The challenges of doing gender research in developing countries : Focus on Malawi

We aim to illustrate how the tenor and character of the initial importation of the concept of gender into the Malawian context continues to impact, construct, and pose challenges to the study of gender in Malawi. Ultimately, we argue that complex and ethnographically grounded approaches to studying gender in Malawi may serve as a corrective to a long legacy of conceptual imperialism that has ramifications for both the researched and researchers in Malawi.


Introduction
In August 2008, a woman living in rural southern Malawi provided us with her own definition of gender.

Gender means there is no difference between women's work and men's work. Even though gender [in English] has come [yabwera] I still feel like there are other activities that women cannot do.
Her use of the English word for gender (amid the other words in Chichewa) and her suggestion that gender "has come" to Malawi signal that she, and many other rural Malawians, view gender as new and brought into Malawi from outside. To some degree, this characterization is not completely inaccurate. After all, it wasn't until the 1990s era of democratization and the accompanying influx of NGOs that concepts such as gender, human rights, empowerment, and democracy became prominent topics of study and intervention in Malawi. The importation of the concept of gender, entangled as it was and is in projects of liberation, empowerment, feminism, and human rights, has thus taken on a technical hue in Malawi, where the main thrust of much of the research and intervention around the topic has been to address the special vulnerability of women to social problems like AIDS or poverty and to alter traditional practices that comprise women's liberation, autonomy, or equality. Among other things, this has facilitated a chasm between gender theory and "the gender issue." While current gender theorists in the ivory towers of developed world universities tends to concern themselves with issues of gender performance, the links between gender and sexuality, queering feminism, or post-gender perspectives (Butler 1999;Butler 2004;Edelman 2004;Halberstam 2005), activists and those charged with incorporating gender into developing nation policies concern themselves with the "gender issue;" how to increase equality for women in societies deemed patriarchal and traditional, how to bring attention to violations of women's rights, or how to erode cultural norms of domestic violence, for example. Our dual challenge as feminists and researchers, then, is to find a way to imbue the analytic category of gender with power to bridge the gap between theory and the real world, to link gender as analytic category with gender as lived experience. Anthropologist Henrietta Moore suggests that we should ask what the fragmentation, ambiguity, and multiplicity [of gendered identities and the term gender itself] are doing for us as social scientists and for people in their lives (Moore 2006: 27). This paper draws attention to and makes gestures toward closing the gap between "our" world of theory and "their" world of practice.
We aim to illustrate how the tenor and character of the initial importation of the concept of gender into the Malawian context continues to impact, construct, and pose challenges to the study of gender in Malawi. Ultimately, we argue that complex and ethnographically grounded approaches to studying gender in Malawi may serve as a corrective to a long legacy of conceptual imperialism that has ramifications for both the researched and researchers in Malawi.

Method of study
The paper is exploratory in nature and draws on textual analysis of archived reports at Centre for Social Research in Malawi and Malawi National Archives, the experiences of one of the authors as a consultant on gender-centered projects in Malawi and a sociologist whose own research is centered on gender, as well as the experiences of the other author who had been doing PhD research in Anthropology for 16 months in Malawi between 2007-2008. In the US and other developed nations, gender has become a central focus of academic inquiry and its study has, in fact, been institutionalized into universities through the rise of women's studies or gender studies programs and emerging sub-fields such as sexuality studies or queer theory. Although gender is a topic included in the curriculum at University of Malawi, its main institutionalization into the southern African nation has been through the work of NGOs or development organizations who have struggled to bring gender to rural villagers through "sensitization" or "social mobilization" programs and to mainstream gender into national policies and programs. It should here be noted that the institutionalization of gender as an analytic category globally has been made possible as a result of the gains of the feminist movement. The history of this movement can be briefly summed up as a progression from making explicit the differences between men and women, then the differences among women, and, finally, the differences within the category woman, always with a commitment to the liberation, empowerment, and equality of women. For many reasons, the developed and developing world do not seamlessly share conceptions, definitions, or interpretations of gender. Instead, translation of the concept across both the theoretical/technical dichotomy and across cultural boundaries is fraught. For example, Mukhopadhyay and Wong, writing on how the theoretical concept of gender was translated into development (technical) initiatives, argue that the development apparatus adopted only those aspects from gender theory that were tangible, measurable, and observable (Mukhopadyay 2007). Thus, conceptions of gender in the developing world have tended to focus myopically on the division of labor and access to resources, while the important symbolic and complex ideational processes of thinking gender and performing gender went by the wayside. Similarly, many have pondered the applicability of the concept of gender and of 'feminism,' more generally, to African contexts (Molawole and Modupe 1997). However, while many have illustrated the diversities of feminisms possible across different geographic and cultural spaces (Kaplen 1999;Phillips 2006) and critiqued Western feminism (Mohanty 2003), there is a paucity of work exploring the translation, articulation, and diversities of interpretations of gender across such contexts. However, the term is a central concern of the Millennium Development Goals (MDGs), international development policy, and NGO priorities, all concerns that seem to call for a clear understanding of its meanings and valences.

