Contextualizing gender in climate change adaptation in semi-arid Zimbabwe

Over the past two decades, climate change has increasingly become recognized as a threat to sustainable development, with specific threats on various sectors such as health, agriculture and water, tourism and energy, among others. There are specific impacts on household and community livelihoods, especially in Sub-Saharan Africa, where nations and communities tend to be vulnerable to the effects of climate change (IPCC, 2007, 2014). However, these impacts tend to affect different socio-economic groups in different ways, making gender critical to understanding the way that climate change is experienced. Gender dimensions of vulnerability are hinged upon differential access to resources that are crucial for adaptation (Kleiber et al., 2015). Socially constructed gender differences have a bearing on the extent of exposure to climatic shocks, leading to various patterns of vulnerability to these shocks (IPCC, 2014).

While there are new findings to the differential impacts of climate change by gender (IPCC, 2014), there appears to still be limited literature on gender and climate change adaptation. The bulk of the literature in this regard highlights women as more adversely affected by climate change, yet they appear not to be part of the decision-making processes and bodies and not to possess virtues of environmentalism, especially in the African context (Arora-Jonsson, 2011; Denton, 2002, 2000). Predominantly, literature on gender and climate change highlights the rural woman in the south in two ways: that this woman needs special attention because they are highly vulnerable and that they have a high mortality rate in times of natural shocks driven by climate change (Arora-Jonsson, 2011). These conceptualizations tend to advance the notion of ‘feminization of poverty’ that is problematized by several scholars, given that it tends to erroneously blanket female headed households in whatever context as poorer than other households (Chant, 2010; Sen and Östlin, 2008). This problematization emanates from acknowledgement that the thesis has been instrumental in garnering resources for women, yet there is seemingly lack of rigorous and convincing gender disaggregated data and the existence and extent of all contributions to household income that substantiates the thesis of ‘feminization of poverty’ (Chant, 2010; Kabeer, 2008; Marcoux, 1998).

While there is evidence in the past two decades of these emerging perspectives on gender and climate change adaptation that appear to challenge the dominant narrative of women as vulnerable (Harris-White, 1998; Jackson and Wolinsky, 1996; Kabeer, 2008; Premi, 2001), the work in this paper to contextualize gender remains relevant in this regard given that some of the most recent research still indicates gender discrimination and inequities in adaptive capacity (Araujo, 2008; Carr, 2008; Denton, 2000). Arora-Jonsson, (2011) equally emphasizes the importance of context and embeddedness in gender analyses. This approach tends to challenge the universality of the discourse in which all women are portrayed in the same way. The emphasis on vulnerability, and the portrayal of women as victims, denies women agency and can result in policies that are focused on meeting short term needs with no potential to substantially improve women’s capacity to adapt to change (Dankelman, 1992).

This paper establishes the extent of variations between perceptions of men and women regarding changes in climate. Perceptions tend to be an important component of households’ attitudes and practices in relation to climate, given that adaptation is to a significant extent hinged upon perception of the extent of threat posed by climatic occurrences (Mubaya, 2009). A growing body of literature exists that differentiates women from men in relation to understanding the environment broadly (Bord and O’Connor, 1997; Davidson and Freudenberg, 1996; Gustafson, 1998) and within these variations, what are the indications in terms of vulnerability of females versus that of males. We also focus on the local realities of responses that contribute to both coping and adaptation by gender as opposed to starting from the universalism within this discourse of women as disadvantaged and therefore more vulnerable.

