state, a notion encapsulated in the concept of 'exit' (Fatton 1989).
More recent work by African feminist scholars has overturned the notion that women exist largely outside the state, beginning to analyse women's role in statecraft itself (Tsikata 1989, Manuh 1993, Mama 1995a). Generally these studies examine the important roles that women and their organisations have played in support of nationalist movements and political parties, but bemoan their lack of success in effecting changes in gender politics and advancing women's interests. Tsikata's study targets women's political organisations in Ghana, providing us with a critical and clear-sighted analysis of how it is that, despite the impressively high level of activity, these have not been able to tackle the important problems faced by women in Ghana. This failure is attributed to:

the unfavourable political climate for independent struggles; the political character and practice of the organizations in question; and weaknesses arising from the lack of a tradition of independent political organization in the country (Tsikata 1989:73-74)

Tsikata documents a pattern of co-option of women's organisations by the ruling groups - first by Nkrumah's Convention People's Party and more recently by the Provisional National Defence Council under Rawlings. She concludes that for women's political organizations to outlive the ruling regime, and to be more effective vehicles for women's political as well as social interests, would require a change in the political culture. Such a change would favour independent organisation and the pursuit of objectives emanating from the concrete conditions of ordinary women's lives, rather than objectives thought up by a leadership representing ruling political interests.
Manuh (1993) also writing about Ghana, analyses relations between women, society and the state under the rule of the PNDC. She comes to similar conclusions regarding the need for greater independence of women's organisation. In fact, Manuh's analysis of the women's movement concludes that this has not so far engaged in the necessary questioning of social structures and unequal gender relations. Instead it has continued to mobilise women as another support base for a ruling regime seeking its own consolidation and legitimacy.
Nzomo (1993) concludes Khasiani and Njiro's (1993) collection on the Kenyan women's movement in similar vein, describing the movement as 'muzzled and toothless'. She observes that despite their numerically high membership, the Women's Bureau (MYWO) and the National Council of Women of Kenya (NCWK) have been ineffective in empowering women and advancing their participation in decision-making. She attributes this to the fact that these organisations are controlled by conservative leaders who have consistently supported the ruling regime.
Likewise, Abdallah (1993) also decries the failure of women's organisations to address issues of gender subordination in Nigerian society, or to challenge the conservative agenda of government programmes like the Better Life Programme instituted under the Babangida regime. 'Military regimes' she writes ' are by their nature repressive and undemocratic and cannot therefore undertake responsibility for the liberation of any sector or group in society' (1993: 32).
Mama (1995a), also writing on Nigeria, analyses the emergence, not of a women's movement, mass-based or otherwise, but of a femocracy. This she defines as a feminine autocracy paralleling and servicing a persistent military dictatorship, and advancing a highly conservative brand of gender politics in the name of 'women in development'. This argument reaches similarly pessimistic conclusions about the likelihood of existing organisational forms challenging women's oppression or advancing women's political, social or economic interests.
What then are the prospects for women's involvement in the state and politics in the coming years? This is clearly an area for more study. So far, only one collection of seminar papers (Kabira et al 1993) attempts to consider women's likely role in democratisation . Following the introduction of multipartyism in Kenya, it contains useful discussions of the possibilities democratisation may afford women in the areas of health (Ngechu), reproductive rights (Khasiani), education (Mukudi; Obura), culture and language (Mukabi-Kabira; Adhiambo-Oduol; Ngechu), and political life (contributions by Nzomo; Kameri-Mbote et al & Gachukia). However, there is no critical discussion of whether Kenyan multipartyism is likely to translate into fully-fledged democracy, something which seems to be assumed by the contributors. Nor are we informed about the actual involvement of women in party politics. Nzomo introduces the collection by defining a democractic political system as 'one which encourages and makes possible the free and voluntary involvement of the people in the political life of the nation' (Nzomo 1993:7), noting only approximations of this ideal have ever been acheived. She argues that women must be involved as much as men for a system to be trully democratic. Although she cites facts and figures to point out that Kenyan women have been completely marginalised from politics and public-decision-making, she nontheless lists a series of strategies for overcoming this history of exclusion. Amongst these strategies she specifies the need for a strong and autonomous women's movement, political education and mobilisation and the identification of women candidates.

Recent research carried out in predominantly Muslim areas of Africa has centred around comparison of the gender politics of the nationalist period and those of the contemporary period, during which the rise of Islamist movements has undermined the dominance of secular and socialist discourses. The struggle between secular nationalism and Islamism appear to be most pronounced in the Algerian case (see Lazreg 1990 Knauss 1992, Cheriet 1992, Tlemcani 1992, Baffoun 1994, Bouatta & Cherifati-Merabtie 1994). It is however, also a key theme in gender-aware studies of the state and politics carried out in Tunisia (Baffoun 1994), Egypt (Badran 1994, Shukrullah 1994), Sudan (Hale 1993) and Senegal (Creevey 1993, Callaway and Creevey 1994).
The main themes of this literature can be illustrated with reference to the Algerian studies, which are by far the most detailed. Particularly outstanding is Lazreg's (1990, 1994) work, which furnishes us with a thorough case study of Algerian women's situation from precolonial times until the present. She reanalyses the gender politics of the nationalist discourse espoused by the FLN to highlight the ways in which they accomodated conventional notions of femininity. She distances herself from the Western feminist analysis which implies that Algerian women were duped into participating in a revolution which then betrayed them, providing a more balanced and nuanced understanding of the gender politics of Algerian nationalism.
During the war, in which over 10,000 women are on record as having worked for the FLN, mostl were in the civilian service where they performed tasks that were 'less spectacular, perhaps even tedious, but equally dangerous' (Lazreg 1994: 124), taking charge of food, medical and weapons supplies and the underground network of sanctuaries, and washing clothes for the fighters. Only 11% of the military wing were women, mostly serving as nurses. Very few engaged in highly publicised paramilitary acts of destruction or combat, and it is hard to make sense of the reluctance showed by the FLN on the question of arming nurses working in combat zones for their own protection. Casualties were high for both sexes: as many as one woman in five was either imprisoned or killed by the French during the war. The fact that more than half of the women registered with the Ministry of War Veterans who were killed were under twenty five years of age indicates that many of those who joined up were young, single women who did not have children they would have had to leave.
Lazreg concludes that priority had to be given to driving out the French, and in any case, women did not voice concerns that would be considered as 'feminist' today. Instead they tended to assume that 'they would naturally be recognised later'(1994:140). These factors, combined with the complete absence of women in the leadership are used to explain why the FLN failed to take any concrete steps to redress gender inequality.
At the end of the war, victory was characterised by a restorationist rather than a revolutionary mood, with young vigilantes, and Mohammed Khider, the leader of the FLN ushering women back to the kitchens. No women served on the FLN Central Committee, either during Ben Bella's government (1962-1965) or during Boumediene's (1965-1978). Nor has any woman ever sat on the National Council of the Revolution. In 1968, the National Union of Algerian Women, formed in 1962, was incorporated and subordinated to the FLN (Knauss 1994, Lazreg 1994).
Analysis of Algerian nationalism is followed by an impressively thorough study of how women have fared under the different regimes that have governed Algeria since independence. Broadly speaking, after independence, official statements and documents suggest that FLN government did initially intend to pursue the gains made by women through their participation in the war, but never went beyond polemic. The 1964 Charter of Algiers, for example, made general recommendations, but the only concrete prescriptions for women pertained to widows' pensions. Lazreg sums up the early 1960's period thus:
'Women, as a group, were seen as necessary to the building of the state, but as contributors, not participants...Sacrifice, not duty complemented by right, was the cornerstone of the new state's view of women (Lazreg 1994:146).
Subsequently, the neither the National Charter (1976) nor the constitutions have committed the state to defending women's rights. Gradually, committments to socialism have been superceded by a liberalism more concerned with appeasing the growing Islamic brotherhoods than with challenging them. Until the 1980's, gender relations provided an uncontested site of compromise.
The Family Code was drafted in the 1970's, but not passed until 1984, largely because it proved so contentious, not only to women, but also to conservative and liberal men, for different reasons. For women it heralded the institutionalisation of the gender inequality that had increasingly been encroaching on their lives over the previous two decades. Public demonstrations and lobbying made feminist activity visible for the first time. Women were unable to prevent the passage of the Code, and the fact that it took this codification of sharia law to draw women out of their political stupor is telling. However, this setback has also provoked the emergence of a small but independent women's movement which Lazreg regards with optimism.
What the afore-mentioned studies of gender politics in predominantly Muslim African states show is a pattern in which nationalist and secular forces have repeatedly given ground on gender issues in futile efforts to appease and accomodate the gathering forces of political Islamists. This they have done by condoning the reassertion of patriarchal values and institutionalising gender inequality, often in the name of cultural authenticity.