Gender yabwera: The importation of gender into Malawi
An examination of the archived cascade of reports and executive summaries produced by the University of Malawi's Centre for Social Research, usually in collaboration with NGOs or foreign universities, illustrate a concern in the early 1990s about such issues as "including women in poverty alleviation projects," and a consideration of methodologies such as women-directed income generating activities and women only cooperatives (Gaynor 1991). Documents from this time period illustrate that a major concern of the authors of the reports was providing NGOs with steps to take to ensure that gender was analyzed and that women's participation was encouraged in all projects. With the 1984 establishment of the National Commission on Women in Development came a subsequent widespread call for the incorporation of gender awareness and gender planning in the program design of social action and poverty relief schemes (Chilowa 1992). Much later, a 2002 gender research training manual points to a "lack of gender mainstreaming in the research process" and frames its pedagogical role thusly: "to assist participants to think through how they can introduce gender to their organizations" (Chipeta 2002). Overall, however, a close reading of these documents draws our attention to two main trends. First, the documents universally endorse the necessity of incorporating gender in programming, policy, or plans but none of the authors define the meaning of the term "gender;" perhaps it is assumed. Second, the language used in these documents tends to portray the national context of Malawi as a clean slate, devoid of gender itself.
From the early days of the institutionalization of the social sciences in the new postindependence national university, national and international researchers working in Malawi have targeted their research toward problems of national development and "relevance;" in his speech to inaugurate the new university, then president Kamuzu Banda drew a sharp line between ivory tower research (deemed inappropriate in the Malawian context) and research to impact national development ("relevant" research) (Holland 2006). This legacy of "policy relevant" research persists to this day in Malawi, where much of the research focuses on solving practical problems that exacerbate human suffering and, especially, poverty. The introduction of the concept of gender in Malawi is also tied up with policy and programming to alleviate poverty. An emphasis on the unfair distribution of material or other resources both among and within nations led many in the 1990s to call for attention, also, to how this unequal distribution might make women especially vulnerable. Studies of gender in Malawi and other developing countries at this time focused on the gendered division of labor and resources, or material aspects expressed as inequality between the genders. Interventions staged to address this disparity tended to propose income generating activities for women. However, the insights of gender theorists that gender is a social relation, something produced by culturally shared conceptions and ideas about men and women, also permitted activists and practitioners in the developing world to imagine that social relations between men and women were mutable and open to change.