In the 1950s and 1960s, women’s activism in Africa was linked to nationalist struggles for independence, given that race and class relations were already integral to struggles African women were engaged in (Lewis, 2002). From the 1970s, there was a proliferation of women centered research in Africa (Lewis, 2002; Mama, 1995, 1996; Manuh, 2001; Nzomo, 1998; Pereira, 2002). This research falls within the Women in Development (WID) context that was later spearheaded by Boserup around the 1970s. Among the factors contributing to this proliferation was the international women’s movement and the emergence of state feminism” (Mama, 1996). This research has been rooted in the theory of feminism advanced by Boserup (1970), leading to the introduction of separate programs for women as a major policy choice from the outset. Plans were laid early in the 1970s for the collection of sex disaggregated data to demonstrate the value of women to the functioning of the household economy (Okali, 2011). In the context of agricultural production, women were invariably shown to work longer hours than men; to be more likely to grow food crops for home consumption than men; to have more limited access to physical assets than men for their independent farming; to spend more time caring for children and the sick; and to undertake more household chores including the collection of firewood and water (Okali, 2011; Okali and Naess, 2013; Whitehead, 2000; Quisumbing, 1996). Comparisons such as these provided the material for the feminism oriented narratives that have driven much gender and natural resources policy over the past four decades.

While the beginning of the period between the 1970s and 1980s is characterized by a thrust in women and developmental debates with a continued portrayal of women as a single, isolated category (Lewis, 2002), there was a shift in development focus to gender and development (GAD) and also the adoption of a strategy of more fundamental gender mainstreaming toward the end of that decade. Considerable attention has been given to the development of frameworks for gender mainstreaming, since the 1980s, as opposed to women and development. Among these frameworks are the Moser, Empowerment and the Harvard/Gender Roles Frameworks (March et al., 1999). The idea was to find a different model for promoting gender equity, which would encompass a broader, more multicultural approach and one that took men into account (Jaquette and Rao, 2008). However, it appears that the two paradigms, WID and GAD do not differ much in practical terms. In fact, there are more similarities between the two approaches than some analysts may have realized given that women centered discourses remained central under the GAD paradigm. Understanding each approach in various historical and geographical contexts makes it possible to transcend the discourses within these approaches (Jaquette and Rao, 2008).

Within the gender and climate adaptation discourses, women have predominantly been portrayed as victims that are more vulnerable than men (Arora-Jonsson, 2011). However, research has also started to challenge this generalization (Sen and Östlin, 2008). Women are in some instances portrayed endeavoring to protect their households by the adoption of strategies and mechanisms in the face of shocks and stresses that affect the livelihood and food security of their households (Beraki, 2009, Neogy, 2012). Women are shown to have remained an untapped human resource, which, if given the same access to education, experience and farm inputs currently available to male farmers, can increase yields by between 20 and 30 per cent (Insights, 2012; Quisumbing, 1996; Sprengler, 2001). One case in point is that of Nigerian women, who are often known to adopt coping strategies in the face of food shortage by; allowing their offspring to eat first, eating once a day or picking left over foods at social functions and eating less expensive food, borrowing food or money to buy food, backyard crop and livestock farming and skipping meals within a day (Idrisa et al., 2008; Orewa and Iyangbe, 2010).

The new emphasis on climate change within the context of gender appears to have created opportunities for environmental feminism and ecofeminism to advance. Ecofeminism suggests a responsibility to challenge a patriarchal system that is deepening the impoverishment of the earth and its people (Bowerbank, 2004; Low and Tremayne, 2001; Ruether, 2004). The normative thinking on gender and climate change indicates that women are more vulnerable to climatic shocks than men (Denton, 2002). The key elements which conflated gender with women highlight rural women as victims, vulnerable and poor although equally playing roles in food security and nutrition for their households (Quisumbing, 1996; Okali, 2011; Okali and Naess, 2013). This past literature has generally portrayed the rural African woman as docile, unresourceful and utterly dependent on her male counterpart. Since 1970, sex differences in economic and domestic roles have been relentlessly described, but they have also been confounded with individual interests and therefore presented as fixed and unchanging, yet there is data that shows an evolution of these roles. Reed (2000) emphasizes the embeddedness of women responses to shocks and highlights that women’s agendas are determined by their circumstances.