South African studies on women and gender display the most radical gender politics in the region. Unapologetic in their commitment to women's liberation, and their discussion of feminism, South African scholarship displays great awareness of the failings of African national liberation struggles when it comes to the question of gender equality. Perhaps because there is a long tradition of activist scholarship in the region, women scholars here have not hesitated to be critical of the conservatism that has, at least until recently, characterised the gender politics of African as well as Afrikaner nationalism. Occasional progressive statements in favour of women's participation in the struggle notwithstanding, several authors have drawn attention to the unreconstructed practice of politics: the exclusion of women from leadership positions within the nationalist movement, the view that women's liberation is a divisive issue that must only be considered insofar as it facilitates and remains subordinate to, national liberation (Hassim 1991, McClintock 1991, McFadden 1992). Until the end of the 1980's at least, it was pertinent to pose the question:

'Why has the oldest anti-colonial struggle on the African continent not moved beyond the conventional nationalist position on gender?' (McFadden 1992:512)

Several researchers have homed in on the way in which ideologies of motherhood have been mobilised, not just by conservative forces such as the Afrikaner nationalist organisations and the Inkatha Freedom Party, but also by the African National Congress (e.g. Gaitskell and Unterhalter 1989, McClintock 1991). In focussing on one of women's roles to the exclusion of others, all such ideologies fix women in accordance with their biological proclivities, and ignore the fact that, important though motherhood may be to many women, not all women are mothers, and many women are other things besides the producers of subsequent generations of Zulu, Boers or African nationalists.
South African studies of gender and national politics have examined the momentuous changes of the last few years from a gender perpective. Whereas some retain a healthy cynicism with respect to the prospects for women in a new and democratic South Africa, at least one analyst finds grounds for optimism. Seideman (1993) draws our attention to the structure of the South African economy, in which over one third of African women are were employed in the formal economy by 1985, making South African women the most proletarianised in the region. She also extols the high level of independent political organisation by women at local and community levels, and the extent to which feminist activists within the nationalist movement have managed to push the debate on gender forward and into the public eye over the last few years, despite the historical conservatism of the ANC Women's League. In analysing the changes within COSATU and the ANC, she concludes that since the beginning of the 1990's there has been a qualitative change in gender politics, not emanating so much from the political leaders as from community organisations and rank and file women. Issues formerly ignored by the mainstream - domestic relations, sexual harassment and women's representation in decision-making - are no longer deferred quite as easily as they have been in the past. In 1991 for example, the ANC Women's League, long viewed as a 'ladies auxiliary organisation' chose to insist on a 30% quota of women to be elected to the national executive committee of the ANC. The fact that rank and file men, comprising 83% of the membership successfully opposed the motion until it was withdrawn does not detract from the fact that the issue was raised, and provoked the most heated debate at the entire conference. Her evidence suggests that in the South African case at least:

'...feminist demands within the South African nationalist movement emanate not from a few educated women or the national leadership but from within precisely the popular base from which the ANC draws its support' (Seideman 1993: 315).

More generally, scholarship on women, gender and the African state over the last five years shows a growing attention to the dynamic interaction between state and society, and between competing groups within civil society. Even though the concept of the state is still monolithically patriarchal, society begins to be viewed as comprising men and women, with women's actions and responses having unforseen effects on the male-dominated state, and the relationship between women and the state combining various degrees of autonomy and reciprocity. This change in emphasis is partly an effect of the different disciplinary and methodological influences that studies of women and the state display. Not being limited to macrosociological methods, women's studies have introduced microsociological techniques, and these have in turn brought new insights. Even the most oppressed and marginalised groups are seen as actors rather than passive recipients of whatever the state metes out to the people. Key features and dimensions of state-society relations - the relationships between what is official and what is unofficial, formal and informal and between the centre and the periphery - are all illuminated through the lens of gender.
The state itself must then be viewed as a site of contestation, an entity embedded in complex relationships of reciprocity and conflict with different social groups. The 'woman question' has generally been treated opportunistically and exploitatively by insecure regimes on the one hand wishing to retain creditability within the international community, but on the other, seeking to authenticate themselves with populist appeals to anachronistic notions of masculinity. With this more nuanced understanding of state processes comes an extension of the boundaries of what is deemed to be political, which in turn demands an expansion of the African political science agenda. No longer can political science be limited to the study of the formal and public sphere, but instead it must also address the dialectics of private and public life, of the household and the community, of the formal and the informal.
There are also great gaps in the literature. While a great many seminars have been held on women and politics, or women and democracy, these have tended to focus on multipartyism and mobilising women to run and vote in elections. Meanwhile, a subject as pertinent as gender and militarism remains neglected. Nor are there many studies on the participation of women in the purported clan and ethnic disturbances. What role have women played in the destruction of states, and what will a gender analyses of the genocidal regimes in central Africa uncover? Media footage suggests that Liberian politics is an all-male affair, but is this indeed the case, and if it is, where are the women of Liberia and what are they doing?
The above mentioned studies of gender politics in contemporary African states all agree that women's interests would best be advanced by the existence of national, independent and united women's movements. Several draw attention to instances where this has been an effective check on the unmitigated assertion or re-assertion of patriarchy. The analyses put forward however are more informative on what has gone before and what should be avoided than on current situations and how women's involvement might be strengthened. Politically, African women are no longer novices, but it seems they still have much to overcome.

III. CULTURAL STUDIES

A significant proportion of the research on African women carried out over the last five years falls under the cross-disciplinary rubric of cultural studies, looking at the various aspects of women's lives and their role in the production and reproduction of culture. This interest may be partly due to the fact that many more women social scientists have backgrounds in anthropology, history or sociology rather than in the drier disciplines of economics, political science or international relations. It can also be traced to the methodological innovations of women's studies which have favoured the use of ethnographic methods, biographies and oral histories, all methods more amenable to holistic and cross-disciplinary analyses, all of which are easily accomodated under the umbrella of cultural studies. A third possible cause of the popularity of cultural studies lies in the fact that 'culture' has long been identified as that pervasive terrain on which patriarchy manifests, resisting change and perpetuating women's oppression. Changing the longstanding cultural practices - the traditions - which sustain and reproduce gender inequality is therefore fundamental to the agenda of women's movements worldwide, and research is often viewed as a necessary prelude to such change. The reverse is also true: it is where attempts are made to change cultural practice that the status quo most clearly springs into view and tradition asserts itself most strongly. This means that feminist activism itself has often led to the revelation and reassertion of patriarchal culture, reinforcing the choice of culture as a worthy target of feminist scholarship. Although ethnographic work has often been sensitive to gender, early studies were often imbued with patriarchal biases resulting from the male dominance of the discipline, the reliance on male informants and the preconceptions that researchers from patriarchal Western societies carried over into their analysis. The earliest studies of African women however, (eg Paulme 1963) introduced a focus on women's power in traditional societies which many more recent studies have continued. Cultural studies in and on Africa have begun to address the dynamism, diversity and complexity of gender relations undergoing rapid change as well as producing detailed analyses of the changes wrought in women's lives and realities by a range of political, economic and social forces.