As both poverty and AIDS represent key contemporary points of consensus among donor agencies and national actors in Malawi, the main methodological approaches to the study of these issues has continued, since the early post independence years, to take such forms as: commissioned research, situation analysis, baseline studies, and monitoring and evaluation; such studies provide a quick and easy approach to constructing the reality against which policy gaps are identified, solutions are proposed, and progress is tracked (Holland 2006: 161). The dominance of these methodologies has generally meant that another rising concern of the international community, gender, has become merely an appendage; gender becomes a category of specific vulnerability or another lens through which to view larger issues or social problems. However, unlike poverty or AIDS in Malawi, gender has never been the body, meat, or unique subject of a research project; its meaning is assumed even as it is appended again and again to proposals or policies. In 1995 at the UN Fourth World Conference on Women in Beijing it was agreed that gender equality is a primary goal in all areas of economic and social development. In 1997, following these discussions, the following definition of the strategy towards this end (called gender mainstreaming) was adopted: Mainstreaming a gender perspective is the process of assessing the implications for women and men of any planned action, including legislation, policies or programmes, in any area and at all levels. It is a strategy for making the concerns and experiences of women as well as of men an integral part of the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres, so that women and men benefit equally, and inequality is not perpetuated. The ultimate goal of mainstreaming is to achieve gender equality." 1 This emphasis on mainstreaming signaled a shift in how development looked at women. In the early 1970s when the term "Women in Development" was coined, the focus was on integrating women in projects, and the main problem identified was women's potential exclusion from such projects. In the 90s, when gender mainstreaming became mainstreamed, the emphasis shifted to reflect a move away from segregating women as a separate group to focusing on how women fit into their societies, or how relations between men and women are socially constructed (reflected in the adoption of a new acronym GAD-Gender and Development). However, although the shift from WID to GAD necessarily complicated the ways people approached development, the supposed global aspirations of the gender mainstreaming strategy still raise a number of questions. What is gender in Malawi? Does gender inequality mean the same thing in the US that it does in Malawi? Is access to the "concerns and experiences of women as well as men" obtained in the same way in Fiji as it is in Germany? How do we as social scientists begin to answer these questions? In some ways, international policy and the localized programming that emanates from it have put the cart before the horse, implementing and doing before understanding local and potentially disparate definitions, translations, and conceptions of gender.
Gender mainstreaming requires the movement of gender into the interstices of a national context that, contrary to discourse that frames "gender" as completely foreign to Malawi, was and is already laden with meanings and interpretations of the sorts of things we might associate with the term gender: conceptions of proper relations and exchange between the sexes, proper comportment for men and women, issues of female modesty or immodesty, traditional reproduction of proper social roles for men and women, and division of labor at the household level. With the emphasis on gender mainstreaming in international and national Malawi policy, "gender trainings" have become a common event. It should be noted that even this seemingly technical, neutral phrase itself smuggles in the assumption that Malawians are, until trained, ignorant of (a certain kind of) gender. These trainings target audiences ranging from ministers in government, workers at NGOs, to rural villagers. Typically this involves staff from an organization coming for a day or two to 'teach' people about gender. As Mukhopadhyay and Wong, again, point out, 'training' or 'teaching' is seen as the main way to increase gender awareness and change (presumably bad) practices. But even as people everywhere are teaching gender, there is little focus on the content of the concept itself, on its meanings. Instead, the desired outcomes are the endpoint; e.g., how many people across the country were 'sensitized' on gender in a given time period? This illustrates anthropologist Sally Engle Merry's point that experts often take a thin view of things and are not concerned with the particular (that is: what does gender mean in Malawi, what context is it entering into?) so much as figuring out how to apply universal standards to a place. Merry argues, and we agree in the case of Malawi, that transnational ideas become translated at a superficial level, as mere "window dressing" (Merry 2006b). Anthropologist Lila Abu-Lughod, writing on the way in which television soap operas take on meaning to audiences in contemporary Egypt, suggests that 'viewer's responses [to TV programs] are dictated by often vast distances between realities dramatized in serials and people's everyday lives' (Abu-Lughod 2005: 237). We might view the performance of the gender trainings themselves in much the same way as Abu-Lughod's television soap operas; these are scripts that may have little resonance with everyday practice and reality. Writing on gender training in India, Dasgupta points out that a model that necessitates cost effectiveness or quick, observable outcomes ultimately defeats the critical potential of feminism or gender (Dasgupta 2007).

Researching gender in Malawi: Some experiences
In the above section we have provided a brief summary of how the concept of gender entered into the Malawian context. In this section, we provide some ethnographic examples and stories from our combined experiences conducting research in urban and rural Malawian to show how the triad of poverty, gender mainstreaming, and quick and easy research methods have served to reproduce a vacuous, empty notion of gender that fails to engage with or incorporate diverse and complicated local ideas around this concept.