Whitehead (2000) starts his work on gender with an idiom. He says:

In this regard, the gender discourse is said to have its roots in politics rather than in social science. This paper uses this overarching model of gender and climate change as the basis of understanding contextual manifestations of vulnerability on the ground using Nyaminyami District as a case study. The paper problematizes this model, which suggests a universal application in relation to all women in all places and situations, particularly in the South and suggests a more context specific narrative and consideration of gender relations as a basis for policy.

The paper further challenges this thinking of ‘woman as victim’ by contextualizing gender in climate change adaptation in a rural set up with the help of data collected in Rural Kariba and other supporting literature. There is recognition that capabilities are not necessarily matched by opportunities; thus, they do not automatically translate into outcomes, translating into diminishing disparities between women and men in recent decades (Chant, 2006; Mubaya, 2009). Social science literature on the contrary (Guyer, 1986; Moock, 1986; Chiappori et al., 1993; Hart, 1997; Reed, 2000) realizes a different view of gender in climate change issues e.g. vulnerability is measured in the field as the existing or non-existing asset base (physical, human, financial, social, natural) and considers the scale of their agricultural productivity accordingly. In essence, this literature acknowledges the importance of context in gender analyses. For instance, there are already glimpses that some aspects of social change in Bangladesh have already started improving women’s lives (Cannon, 2002). It is also quite clear that evidence that the impacts of climatic hazard impacts are worse for women is inconclusive and variable. We therefore ride on the suggestion by Acosta et al., 2016) that a gender responsive policy that takes into account real gender differences needs a different starting point, with context specific social and gender relations. This approach is likely to be more successful in supporting policies that result in increased and sustainable adaptive capacity. As it is, many policy responses tend to be based on the understanding that women are the most vulnerable victims of climate change and its impacts, limiting opportunities for these policy responses to address power imbalances (Arora-Jonsson, 2011; Flegel et al., 2009).

The study used both qualitative and quantitative approaches. Among the qualitative methods used for data collection are focus group discussions (FGDs), key informant interviews and in-depth case studies. The quantitative approach entailed conducting a questionnaire survey to complement data collected from other methods.

Omay is situated in the North Eastern part of the Zambezi valley (Murombedzi, 1993) and falls under Nyaminyami District, which is made up of three communal lands (Omay, Kanyati and Gatshe Gatshe). Omay is on the Southern shore of Lake Kariba and is adjacent to Matusadonha National Park. The Park is located at the center of the district and constitutes approximately 30 per cent of the total area. Omay is characterized by low agricultural potential and is located in agro-ecological Zones 3, 4 and 5. The environment is semi-arid and there is generally a seasonal rainfall pattern that is highly variable and barely amounting to 650 mm per year (Murombedzi, 1993). Omay has a hot climate of maximum temperatures above 40°C and rarely falling below 17°C. The soils are generally poor, resulting in consistently poor yields for the locals. The situation is compounded by the topography which is largely not amenable to intensive agriculture of any kind (Murombedzi, 1993). In addition, Omay Communal lands are Tsetse infested, a characteristic that makes livestock rearing almost impossible. While Tsetse control has been implemented over the years and communities have in some cases increased the number of cattle they keep, largely, only goats are still kept by the communities. Omay Communal Lands are made up of nine wards which are under the jurisdiction of Nyaminyami Rural District Council. We studied two out of the nine, namely, Nebiri and Negande (Figure 1).

Researchers for this study carried out key informant interviews with council officials, chiefs, village heads and local school heads. Data collected through these interviews include perceptions on climate changes, gender variations in responding to climatic and other shocks and the role that stakeholders play in advancing gender in climate change adaptation, among others. FGDs involved gathering a group of people from each of the sampled villages to generate data on a variety of the variables under the interviews, similar to the ones highlighted under interviews. The number of participants in each FGD ranged from 8 to 12 participants. Both men and women, young and old, formed part of the focus groups. Researchers separated men from women in some of the discussions. This was done to ensure that ideas came from both men and women to capture the reality for both gender groups. Various issues such as the livelihood strategies and access and endowment of these by gender were also unraveled through FGDs.