Ideologies of Domesticity
In a somewhat different vein, several studies and at least one major collection (Hansen 1992) have examined the colonial attempts to instill Western notions of femininity and domesticity in African women, efforts intended to ensure the creation of a select cadre of 'suitable wives' for that class of African men entering in colonial serviceas clerks and junior adminstrators. Several of the contributions to Walker's (1990) collection address the role of mission settlements and elite schools in furthering this end in South Africa. Hunt (1990) documents the social clubs were set up to domesticate 'native' women the Belgian Congo. These organisations were not explicitly designed government agencies: they emerged instead out of the 'good intentions' of the wives of missionaries and colonial officials (see also Musisi 1992 on Uganda, Denzer 1992 on Yorubaland, Ranchard-Nilsson 1992). Nevertheless, their agenda was clearly compatible with the interests of the colonial state, which welcomed a development which was to see at least a section of African womanhood being trained in the social graces of Victorian fashion and etiquette, cake making, needlecraft and flower arranging. The ideology of domesticity was also a way of ensuring that women's reproductive work remained outside the public sphere, so enabling the appropriation of labour that had been so integral to the development of Western capitalism. That this sometimes resonated with African traditions of female domesticity is amply demonstrated in the literature on African households (see for example Mack's (1992) discussion of harem domesticity in Kano, Nigeria).

Religion
Religion has been a popular subject of study, presumably because religion, like custom, is a major vehicle for gender ideologies that oppress women. Feminists are in agreement that, on the balance, all the major religious texts have provided justifications for the oppression of women and assertions of male superiority, whether one regards this as intrinsic to the texts themselves, or attributes it to men's success in appropriating these texts. Perhaps because of the predominance of Western scholars interested in societies different from their own, more studies have addressed themselves to Islam than to Christianity within gender research. Fewest of all are studies of the role of women in African traditional belief systems (but see e.g. Amadiume 1987). Major texts on African philosophy make little reference to gender (Hountoundji ), although it is often suggested that women were more powerful in these than in the more recently arrived monotheistic faiths (Diop ).
The preference for studying the effects of Islam on women may stem from the widely held assumption that Islam has generally been more oppressive to women than other religions, an idea that has gained currency with the contemporary proliferation of islamist movements which re-assert conservative gender ideologies. This assumption is questioned from time to time, but not really laid to rest by their evidence, in Callaway and Creevey's (1994) book The Heritage of Islam. Addressing itself to 'women, religion and politics in West Africa', this major work draws primarily on data collected by Creevey and Callaway in Senegal and Nigeria respectively, with occasional reference to other countries in the subregion. The comparative framework they use compares the various forms of Islam practised in Senegal (Wolof, Serer, Tukulor and Dyola), not with the various forms of Islam practised in Nigeria, but with the Islamic culture of the Hausa in northern Nigeria. This creates the impression that Senegalese Islam is relatively favourable to women, a position which would not had been upheld had they either limited themselves to an inter-ethnic comparison of say, Tukulor Islam with Hausa Islam, or compared the various forms of Islam that exist in the two countries . In defense of Nigerian Islam, one would not challenge their observation the Hausa Muslim interpretation is amongst the most conservative in Africa, but rather draw attention to the very different practice of Islam by the very large Yoruba, Nupe and even non-Northern Hausa communities in Nigeria, not to mention the Islam practiced amongst the numerous smaller ethnic groups scattered across the country.
In Senegal, where Islam has been established for longer than amongst the Hausa Muslims of Northern Nigeria, it has also been tempered by the French insistence on limiting the influence of Islam on the national body politic. Another distinctive feature of Senegalese Islam that Callaway and Creevey draw attention to is the system of brotherhoods, some of which they see as allowing women to play a greater role in the public affairs than is countenanced among the Muslim Hausa. Although women are not technically accepted as members of brotherhoods, women do identify themselves as belonging to these associations. They also consult marabouts about their problems as often as men if not more, and play active roles in the social and religious life of their communities.
The strength and coherence of this study lies in way their account centres on the question of whether religion shapes or reflects society, a question they pursue in relation to traditional cultures, women's education, their roles in the formal and informal economies and questions of political empowerment. This analysis takes us some way beyond the assumption that Islam is responsible for the oppression of women, generating an interactive account in which the influence of Islam continuously combines with old and new cultural forces which can, but by implication need not necessarily, subordinate women. Nontheless, it is clear that women are far from equal to men even in the 'more progressive' Islamic context of modern Senegal. Where women have made the greatest advances, it is not due to Islam, but to the mitigation of the effects of Islam by traditional cultures which allowed women greater power, access to Western education and paid employment, or by political or legal conditions which granted women some rights not granted within Islamic law as it has been interpreted in either context.
The fact that the Senegalese Muslim feminist organisation Yeewi-Yeewi is unrestrained in the advocation of gender equality bears a marked contrast to the careful conduct and conservative tone taken by Nigeria's Federation of Muslim Women (FOMWAN) .
Generally speaking, recent studies of Islam in West Africa (Creevey 1991, Callaway and Creevey 1994, and Imam 1994) afford us a more nuanced understanding of the impact of Islam on the lives of specific groups of women - one which acknowledges that Islam has both undermined and strengthened women's position in society at different historical moments, but has always had its effect in conjunction with other variables. In other words, religion does not operate as a determining factor on its own, and it is therefore virtually impossible to assert whether a given faith is more or less favourable to women as compared to another.
Other studies have focussed on the corrosive effects of Islamic fundamentalism on women's situation in various African countries, including the Sudan (Hale 1992), Egypt (Shukrallah 1994), Algeria (Bouatta and Cherifati-Merabtie 1994 ) and Tunisia (Baffoun 1994). All these studies draw our attention to the way in which gender has been a key site for the production and proliferation of Islamic fundamentalist ideologies. The circumscription and curtailment of women's activities and visibility, for example, has been a major dimension of militant Islam in all the countries under study.
Shukrallah (1994) outlines the conditions giving rise to the emergence of the Egyptian Islamic movement, arguing that the failure of modernisation for the majority of the people has led to a crisis in nationalist secular discourse. More than this however, she draws attention to the way in which the leadership of the nationalist movement compromised on the gender question, never really upholding their right to be full citizens in the emerging nation-state. It seems that ever since that time, Egyptian women have faced contradictions between their status as citizens of a supposedly secular nation-state, and their explicitly circumscribed status within the umma (religious community). Within conservative Islamic discourse they are cast as bearers of authenticity, and subjected to a narrow reified identity.
Also writing on Egypt, Badran (1994) looks at the different discourses positioning feminists, pro-feminists (women with feminist sympathies who do not wish to be publicly known as such) and Islamist women, noting that the rise of conservatism in Egypt occurred during the 1970's, just as second wave feminism was sweeping other parts of the world. She uses the term 'gender activism' to accomodate the fact that there are competing discourses on femininity emanating from these different groups of women. In today's conservative-dominated Egypt, feminists, most of whom are Muslim, speak with a muted voice and are reluctant to declare themselves to be proponents of women's liberation, whereas Islamist women are loudly proclaiming their views on women, and are avaowedly antifeminist. They disown historically famous women like Huda Sharaawi as morally corrupt, and insist that women's rights are better protected within Islam, from behind the veil, as it were. It is their efforts to secure women's rights, as prescribed by Islam which qualifies them for the term gender activists. They play an important role in mobilising women for the Islamic movement. It is a situation which is differs markedly from the 1920's, when Egyptian feminists held the centre of the public stage within and beyond the boundaries of the nation.
Badran (op cit) perhaps optimistically, argues that a general widening of Islamist discursive space has allowed the emergence of younger 'Islamist feminists' who do not reject feminist writings entirely, but read them critically, returning to the Koran to appropriate the more acceptable elements of feminist thought. Less scornful of other women's struggles, they call for a movement to liberate women from within Islam, and engaging in the ideological task of reinterpreting the Koran.
Boutta and Cherifati-Merabtine 1994) analyse the FIS newspaper El Mounquid as a way of investigating the representation of women in Algeria's Islamist movement. Although this organ officially denies the need for any consideration of 'the woman question' on the basis that 'Islam has provided all rights to the woman' its pages are nontheless replete with articles prescribing Muslim femininity. The new 'Muslim woman' is construed in manner which is both detailed and contradictory. She is simultaneously celebrated as being superior to all other women, and subjected to a series of taboos, all of which are viewed as leading to depravity and a return to the ever-threatening Jahilia (pre-Islamic state of ignorance). Like the Egyptian and other Islamists, the writers of El Mounquid use biological arguments to justify these taboos: women are the psychological and physical inferiors of men, born to be wives and mothers, and clothed in the cloth of honour and purity - the hijab. The authors point out that this simplistic and static construction of Islamic femininity denies both the diverse realities that constitute modern women's lives and the collective history in which the moudjahidates played a heroic role in the war for national liberation.
Hale's work on the Sudanese Islamic movement explores the role of women in the National Islamic Front (NIF), taking note of the fact that being activists within it has had some dividends for the elite women who have become main spokeswomen. She argues that it is clearly not an emancipatory development, but one that reflects class interests and which is negotiating a new sexual division of labour which has selectively removed women from prestigious professional positions, while allowing less privileged women to continue to work in unrewarding menial capacities, when they have to support their families. It is a renegotiation which accomodates the realities of an economic crisis which has meant that fewer men than ever are able to live up to the ideal of keeping their wives at home, and ensures that women can still be exploitated in marginalised sectors of the labour market while denying them any significant place in public life.
Arguing in a similar vein, Arlene MaCleod (1991) has conducted a rich and detailed study of the lives of lower middle class women in Cairo, particularly examining the conditions which have given rise to the 'new veiling'. In the best methodological tradition of women's studies, she spent years living and socialising with the group she interviews in depth, and presents her findings in a clearly-written and unpretentious manner which allows an intimate level of identification between reader and researched. Although like Hale, MaCleod has an eye for the contradictions between Islamist discourse and socio-economic realities, the women in her study are very clearly not just victims of macrosociological forces, but architects of their own destinies. The openness of her methodology allows us to comprehend the deliberate manner in which those who are adopting the new veiling weigh up their situation and make the choice, invariably as a strategy for attaining other goals. Here are women who will don the veil because they want to continue to work outside the home and at the same time remain respectable, in a socially conservative climate. Furthermore, the veils they wear are modified: no longer a uniform black cover, the modern veil has become a display of indigenous fashion. A range of elegant stylistic variations allow a much fuller expression of individuality, one that is not a mere mimicry of of Western fashion, but culturally grounded. It is the series of compromises and innovations that the new veiling encompasses that has led MaCleod to the term 'accomodating protest', which aptly denotes the constant negotiations and changes that are occuring in urban Egypt at the present time as a result of unique conjunctions of international and local forces.