Early and continuing emphasis on research that will alleviate poverty in Malawi, has tended to view gender as one possible category of vulnerability. Often this framing depends on the characterization of Malawi as a patriarchal, traditional society whose culture disadvantages the earning power or economic capacity of women. This has contributed to the widespread and authoritative assumption that women in Malawi are vulnerable or, for example, engaging in sexual intercourse for financial gain, or enmired in poverty. In general, this has led to the production of a vast body of research on women in Africa that might be termed, "depressing research." In this regard, much research on the African continent has centered on the vulnerability' of women and girls to gender based violence, HIV infection, lack of decision-making power, and the general disadvantaged position of (apparently) all women in (apparently) all circumstances (see, for example : Carrillo 1992;Heise, Pitanguy, and Germain 1994;WLSA Malawi 2002;Liwewe and Matinga 2005). While we do not wish to argue that women enjoy high economic mobility and power in Malawi or trivialize the real burdens that they face, we do encourage the reader to reconsider the portrayal of Malawian women as wholly and only vulnerable. Our analysis of the 'depressing' portrayal of the position of women may lend support to Ifi Amadiume's suggestion that anthropologists and others were complicit in reproducing and legitimating a picture of Africa and African women that reflects their own conceptions of class, feminism, and patriarchy (Amadiume 1997). More recently and specific to the Malawian context, sociologists Susan Watkins and Linda Tawfik, suggest that in policy centers like Geneva and Lilongwe, women's adultery, for example, is automatically interpreted as resulting from poverty. However, in their research in Balaka, a central matrilineal district in Malawi, they found that adultery was a strategy by which rural women aspired to higher standards of living, or took revenge for their husband's initial infidelity (Tawfik 2007). We argue that perhaps traditional methods of data collection and the tendency of research to replicate authoritative knowledge (Moore 2001) have tended to unproblematically accept that women are powerless, vulnerable, and living in a male dominated society. Furthermore, in the absence of emphasis on ethnographic or intensive research of gender itself, it is essential that we interrogate the affects, assumptions, and positioned experiences that prompt us to uncritically accept, for example, characterizations of Malawian women as always only powerless, poor, or vulnerable.
Another example that brings Amdiume's critique of the complicity of the researcher in constructing facts that reflect one's own positionality to life is drawn from the experience of Ntata, Bisika, and Konyani (2005) working on a commissioned study about gender based violence in Malawian primary schools. A collaboration between UNICEF, an NGO, and the Centre for Social Research, the study hypothesized that one reason why girls underachieve in Malawi is due to gender violence. The study sought to identify sources of violence, to understand girls' definitions of violence, and to ultimately intervene on the issue. Four main issues emerged out of this gender based violence research project that can inform our consideration of methodology and theory in this paper. First, the study began in the context of emotive media propagated images and representations of the school teacher as the main culprit in terms of violence against school girls and the powerful and widespread image of the sugar daddy with a young girl in tow. However, it emerged from the study that not only was the assumed perpetrator inaccurate (girls felt more threatened by their male peers in school), but the manner in which the girls thought of violence itself was different from the typical notion of GBV which tends to focus on rape or physical attacks. The girls themselves defined violence much more broadly and in a way that reflected their every day life paths; for them violence could be bullying by boys, verbal abuse, teasing, etc… Of course some might say that instances of horrible violence or violence by teachers might be underreported in a study like this one; but we might also interrogate our devotion to proving the things we set out to prove, the way in which we hold on tightly to facts that might be believable, but maybe not a good match with reality. In addition, the study focused on 'girls' as a category, without considering the sheer diversity which might fall beneath it. For example, girls told the researchers that older women in their community looked at the 'violence' boys were committing in school and saw it not as violence at all, but simply as part of growing up, where boys 'play around' with girls. In this way, we note that definitions of 'violence' are hardly stable or uniform even within a very small community.
Finally, we point to this study as one of many that attempted to understand gender without soliciting participation from males. Interesting, as well, is that this study found that there was a large amount of violence against boys by adults in the schools (notably not in line with what we traditionally think when we hear the term); this aspect of violence was left unexplored, however, since the study focused exclusively on females. A standard global assumption that gender means "women" and that only women can be victims informed the study proposal. Examples such as this one show us the importance of coming to terms with assumptions that become facts by virtue of their believability, the importance of allowing a relevant definition to emerge from the research instead of imposing one on the participants, the centrality of recognizing that even within a small community, women may not share the same ideas about gender and the need to include males in studies that aim to explain something about gender. In short, this example illustrates the pressing need not for more studies that assume a meaning for gender in the Malawian context, but for studies that seek to characterize meanings and interpretations of gender both from and with meaning to the local context and community.