In-depth case studies were also used to capture lived reality from sampled individuals. Researchers also used purposive sampling to identify four farmers, two from each area, who were known to be engaged in one way or another in significant response to identified challenges as part of the in-depth case studies. Extension officers who work within the committees and are familiar with the households assisted the process. Key informants were picked from village heads, health personnel, local business people and council officials. Participants were both male and female in three categories of age groups (the youths, the middle aged and the elderly). This was to cater for varied experiences of different age groups and also to trace changes over time. Agro-ecological zones and food insecurity were important criteria for selection of the district in the first stage of the sampling process. Agro-ecological zones are strongly related to distribution of rainfall and are, therefore, a proxy for food insecurity and climate vulnerability. Kariba District is one of the districts that fall in the semi-arid and arid regions of the country and is therefore subsequently related to food insecurity. Food insecurity trends over a five-year period were analyzed in the Zimbabwe Vulnerability Assessment Committee (ZimVAC, 2014) reports from 2010 to 2014 and there was a realization that the district in question is among the consecutively food insecure districts. In the second stage, three wards (two in Mola and one in Gatshe gatshe) were purposively selected. Quantitative data were analyzed through the Statistical Package for the Social Sciences and used the Z-test and the chi-square test to determine differences between means for men and women data. Significant differences were declared at P ≤ 0.05.

Of the 116 respondents in Negande Ward who said they were aware of climate change, 54.3 per cent of them are men and 45.7 per cent are women (Figure 2). There was no evidence of a significant association between awareness of climate change and sex of respondent, implying that climate change perceptions are not influenced by gender. Further, a Z-test to compare the sex of respondent proportions on awareness of climate change suggests that there is no significant difference between the male proportions and the female proportions who indicate awareness of climate change. The study findings negate the popular notion that men tend to be more conscious of the environment than women and subsequently more likely to successfully respond to climatic shocks. Within this context, there is a suggestion that this finding points toward inherent potential within the study area for equal opportunities for both men and women agency to address levels of vulnerability to climatic shocks and, by implication, potential to challenge patriarchal structures that tend to characterize these study areas. This inherent characteristic is consistent with Aguilar’s 2010 assertion that women can actually be ‘powerful change agents’ instead of helpless victims of climate change.

Within this perceptions discourse, researchers further examined the dynamics of sources of information on climate that may contribute to the level of awareness of this climate phenomenon, particularly for the fellow farmer source, which emerges as the predominant source of information (Table I). Of the 73 respondents in Negande Ward who said they knew about climate change through fellow farmers, 54.8 per cent of them are men and 45.2 per cent are women. From the analysis, there is no evidence of a significant association between the sources of knowledge of climate change and the sex of respondent. Essentially, how a person knows about climate change is generally not influenced by gender.

In Nebiri Ward, there is a significant difference in the proportions of males and females who know about climate change through fellow farmers (p ≤ 0.05) (Table I). Females tend to rely on fellow farmers for information on climate change more than males, indicating that females come together more than males do. This could also explain the successful efforts to mobilize women for activities that build household resilience (Box 1) and emphasize that within this context, there is a healthy breeding ground for women agency toward adaptation. However, this scenario does not present an opportunity for full women empowerment and engagement in the absence of deliberate efforts by stakeholders targeted toward equal treatment. There is evidence of the threat of the persistence of patriarchal structures that are evident within the ecofeminism paradigm and that may advance unequal potential for agency by both males and females (Box 1).