Much remains to be investigated regarding the legacies of centuries of Judaeo-Christian influences in Africa. Christianity has presented itself as uplifting and progressive, emphasising the role of mission schools in educating and raising the 'heathen'. In opposition to this view, anti-imperialists and cultural nationalists alike have argued that the European religion played a key role in the subjugation of African people, derogating indigenous cultures and instilling a psychology of self-hatred and contempt for African ways. Gender analysis of the legacies of Christianity throws new light on the whole debate, revealing a complex and varied picture in which there were both gains and losses in women's status, changes which were part and parcel of changes in notions of masculinity and femininity, the sexual division of labour and in marital and sexual practices. All of these were inextricably intertwined with the radical transformations that accompanied the different stages and forms that colonialism took at different moments and in different places. Not only did the impact of Christianity depend on what existing gender relations were like, but also on the class and caste position of the women in question. Exemplary historical studies have been carried out in South Africa (Meintjes 1990, Hughes 1990 and Gaitskell 1990), using official records and archival sources to investigate aspects of missionary communities, schools and church organisations between the 1800's and 1945.
Meintjes (op cit) examines the changes in gender relations in a Wesleyan settlement in Natal, noting that the doctrine introduced by James Allison, catechist and missionary, was not passively imbibed by the multiethnic community that constituted Edendale, but that there was rather a complex synthesis between this and traditional norms, a synthesis which was largely governed by the exigencies of survival in a changing colonial economy.
This meant that while the Wesleyans favoured the exclusion of women from work outside the home, African women were traditionally engaged in heavy agricultural labour, with the result that women performed a combination of domestic tasks decreed by Christianity and poorly remunerated farm labour. Boys were taught wagon-making and masonry, whereas girls would be tutored in the skills required of 'angels of the house', namely cooking, preserving, making Victorian clothes and laundering. African women, far from being lifted out of drudgery, then found themselves combining these additional domestic responsibilities with their traditional duties of rearing animals, hoeing, weeding, harvesting, food storage, the making of mattresses and so forth, something undoubtedly facilitated by the emphasis on duty, subservience and quietude in the home.
Although Christian marriage rites were introduced, African marriages continued to be governed under customary laws which afforded women the status of minors under the full control of fathers, then husbands, and on widowhood, husbands brothers or even their own sons. Those women who worked (often before or after marriages) invariably found themselves taking in laundry, becoming maids or seamstresses. As if seeking a positive note, Meintjes also highlights the possibilities for socialising between women that emerged in church groups, and how they took pride in their cooking, sewing and other skills and found avenues for self expression in their homes and elaborate dress styles.
Hughes (op cit) study of the then revolutionary establishment of the first boarding school for African girls presents fascinating study of both the concern to regulate the femininity and control the sexuality of the favoured daughters of converts. African education was almost exclusively maintained by missions and as such centred around the religion. As a school for the daughters of the African elite, the Inanda seminary developed an emphasis on academic education rather than restricting its courses to training suitable for future maids as many other schools did. Nontheless the perceived options, even for relatively privileged African women centred around the dominant notion of femininity: they were to be helpmates to their husbands, teachers or nurses. The conflicts that these girls experienced and the insights they had into their world are glimpsed through Hughes' rare use of excerpts taken from school essays.
The highly visible and active participation of women in African church organisations at times appears to suggest that there are instances of Christianity empowering African women, an argument taken up by Gaitskell when she concludes her study of manyanos, the higly popular and highly cohesive women's unions that accompanied the spread of Christianity to urban areas. Engaging in evangelical and fund-raising activities, and adopting enthusiastic revivalist style praying and preaching, these organisations appear to have been partly responsible for the preponderance of women converts. she discusses the different motivations that men and women had for converting, suggesting that whereas men may have more often bee seeking self-improvement, women may have more often been 'refugees' from tradition, fleeing to mission compounds to avoid forced marriages, escape domestic violence or avoid the misery of widowhood. However, having converted, women at best received a 'contradictory package', with their traditional powerlessness being replaced by subjection to Christian patriarchy and domesticity. It is in this context that Church unions seem to have offered women both a means of self-expression and release and a potential means of empowerment, while at the same time introducing clearly African variants of Christianity.
Studies of the church in contemporary African have continued to show the various churches having contradictory results for African women, sometimes facilitating their oppression and sometimes offering a means of protection or even, on occasion, empowerment, whether through granting women access to Western education, or through women's direct participation in church activities and networks. Two recent studies draw attention to the complex interplay of gender and power within African churches.
Crumbley (1992) for example, looks at the role of women in the indigenous Christian movement that has given rise to the Aladura churches in Nigeria. Characterised by the central role of prophet healers, belief in faith healing and the efficacy of prayer to change material circumstances, by incoporation of features of Yoruba traditional religion and by belief in the validity of divinely revealed knowledge, the Aladura churches have proliferated in and beyond West Africa. She looks at three denominations: the Celestial Church of Christ, The Christ Apostolic Church and the Church of the Lord-Aladura. Only the last of these allows women to play leadership roles, having parallel male and female offices at each level of its structure, and training both prophets and prophetesses. Regarding itself as a progressive and pace-setting church, the Church of the Lord-Aladura is radically different from the Christ Apostolic Church and the Celestial Church of Christ which observe strict menstrual taboos, therefore excluding women from religious participation for a significant proportion of the month. Women are furthermore forbidden to speak unless called upon. In discussing the different 'ideologies of impurity' in these different churches, Crumbley reveals the complexities of cultural analysis which have to take historical and stuctural ambiguities as well as cultural legacies and the role of ritual into account.
Marshall (1991) analyses the rise to religious and social prominance of the newer charistmatic or Pentecostalist Christian movement, thousands of new churches and groups having been set up all over Southern Nigeria. Her analysis links the personal and social processes of rebirth to the creation of autonomous spaces which defy existing power monopolies and propogate new power relations, presenting new opportunities for survival. Despite the doctrinal emphasis on women's primary obligation being to her family, being a born-again woman does offer some avenues for empowerment, offering strategies for resisting the demands of powerful men for sexual favours, and demanding sexual restraint, marital fidelity and monogamy from male as well as female members of the community. In the context of widespread commoditisation of sexuality and the profligate habits of urban men, the fact that a man may be denied sexual favours, or a husband called to account enables women to regain an exceptional degree of control over their sexuality. Overall the movement offers ways of replacing disintegrating extended family systems and patron-client networks with relatively egalitarian and dignified systems of spiritual solidarity and mutual support, systems equipped with the new spiritual power of the born-again community, power believed capable of transforming the social and material world as well as the individual.