Gender mainstreaming, as mentioned earlier, is a strategy that depends fundamentally on the assumption that "gender" is foreign to or has not yet taken root in Malawi. What implications might this have for how local people assess, evaluate, and assign moral meaning to the concept of gender? Anthropologist Ulrika Ribohn points out that cultural practices in Malawi are often gendered in human rights discourses, meaning that women are identified as the carriers of traditions and that these very traditions are viewed in the postcolonial era by official agencies and, in some cases, the state, as a threat to women's or human rights (Ribohn 2002: 166). In a place where cultural actions and culture itself have been characterized so strongly as inimical to human rights, any change in the roles, status, or actions of women might be interpreted as a product of outside influence or western ideas.
For example, in Malawi, questions of female identity are very much intertwined with material benefits and this prompts many women to fear breaking away from or challenging the dominant cultural, kin based, discourse of respect and proper womanhood. The powerful opposition between culture and human rights or women's rights is also an opposition between the Malawian and the western, and fits well into stories that both authors have heard from chiefs and villagers across Malawi; that youth these days are corrupted by western values and that local systems of order and respect are unraveling at the hands of outside influence. Whether or not these stories are mere precipitates of nostalgia matters little; they are widely recited and, thus, inform the ways in which concepts like gender or women's rights enter local settings in Malawi. Ultimately, the concept of gender itself and its increasingly wide circulation at the local level may contribute to women's disadvantaged position by prompting men and women alike to include it among the very thing they feel their way of life to be most threatened by: westernization. In the process of conducting interviews with men and women in Balaka and Salima districts, one of the authors often found that her respondents had a lot to say about gender. One common trend across almost all her conversations with these rural Malawians was their tendency to associate gender with human rights and democracy. As with any society, the range of moral interpretations of these terms varied. However, on the whole, people tended to view both human rights and democratization as "from the West" and as threatening to local ways of life. For example, people recounted many stories involving the decline of community policing of witches and bloodsuckers as a result of "these human rights." Instead of feeling protected or safe, these rural dwellers claimed that human rights enabled bad people like witches to "do whatever they like, without punishment." This evolution is commonly described by villagers, and captured in the words used by a 43 year old woman in Mchinji: "Eee! [Human rights] kusongoneza kwambiri (they are very disruptive). It is essential that we consider the ways that people assume a similarity between concepts that were imported into Malawi around the same time and in the same modes (i.e. sensitization, social mobilization, or trainings), because the morally inflected interpretations of any one of these terms can thereby influence interpretations or evaluations of any of the others; all are deemed to be "foreign" to Malawi, a fact made especially apparent when Chichewa (vernacular) speakers employ the English words for all three concepts.
Another element we might consider is the fraught history of foreign involvement in Malawi, which progressed (in summary) from an era of colonial indifference toward NGOs and other outside organizations to an era of severe suspicion or mistrust toward NGOs during Banda's rule to a laissez faire attitude following democratization. Malawians who grew up during the era of suspicion toward outside organizations, and still hold fond feelings for the late Dr. Banda, may still hold deep seated suspicions of these kinds of organizations and the messages or ideas they are bringing to villages.
Anthropologist Pheng Cheah suggests that human rights NGOs, for example, have come to see their major task as overcoming or fixing the ignorance of the rural poor, even though this process of consciousness raising or sensitization emerges out of the same system that exploits the rural poor and serves to exacerbate social inequalities and knowledge gaps already in existence (Cheah 2006: 167-68). Donna Murdock, another anthropologist, points out that funding structures and the dependency of civil society organizations on donor monies have, in some contexts, facilitated competition between so-called 'local' women and so called "NGO women" (those who are educated and can write better grant proposals), usually to the tune of the marginalization of the former and, thus, the fragmentation of any sort of idealized locally grown feminism (Murdock 2003: 524). This kind of competition and the drawing of so-called grassroots organizations into funding structures with prescribed goals can also serve to reproduce and replicate concepts, such as gender, without querying their utility at the local level. As local groups struggle to stay afloat through accessing funds from international donors, they tend to pay little attention to the meanings of the terms that have now become impossible to leave out of proposals that seek to be successful and may feel significant ire or jealousy toward or distrust of more elite women who may end up coming to "teach" them about gender. This can lead to gender becoming a concept with empty or overflowing meanings, rendering it useless as a basis for either social change or theory building.