Researchers examined responses to climatic shocks by gender based on five analytical risk management and adaptation classifications for the analysis of livelihood and adaptation in the study areas as suggested by Agrawal et al. (2008) and Agrawal (2010). These categories include communal pooling, storage, diversification, mobility and market exchange. There was no significant difference between the male proportions and the female proportions in both wards for all the categorized response strategies, except for the communal pooling strategy (Figure 2). However, there was significant association between the communal pooling strategy to shocks and sex of respondents in Nebiri ward (p ≤ 0.05). The proportion of males who use communal pooling as a strategy to climate-related shocks (20.7 per cent) is greater than the proportion of females (2.3 per cent), and (Figure 3) the proportion of females who use communal pooling as a strategy to climate-related shocks (97.7 per cent) is greater than the proportion of females for any other response strategy.

These patterns highlight that within this context, considerations for mainstreaming adaptation responses to climate change do not necessarily have to consider women as a special social group in isolation but, rather, implications for both men and women. If anything, women appear to have transcended custom which suggests that women tend to wait for assistance and lack a sound degree of agency (Box 2). Anecdotes of women being custodians of food security in the home not only for the children but also for the husbands and other males in the home are outlined in Box 3. Women in this context tend not to lament the ‘vulnerable status’ bestowed upon them but take a hold of the situation and emerge as victors and successful by adopting diverse household resource-use strategies to cope with food deficit situations, especially during lean seasons and natural disasters (Mohammed and Niaz, 2010). Mabonje (wetland maize cultivation) appear to be a very useful way to deal with episodes of insecurity and women have actually moved into mabonje throughout. It is only in these fields that they get meaningful yields. Reports from participant discussions highlight that inland fields hardly get yields, perennial droughts persist and mabonje have become part of strategies toward household livelihood security.

Thematic analysis of the FGDs and key informants highlighted that some of the gendered traditional roles appear to be in a state of flux. Historically, men in this area were fishermen and hunters. However, these activities are now heavily regulated to an extent that they are no longer feasible for these men. A depleted wildlife has moved further down the Zambezi valley and is protected by wildlife conservation authorities. For these reasons, hunting is no longer possible for men and is now viewed as poaching. The same is also true for fishing (generally a man’s domain), where the community’s use of Kariba dam is regulated, with specific designated fishing areas which are reported to have since been over fished, with subsequently no meaningful yields. This has emasculated the male population and relegated them to spectators in a game women seem to be play makers. Traditionally, gender roles have tended to determine the extent of involvement by both men and women in livelihood activities. The turn of events and apparently shifting of roles in the study areas lead to the challenging of historical perceptions that portray women as bystanders and secondary to livelihood issues. The finding resonates with the notion of translating into diminishing disparities between women and men in recent decades (Chant, 2006; Mubaya 2009). One of the local headmen indicated in an interview that:

This idea of shifting roles is emphasized by a number of informants. Another example that is highlighted is that nowadays, even the Chief’s wives go ‘kunoforera basa’ (looking for work), yet, customarily, the royal family received food from the community as part of the latter fulfilling social obligations.

In this paper, the general finding is that dominant and overarching narratives of women vulnerability in gender analysis in climate change adaptation appear to be in a state of flux, outdated and misplaced in some contexts. For this reason, we conclude that:

• There appears to be inherent potential within the study area for equal opportunities for both men and women agency to address levels of vulnerability to climatic shocks and by implication, potential to challenge patriarchal structures that tend to characterize these study areas.

• The contextualization of gender analysis remains elusive in the face of increasingly shifting gender roles that traditionally defined the fate of women as victims to everyday vulnerability and more recently in conjunction with climatic shocks.

In this regard, this research contributes to emerging perspectives on the potential role of ‘woman as heroine’ and challenges the perpetuation of ‘woman as victim’ in environmental management. Within this context, considerations for mainstreaming adaptation responses to climate change do not necessarily have to consider women as a special social group in isolation but, rather, implications for both men and women and caution that embeddedness remains key for gender considerations in any rural context.

The authors acknowledge funding from the National Research Foundation, South Africa.