Sexuality
Considering that sexuality has been a major area of interest within women's studies internationally, the first question one asks in surveying the African work is why there are so few studies of sexuality in contemporary Africa? It is possible that the historical legacy of racist fascination with African sexuality has deterred Western researchers. For African scholars, sexuality remains a minefield few venture into for fear of further jeopardising careers already endangered as a result of addressing gender issues. The virtual absence of good research in this field means that we lack an analytical understanding of contemporary sexual relations in African societies, at a time when one of the worst scourges ever to sweep the region happens to be sexually transmitted. The resultant problems of ignorance and prejudice are exacerbated by the fact that the work that has been carried out has tended to focus on the most problematic aspects of African sexual behaviour. So it is that we have studies concentrated around the issues of female genital mutilation, prostitution, population control and disease.
This leaves a great many areas of African sexual culture unresearched since the early colonial studies of initiation, puberty and marriage rites conducted within the tribal paradigm, or the coffee-table photographic collections of Leni Riefenstal and Mirella Ricardi. Analytical studies of the transformations in the type and form of relationships between the sexes that have accompanied the wide ranging social, economic and political changes are rare, as are studies of masculinity, femininity, homosexuality or the changes wrought in these by industrialisation, urbanisation, militarism, civil war or any aspect of development and underdevelopment.

In view of the change agenda of feminist-inspired research on women and gender relations, it is not surprising to find a number of studies addressing traditions inimical to women. The 1980's saw the publication of several books and reports detailing the various forms of genital mutilation still being carried out in various parts of the region (El Sadaawi 1980, Abdalla 1982 El Dareer 1982, Warsame and Ahmed 1985, Koso-Thomas 1987). It is an area to which African women scholars have worked hand-in-hand with activist groups to confront the uphill task of organising against the various forms of female mutilation. It is an onerous task partly for the historical reason that the colonialists were the first to proscribe and legislate against these customs in the 1940's and 1950's, with the result that the issue became linked with colonial represssion. This meant that nationalists as renowned as Kenyatta proclaimed support for clitoridectomy on the basis that it was integral to African tradition and morality, a defensive position which defined progressives who wished to do away with the practice altogether as 'traitors'. Today cultural nationalists continue to be apologists, insisting that their ancient traditions have been misunderstood and misrepresented, notwithstanding recent documentation by indigenous as well as international scholars.
During the last five years, the genital mutilation of women has continued to be a major concern of activists in the region, and several major books and articles have addressed the topic (eg. Hicks 1993, Knudsen 1994, SWDO/AIDOS 1989). Nontheless, the literature is far from comprehensive, and would be greatly enriched by a deeper understanding of the cultural conditions under which many these clearly harmful practices have persisted.
The recent work reflects the continuing concern over the damaging effects of the various forms of excision and infibulation in the region, and tell of the various efforts to reduce or eradicate them (eg SWDO 1989). The early tendency to address the issue under the less controversial health policy paradigm has been supplemented by studies which pay greater attention to the cultural dynamics of these practices, particularly the sexual politics underlying them. It is also now recognised that in addition to sustaining male control over female sexuality, many women have vested interests in the perpetuation of the custom.
Perhaps the most ambitious study in recent years is that by Hicks (1993) which focusses on infibulation in Islamic North-East Africa, where the practice of the more extreme forms of excision and infibulation are most pervasive. Deploring the lack of comprehensive and systematic data (even in the Horn of Africa where most of the existing research has taken place) she nontheless sets out to map the cultural correlates of the practice across 26 infibulating populations and 20 noninfibulating control populations, a sample taken from the 105 groups she identifies in her targetted area (Chad, Ethiopia, Sudan, Somalia and Eritrea).
Unique in its scope, her use of a diachronous statistical analysis leads her to isolate and hierarchically order 21 significantly associated variables, and to identify patterns amongst these, eventually coming up with 8 primary variables and 5 secondary variables which are compositely associated with the practice of infibulation. Of primary relevance are: i) wives retaining full membership in natal group ii) male absenteeism, iii) unstable marriages, iv) low position of women, v) sheep and goats herded by wives, vi) high brideprices, vii) Islam, viii) negative correlation with exogamy. Of secondary relevance were nomadism, a preference for parallel cousin marriage, men being camel and/or cattle owners and the use of livestock in brideprice payments. Although many of the variables she isolates can be found individually in other Moslem pastoral populations, Hicks argues that it is the fact that they are compositely found in infibulating populations which suggests that further studies will reveal their functional importance to the practice.
In her discussion of the composite relationships between these variables, Hicks observes that 'unstable households' relate to the exigencies of differential mixed herd pasturing, where ecological and domestic instabilities combine to result in lengthy periods of spouse separation, and where a norm of separate sleeping arrangments, perhaps combined with the demands of polygamy are all likely to lead to psychological distance between husbands and wives. With regard to Islam, she notes that although

'Islam has no relationship to the origin and practice of infibulation, it has functioned, and continues to function, as a vehicle for its perpetuation.'