Conceptual imperialism
We think the phrase "conceptual imperialism" very effectively captures the way that outsiders assume that people in the developing world can be easily taught and subsequently expected to embrace concepts that may not resonate with their lived experiences. Tied up in this is also the assumption that any context we enter into, whether with our agendas for training or our tools, theories, and paradigms for collecting data, is a clean slate, devoid of already circulating, vibrant discourses on topics and concepts that can only be imported by those with the capacity to empower, i.e., by conceptual imperialists. At an international conference, for example, one Westerner claimed that Africans are unable to hypothesize and concluded that it is thus difficult for them to study science or apply scientific methods.
In the long tradition of those who have drawn attention to diverse kinds of rationality, reasoning, and philosophizing (Evans-Pritchard 1937[1976; Karp 2000;Verran 2001), we can only conclude: There are different ways of hypothesizing. (At another level, the notion that only the western type of science is science could also be challenged). Conceptual imperialists also tend to gloss over or ignore the material factors that may motivate forms of activism or interest they label as 'local' or may prompt people in certain contexts to 'buy into' a concept they care little about in order to access funds. By eliding these fractures and ruptures between theory and practice or the global and the local, conceptual imperialists may succeed in furthering their own objectives or goals (for example, by forging transnational networks, fulfilling quotas or indicators for gender training sessions, or international feminist groups) but, in the process, impose these goals or terms on developing country partners, activists, or academics.

Reconceptualizing agency, and a call for study of the mundane, lived experience of gender
Saba Mahmood, in her book that considers the politics of piety, or the possibility of feminist subjects within cultures of Islam, writes that it is hard for, "us [as academics] to see and understand forms of being and action that are not encapsulated by the narrative of subversion and the reinscription of norms" (Mahmood 2005: 9). Judith Butler, prominent feminist theorist, sees gender norms as produced by language itself, leading to the enactment of normative heterosexual or male/female roles by members of a particular society (Butler 1999). For Butler, agency lies in the fact that norms are produced and, therefore, malleable and open to change at any time by any social actor; this is a departure from traditional feminist conceptions that agency or resistance lie only in emancipatory politics, for example. However, we return to Mahmood's account here, as she builds on Butler's focus on the fact that norms can be reconfigured; Mahmood argues that we should pay more attention to the ways in which norms are lived (not the ways in which they are changed). In short, she asks a question that we find especially relevant to helping us, as people interested in using gender as an analytic lens in the developing world, to recognize the way in which certain paradigms and assumptions dominate our consideration of gender and women's agency: "How do we conceive of individual freedom in a context where the distinction between a subject's own desires and socially prescribed performances cannot be easily presumed, where submission to certain forms of authority is a condition for achieving potentiality?" (Mahmood 2005: 31). Although highly theoretical, we find this question to be an especially useful one for us to ask when considering how it is that we might move away from authoritative, well circulated notions of 'powerless African women' or 'vulnerable women living in the shackles of patriarchal societies.' We mean to point, here, to the way in which the term gender itself (and the string of other words that enter our minds through association: empowerment, women's rights, feminism…) has been presumed to be linked to revolution, resistance, or drastic overturning of social roles. The applicability of this revolutionary narrative and of 'feminism,' more generally, to African contexts is under debate (Kolwawole 1997).