It does this by maintaining women's inferior status, alienating them from economic resources and public life, perpetuating modesty codes, physical restriction, and their symbolic role as custodians of male honour, and reasserting men's exclusive rights over female fertility. This weak social position means that the only avenue for social status open to women is through their roles as wives and mothers, a status which will be denied to the uninfibulated. This explains Hicks' remark that in societies where infibulation is practiced, it is incorrect to regard it as a social problem, since it is the uninfibulated that constitute a social problem. Her approach is based on a cultural analysis which recognises the centrality of infibulation to the lives of women in such societies, and the ostracisation likely to be faced by those who are not subjected to it. As such it marks a change from the sensationalism of many reactions to this custom, and moves away from the medicalisation that has characterised many responses to it.
In her conclusion she discusses the difficulties of effecting change in what she describes as 'closed cultural systems', where infibulation is 'part and parcel of the reproductive process' and the rigid control of women is a direct result of their central role in reproduction. She maintains that the practice cannot therefore be attacked in isolation from the intricate web of social relationships that it is so central to, an insight which suggests that infibulation will not be reduced, let alone eradicated without profound social transformation. It is this that leads Hicks to argue that in areas where the secularised state has virtually no impact on people's lives or livelihoods, means that strategy of using religious authorities to advocate the Sunna circumcision (excision of the clitoris, some of the labia and only slight stitching of the wound) over the more drastic pharonic type, may be the most effective way to effect change and reduce the appalling consequences of the latter for women's sexual and reproductive capacities. The disadvantage lies in the fact that this is likely to cement the link between Islam and mutilation, a link that has been claimed, but not fully institutionalised to date.
Knudsen (1994) provides a rather less systematic preliminary study of genital mutilation in her native Ghana, preferring to use the technically incorrect term 'female circumcision' to avoid the moral judgement implied by the word 'mutilation'. In fact she address a variety of practices, including the enhancement of female sexual organs by stretching as well as their diminuition by surgical removal or scarring. She objects to the fact that male circumcision is not also viewed as a form of mutilation. More disturbingly she draws a link between 'declining sexual morality' (for example teenage pregnancy) and the decline in these customs. Her documentation of the variations amongst Ghana's different peoples is nontheless informative, particularly given the absence of information on the subject in most African countries. Although Knudsen does not specifically address the issue, her information seems to suggest that in Ghana, female genital mutilation is practiced by traditionalists, rather than being conveyed by either Islam or Christianity, both of which seem to be linked to a reduction in the practice of female excision.
From this brief review we can see that the conditions under which female genitalia are excised, infibulated or otherwise modified differ from place to place. It is interesting that since colonial rule many such practices have declined in West Africa, whereas they have gained popularity or persisted in many parts of North East Africa, amongst Muslims and non-Muslims alike. It is also clear that cultural analysis of pre-Islamic and pre-Christian traditions in general could contribute greatly to our understanding of indigenous gender relations. Studies of practices which are inimical to women invariably reflect a desire for the practices to be discontinued, because they are physically, psychologically and socially damaging to women, and to women's capacities for sexual and reproductive fulfillment. One is left to ask, where is the research on traditions which empower women, which give them more control of their sexual and reproductive lives? Are there no surviving traditions which benefit women in ways not apparent within contemporary cultures and religions, all of which arrogate more power to men?

Virtually all other studies in the area of sexuality are linked to one of two major recent international concerns: population and HIV\AIDS. It is fair to say that women have most often been researched as perpetrators of both. This material has for the most part been funded by international agencies and directed towards preventative strategies directed at curbing population growth, or curtailing the spread of the HIV/AIDS pandemic. It is not therefore dealt with in this review, beyond noting that the availability of research funding in these areas is leading to a growing number of publications.
In addition to this, there are a number of international and African organisations, like the Global Women's Network on Reproductive Health, or the Black Women's Health Coalition, which have sought to bring a feminist perspective to bear on these important areas of international policy and intervention, and which have included African women in their activities to varying degrees.

Herstories: Giving Voice to Women
The genre which epitomises the methodological innovations that have accompanied the development of women's studies is one in which women's voices are given pride of place. These are best sited under the broad heading of cultural studies because of their anthropological antecendents and because they necessarily embody a holistic approach to women's lives which defies mainstream disciplinary boundaries. Women's accounts of their life experiences and daily practises, their histories and myths, their reflections on their past and present existences: the collection and analysis of these has been one of the major ways through which feminist-inspired scholarship has sought to redress the gender inequities suppressing women's voices and viewpoints. In social science, women's actual social subordination has often been hidden from view, naturalised and so compounded by androcentric theoretical frameworks and research methods. Having mounted major epistemological challenges to the imbuement of science with patriarchal values and perceptions, feminist-inspired scholars have made recourse to methods which give women a voice as a way of overcoming their absence from the more conventional archival, library and man-to-man fieldwork sources.
Several different methodological approaches are evident in recent studies of this genre. The first are in-depth, gender sensitive anthropological studies of specific communities of women (eg Early 1993), or men and women (Pankhurst 1992). Both of these examples started out with research aims that were abandoned in favour of others, aims which seemed more relevant once fieldwork had commenced, something which suggests an intellectual openess and willingness to hear and respond to what women have to say. Pankhurst (op cit) is more burdened with her own theoretical preoccupations than any of the others, and, although she admits to relying predominantly on biographical interviews with women as her primary source, displays a certain insecurity when she explicitly resists defining her work as being on 'woman and that kind of thing'. The 70 biographical interviews with women that she quotes extensively in her text are duly accompanied by 97 questionairre interviews mostly with men, use of her own field diaries and the usual archival and written sources. Her study of an Amharic-speaking village of Gragn in Menz seeks to unravel the relationship between the state and the peasantry in Mengistu's Ethiopia, and to examine gender relations in this context. Yet, lengthy contextualisations and background information notwithstanding, Gender, Development and Identity is primarily a study of the changes wrought in women's lives by government policy and changing economic conditions, a study which contains rich ethnographic detail, but at the same time provides an analysis of the relationship between the peasant community and Mengistu's state. It illuminates the failure of the state strategy of villagisation, the antagonistic relationship between a militarily powerful and extractive state machinery and the community, and the ineffecacious and tokenistic attempts to involve women in administrative and community structures. Pankhurst's research also reveals women's ability to negotiate and manipulate their restricted circumstances in order to attain the least unfavourable of possibilities. Divorce, invariably initiated by women, is identified as a major strategy deployed by women dissatisfied with their circumstances, with the result that divorce and multiple serial marriages are the norm in a society which is unfavourable to women, but which allows them some choices within the overall patriarchal culture.
Early's (1993) study of traditional, urban Egyptian women in the Bulaq Abu'Ala district of Cairo is a more focussed documentation of these predominantly low-income women's lives; one which deploys social observation techniques to gather and convey not only the activities but also the values, wits and multiple identities of baladi women, some of whom became close friends as well as indispensable informants. A fluent speaker of Arabic, Early spent 3 years observing and gathering material towards the doctoral dissertation which yielded Baladi Women of Cairo. Her theoretical concerns are much less cumbersome than Pankhurst's. Although both are anthropologists, and both use what can loosely be termed participant observation methods - noting other people's conversations and activities as well as asking questions - Early has none of Pankhurst' ambivalence over relying on women's narratives. She is thus able to convey a much richer and more nuanced order of information about the realities and concerns of her target group, one which credits them with a greater degree of agency. She concludes not with a tying together of theoretical concerns, but with a summary of what it means to be baladi, a meaning that has been developed through her study of the narratives and performances of baladi women, often through their own counterposition of themselves to the afrangi (westernised, modernised) Egyptians amongst whom they live and work.
At the other end of a continuum in women's studies are compiled biographies and life histories, all of which (introductory and editorial additions nothwithstanding) are published in the first person (eg Mirza & Strobel 1989, Russell 1990, Staunton 1990). Here the researcher's role, is given varying degrees of salience. Russell (op cit) is credited as the 'author' of Lives of Courage, her carefully presented compilation of South African women's stories. Mirza and Strobel (op cit) present the 'edited and translated' life stories of three women of Mombasa, gathered through interviews conducted in Swahili during Strobel's doctoral fieldwork. Staunton, a Zimbabwean who runs local publishing house, is the most modest of these researchers: she does not even put her name to the carefully gathered and presented stories of thirty Zimbabwean women's experiences during the liberation war. All 3 of these collections of biographies retain the orginal names and identities of the contributors, as well as photographs. Each woman is thus not only given a space in which to tell her story, but does so in full awareness that she will be appearing in a book for public consumption. They are thus part of the project in a way that is precluded by the conventional anthropological use of pseudonyms. Whereas the majority of the contributors to Staunton's Mothers of the Revolution are rural women probably not well known beyond their communities, Russell appears to have made an effort to include many of the famous women active on the South African scene prior to the demise of the apartheid regime, and her book displays accolades from the good and the great (Alice Walker, Nadine Gordimer, Archbishop Desmond Tutu and Oliver Tambo have all been delighted by it). Both Russell and Staunton's collections revolve on the particular conditions of the respective struggles of the people of South Africa and Zimbabwe, a fact which gives them focus and clarity. Mirza and Strobel (op cit) for their part, regard the presentation of three women's lives as informative enough in their own right, and make no explicit attempt to situate them in any given historical or political framework other than that implied by the somewhat diffuse tag of Swahili-ness.
The third variation in the women-speak genre combines the commitment to unleashing women's world views and experience with a commitment to reflection on these and a recognition of the importance of theorising on the basis of these realities (see Bozzoli 1991, Cock 1992, Mcleod 1991). Mcleod's is a rich examination of the lives and strategies of lower-middle class working women in Cairo, whereas the other two are South African studies of women during the apartheid era. All three books are outstanding in their analytical acumen as well as in successfully conveying women's contemporary history. None have any hesitation in asserting the value of listening to what women say, yet neither is this sanctified or assumed to be simple truth. Much attention is given to the social relations of the research process, and the epistemological importance of these.
Bozzoli's authorship of Women of Phokeng includes an acknowledgement of her main fieldworkers contribution: 'with the assistance of Mmantho Nkotsoe'. And so it should. Bozzoli points out that she does not speak or understand the language used by the women of Phokeng, and that as a white woman and an outsider, she would not have been able to attain the intimate level of rapport that Nkotsoe, a university-educated woman from a village neighbouring on Phokeng, was able to develop, particularly in view of both the context (under an authoritarian apartheid state) and the subject matter (women's often 'illegal' migrancy). Several pages are devoted to an exploration of Nkotsoe's relationship with the women she interviewed at great length. Not only does Bozzoli devote several pages to detailing the relationship through which all the material was gathered, but she also points out that at least we are looking at the English version of a Setswana original, which has to be better than attempting to interview in English, as is still common practice. She implicitly includes herself amongst the 'students of South African society who await the day when a new generation of fluent Bantu-speaking sociologists emerges, able to convey to the English speaking world what insights they gain from the analysis of the words of ordinary speakers of their own tongue' (1991: 12). Here I have to insist that white scholars should not merely wait on the laurels of their privileges, continuing the practice of using indigenes as insightful but dispensable assistants. It seems to me that the insights of African scholars can be and are being used for a great deal more than conveying information about their societies to the English-speaking world.
Bozzoli and Mmantho's study uses the oral histories of women born between 1900-1915 to investigate the consciousness, life strategies and movements of women between the so-called homeland of Bophutatswana and the cities. It is the inclusion of the dimension of subjectivity that sets Bozzoli's analysis of these women's lives apart from other studies of life strategy and migration. Whereas the early chatpers are devoted to chronologically organised narrative, the later chapters, while retaining the prominence given to women's voices, provides us with a detailed documentation of the process of politicization. It is a process during which women who consider themselves highly respectable find themselves being repeatedly subjected to intimidation and harassment by the authorities, harassment which leads them to develop a counter-counter-consciousness that resents and defies the persecutors, thus foretelling the demise of the apartheid regime.
This upholds Bozzoli's thesis that consciousness is organised around life strategies devised by women, but remain linked to, and depend upon the wider and changing material world in which they live, a world textured by race and gender relations as well as economic forces. The analysis put forward takes us beyond the caricaturised notions of South Africans, either as victims of apartheid, patriarchy and capitalism, or as heroic revolutionaries. It also transcends the determinism of orthodox materialist analysis, revealing the far more subtle processes through which individual identities are constituted and changed in the course of life, and may be collectivised and politicised, so beginning to resemble ideologies. By conducting her analysis from the bottom up - by grounding it in the life strategies of her target group - Bozzoli is also able to move beyond the structuralism of grand marxist and early feminist theories, and so make a major theoretical contribution to the study of consciousness.
Cock's (1991) study, Women and War in South Africa is original both in its subject matter, and in its methodology. Employing an unstructured interview method rather than the oral history method used in the Bozzoli-Mmantho study, Cock bases her analysis of the gender politics of the South African war of liberation on a comparative examination of the experiences of white women within the South African Defence Forces (SADF) and black women in Umkonto we Sizwe (MK). She begins by noting that both the whites and the Africans come from patriarchal societies lacking any tradition of gender equality, and in a somewhat contentious exercise, likens the apartheid regime, the Congress of South African Trade Unions (COSATU) and the ANC on the basis that none of them have women playing significant leadership roles.
While the numerical paucity of women in the leadership of all three forums is a matter of empirical fact, her comparison would not have withstood the substantial evidence that the ANC and COSATU, while they may be a long way from espousing feminism, have far more progressive gender ideologies than the apartheid state, as any reference to their congresses or the statements made over the years by their leadership would have indicated. Moreover her reference to the diverse African cultures is rather limited, since it is the Zulu culture propounded by Inkatha which she quotes in support of her argument that both African and white cultures, although patriarchal, contain something of a 'tough-but-submissive woman' tradition evident in the images of Voortrekker wives and Zulu mythology respectively.
Several other authors have drawn the comparisons between the gender ideologies of Zulu and Afrikaner nationalism, but these have stopped short of suggesting that the African liberation movement and the apartheid regime shared more than the most superficial similarities in gender politics (Gaitskell and Unterhalter 1989, Hassim 1993, McClintock 1991). The interview excerpts reveal a far more subtle complex of gender values and attitudes than is suggested by the rather global analysis of South African gender politics Cock puts forward as a prelude to her commendable study of the role of women in the armed forces of the apartheid regime and in the struggle against it. Here more space is devoted to the highly conventional gender dynamics of SADF than to 'the resisters' - MK and the End Conscription Campaign (ECC). The picture that emerges of SADF is one of a conventional military organisation with highly circumscribed gender roles, roles which rely on notions of men as 'protectors' of women, and which cannot therefore countenance more equal participation of women. The increased participation of white women in SADF in recent years is attributed to the exigencies of being a losing side, rather than any radical change. The interview material reveals something of the motivations of white women entering SADF: many of them are simply seeking a training ground or a short career prior to marriage, rather than having clear ideological commitments for or against the regime they serve. It is on this point that they differ from the MK women who come across as being politically aware, committed to the liberation struggle and highly motivated, as indeed must be the case for any person (male or female) volunteering to undertake training as a guerilla, and thereafter to participate in actions against the powerful military machine of the apartheid state. Neither SADF nor MK deploy women in combat roles, although perhaps because MK are ideologically committed to gender equality they do give women the same training (including the use of weapons and explosives) as men.
Cock's consideration is fascinating but brief: a more detailed study of women in the pre- and post-apartheid military forces remains very much in order. Her material is used to reconsider the confrontation between feminism and militarism that has been in evidence since Wollestonecraft first argued for women to take an anti-militarist position. A great many feminists have concurred with the conventional view that war is 'men's business', and that women should be kept out of military matters, given their 'special qualities'. As Cock observes, this persistent view is not one which has been espoused by many Third World feminists. For many of us, participation in armed struggles has often been seen as a necessary aspect of the national liberation movement, colonial legacies and military traditions of sexism notwithstanding, and revolutionary movements have (justifiably or unjustifiably) prided themselves on their inclusion of women (see Urdang 1989 on Mozambique's Frelimo, Wilson 199 on Eritrea's EPLF). Feminists remain divided into those who feel that equality means equal participation in the violence as well as the glories of war, and those who feel that women should stay out of this unpalatable aspect of human endeavour. Both schools of thought ultimately seek to build a world in which wars do not occur: the question is whether this can best be done by entering the powerful military establishment and participating in military activities, or remaining outside it, a question which Cock poses and explores through her material but, perhaps wisely, does not answer definitively.
Mcleod's Accomodating Protest (op cit) shares the methodological commitments of Bozzoli and Cock. Like Early (op cit) she has the advantage of being able to speak the mother tongue of her target group. She too is able to develop intimate relationships with women in the community she has targetted, and over a period of time, gathers oral data comprising narrative about various aspects of their lives. Her central interest is to explore why it is that today's young, educated career women have taken to donning veils, when only a few years ago, the same group of upwardly mobile women would not have done so. The answer she gradually uncovers for us is encapsulated in her title: it is about the way in which women delicately negotiate a way of acheiving their goals through the various contradictory forces that impinge on their lives. The dominant, Islamic ideal is one in which these mainly young women with children would stay at home as housewives like their mothers before them. But in todays's Egypt this is no longer an economically viable option, and nor is it one that educated women would wish upon themselves. Thye therefore do go out to work, but wear the veil: not because they are fundamentalists, but because they are breaching the dominant discourse. Furthermore, they do not want to be perceived as stooges of Western imperialism.
What distinguishes all these studies is the degree to which women's agency is revealed, something only possible through the open-ended, woman-centred methods that all of them deploy. It is a paradigm which takes the study of the oppression that all these women experience far beyond studies which leave one angered by the sense of passivity that is inevitably conveyed using more directive methods which impose theoretical frameworks on the subtleties, nuances and innovations that make women into makers of history rather than mere victims of it.