Though we want to avoid falling into the trap of painting a picture of the typical "African" woman, we do wish to point out that the things African women might consider 'worth fighting for' (that is, their values) may relate more to being a mother, to the bonds of family, and to status accrued throughout the life course. Typically, many Western women would not value these aspects in the same manner. Without devaluing the main goals of international feminisms or suggesting that things in Malawian villages or other villages in other countries should just remain as they are, perhaps Mahmood's insights might help us to think about novel, yet unexplored avenues of researching gender. Maybe we should take our focus, for a moment, off the "prize" (say, complete reorganization of patriarchal societies) and start thinking up really good methods for studying the more mundane: how gender is lived every day. We provide two anecdotes here to illustrate Mahmood's focus on lived norms, on the mundane and everyday, on agency within relations we might term "oppressive." Ntata and Kamoto (2004) were working on a research project in Zomba District came across a female chief in Chingale area and noted that she had accrued immense power and respect from her subjects, was judged by all to be fair and competent, and made decisions for her village. Indeed, in Malawi there are still a number of villages where women are chiefs. Of course, these well respected female chiefs illustrate the simple point that capabilities for leadership and for acquiring power and status lie within both men and women. However, this example also points to the fact that a complete overhaul of social structure is not always necessary to "fix" gender inequality, even if our attachment to narratives of revolution is powerful. The female chiefs index the need to appreciate and capitalize on the positive things already happening on the ground in Malawi instead of overlooking them to focus exclusively on negatives regarding women in developing countries.
In another case, Ntata (1998) was working on a UNICEF-funded research project in Ntcheu District where some of the questions asked women "Who [in your household] makes decisions on A, B, C, D?" A very common response from the women was, "My husband and I make them jointly." When they were pressed by the researchers ("I thought men make the decisions here"), however, the women said: "They think they do. To avoid conflict, you need to do things in a way where they think they have the power, where they think they have made the decision." In this example we see a far more sophisticated model of agency within everyday life; rural women's understated ability to mediate gender imbalance might be more successful than notions of open confrontation, challenge, and revolution. As researchers committed to employing complex notions of gender in our work, we must look beneath the surface and question our own conditioned ways of seeing.
In her edited volume entitled "Feminism and History," prominent feminist theorist and historian Joan Scott suggests that, "The desire to legitimize feminist claims about women in order to consolidate an effective feminist political movement treats 'women' uniformly and so ahistorically. But the creation of women as subjects of history places them temporally in contexts of their action, and explains the possibilities for such action in terms of those contexts" (Scott 1996: 4, emphasis added). Though Scott's focus is on historical differences between categories of women (she argues, for example, that we can hardly find common identity between religious women of the Middle Ages who sought transcendence of their bodies and twentieth century sex workers whose bodies serve as a source of income), we can extrapolate her insights about the problematic of common identity to the plethora of identities for women that exist alongside each other in our global world; obviously the material and social interests and contexts of, for example, a transgendered person in San Francisco and a woman at the borehole in rural Malawi are not aligned and, in fact, these identities are possibly mutually unintelligible to each other. We highlight the obvious point that many have made that "woman" is not a unified or essentialist category (Nnaemeka 1998) not to rehash old terrain, but to suggest that in a globalized world where the gendered identities available to both men and women are proliferating, a keen and contextually informed understanding of the term 'gender' and the multiple factors that color its local and global interpretation is essential to effective research, policy, and programming.
Finally, drawing on Scott's point that privileging or viewing as the main endpoint the creation of a feminist political movement can lead us to gloss over or minimize crucial ruptures between this goal and lived experience in local contexts, we suggest that research on gender could benefit from methods that allow us to access the very spaces and contexts we enter into with our preconceived ideas about gender or whatever else. As well, we should be mindful of the association and complex entangling of gender and a feminist, activist project. Without taking away from the laudable goals of the latter, we boldly argue that detaching gender from a narrative of liberation, at least momentarily, may help us fill it up with local meaning that will, ultimately, provide great analytical purchase in Malawi. As many have argued, locations are not just spaces but crossroads of political, economic, and social forces that have resulted in the production of a context that is complicated and messy.

Conclusion
We return again to the insightful words of feminist historian Joan Scott, who writes, "We [must] recognize that 'man' and 'woman' are at once empty and overflowing categories. Empty because they have no ultimate, transcendent meaning. Overflowing, because even when they appear to be fixed, they still contain within them alternative, denied, or suppressed definitions" (Scott 1996: 174). We suggest that one of our tasks as social scientists of and from the developing world is to challenge or improve taken-for-granted and internationallylegitimated definitions of gender by uncovering and bringing to the fore the multiple definitions which remain hidden but extremely potent within our local research contexts.