IV WORK AND ECONOMY

Conceptual issues
Studies of women's work have led to major reconceptualisations of what we mean by the terms 'work' and 'economy'. The fact that men and women perform different tasks in most societies has led scholars interested in gender to study the sexual division of labour. To say that there is a sexual division of labour means that not only do men and women perform different kinds of work, but that their labour is also differentially valued and remunerated in accordance with the gender of the workforce performing it.
A separate but related issue is the invisibilisation of women's work. It has long been apparent that a substantial amount of the work that women perform in most parts of the world has not been included in the formal definitions of 'work' or 'labour' as these have generally referred to waged employment in a formal economy in which men have predominated. Many of the tasks carried out by women in and beyond the household have thus been taken for granted, rendered invisible and devalued. The most obvious tasks included here are housework, child-bearing and childcare and food preparation, but in many contexts 'domestic' labour has also included household-level production of both food and goods.
In capitalist economies, work has been unjustly divided into productive and reproductive spheres, with production defined in economistic terms denoting production of tradeable goods and expropriatable surplus. Work concerned with biological and social reproduction has not been included in most macroeconomic calculations, except where these are performed as waged work. For the most part such work has not been renumerated, or if carried out as waged work, is so devalued that it is poorly paid. In short much of what has tended to be defined as 'women's work' has been unpaid, whatever its centrality to the functioning of the formal labour force, and indeed, of the society as a whole. This unequal scenario has been facilitated by gender ideologies: notions of maternal altruism, of wifely duty and of men's right to women's service and nurturance and to their reproductive capacities.
Whether this has always been the case in all societies has been the subject of much historical and anthropological study and speculation. Historical and cross-cultural studies have revealed that the sexual division of labour prevailing in Western capitalist societies is far from universal. Even here, it has changed radically over the generations, influenced by the advent of wars, economic booms and slumps and other macro-economic phenomena. Nontheless, at the ideological level, notions of what women should and can do have been unreasonably consistent: women's work has been taken to mean the unpaid labour of housework, childbearing, cooking and caring for children, husbands and other relatives, and a legion of other unremunerated domestic and reproductive tasks, most of which are performed in increasingly atomised individual households, headed by male authority figures who, for their part, are supposed to be the economic providers.
Research shows this scheme to be a patriarchal ideal which has seldom been attained by the vast majority of households within the West, and even less often achieved by households in other parts of the world, however much it has been extolled. Nontheless, because the ideological currency of this ideal has been so powerful, women in many parts of the world perform a great many unremunerated tasks, regardless of whether or not men are providing for them. Discussions of women's work cannot escape acknowledging the influence of this prototypical sexual division of labour on the actual conditions under which women work in and beyond the home and workforce.
African research into gender relations and work is particularly fascinating because of the enormous range of activities that African women engage in and the permutations that these have undergone with the changing circumstances of precolonial, colonial and postcolonial societies. Here we see that the sexual divisions of labour have been subjected to such major transformations that any notions of the immutability of what men and women can or should do are rendered nonsensical. Consideration of the activities that women engage in subvert many of the dichotomies that have been developed to characterise labour, notably those between formal and informal work, between productive and reproductive work, and between the household and the market.
Above and beyond all this, the research reveals the remarkable creativity with which African women, widely perceived as beasts of burden hopelessly locked into their exploitation, have addressed their situation, negotiated new terms in their relationships both with individual men and with governments, and have come up with innovative ways of ensuring the economic survival, not only of themselves, but of entire families and communities.

Women's Work in Precolonial Africa
Little of the recent gender research carried out on Africa has addressed work in the precolonial period in its own right. Set against the burgeoning literature on labour relations in the colonial period, the few available studies suggest a great deal more than they state. Most of the insights contained therein derive from observations about the impact of colonisation, often gathered from archives and records, journals and travelogues, and from oral histories obtained from