Thus far, we have illustrated the historical development of gender as an imported concept in Malawi and highlighted some of the difficulties of interpretation which result. We want to close this paper by summarizing four more pragmatic challenges for those who research gender in Malawi and, most likely, other developing countries. These challenges will inform our quest for a more robust theory of gender that emerges out of the Malawian context.

Research in desperate conditions
First and foremost, we want to be clear that all research in a developing country is complicated and fraught with ethical issues. Namely, a long history of research projects and the poverty that characterizes Malawi have prompted many participants in any kind of research project to adopt expectations that often do not align with those of researchers (Biruk forthcoming). Most basically, participants typically expect money or compensation for their time, but, specifically related to research on or about gender, is the potential expectation that things will change as a result of shared information. If a woman shares her personal stories of domestic abuse with a researcher, she likely assumes that help is on the way when most likely it is not. In rural areas of Mchinji, Balaka, and Salima, one of the authors spoke with hundreds of villagers who all assumed that if research was being done, it meant that help was on the way. Maybe next year, they would have a new school, or a hospital, or more fertilizer. Unfortunately, in most cases, help wasn't on the way. Also in the vein of research ethics, research into topics such as GBV or domestic violence or rape (common across sub-Saharan Africa) necessitate the implementation of a system of protection for respondents, who could face backlash for talking with outsiders about unsavory issues. Because gender is associated with social change and is seen as something 'new' in Malawi, it also has the potential to anger those who stand to lose if injustices in their community are brought into the open.

Marginalization of social science and qualitative methods
At a conference where researchers disseminated their findings held by the Malawi National Research Council in November 2008, very few of the papers presented utilized social science methods, and those that did had their 'significance' or 'representativeness' questioned (both authors attended this conference). In general, Malawian researchers are locked into a paradigm that privileges scientific, quantitative, and 'practical' research. The marginalization of the social sciences combines negatively with the fact that much of the research that happens in a place like Malawi is donor-driven. This often leads to a preference for quick 'in and out' studies that must adhere quickly to a set timeline for data collection, one that tends to privilege quantitative data collection methods. Furthermore, questions are usually already identified, as opposed to emerging inductively from a context. This leads to the paucity of research on gender itself; such an endeavor would likely be deemed policy irrelevant or too costly.

The lack of 'home grown' theory
Although donor driven research likely contributes to the compromised intellectual autonomy of local scholars and research (Zeleza 2003; see Holland 2006 and Biruk forthcoming on the rise of consultancy as a means of survival for faculty at the University of Malawi), it is hardly the only issue here. In addition to developing methodologies for alternative research or building new, locally grown theories, African researchers must also gather and analyze data in a manner that will satisfy the basic scientific requirements of validity, reliability, and replicability (Steady 2005: 321). Often, this disallows them to propose new directions for study, or to stray too far from the definition of gender already so entrenched. Finally, their challenge is heightened by the fact that "Africa" is diverse in culture and contexts; any theories or methodologies have to be adjusted for contextual validity and the potential "non-universality" of a theory of gender in Malawi likely discourages international interest in funding such research.

Lack of a complex notion of gender that is attuned to reality
Fundamentally, research on gender has suffered from the lack of a more compositional model of gender, meaning one that focuses centrally on the social relations between men and women, on the fact that gender is not just about women. In Malawi, also, gender seems confined to the village; it is often assumed that only the poor or those who live in rural areas need to be enlightened about gender or suffer from health or social problems that emanate out of gender inequality. In reality, Malawi is much more complex-obviously, there are rich people, poor people, urban people, rural people, men, women, different ethnic groups. This ties in with the point that the assumed universal subordination of women ignores ways in which social location is also based on race, ethnicity, class, and color.
Even as we focus on problems as the entry point for studying gender in the developing world, there are many problems with the study of gender itself in the Malawian context. We started with the assumption that we would be writing about the challenges of doing gender research in Malawi, but in the course of formulating this paper we both came to realize that the field has yet to emerge. So the greatest challenge we describe in this article is with regard to introducing well theorized and methodologically innovative research focused on capturing the diversity of lived experience of gender in this context.