In this introduction[note 1]The five papers discussed in the present issue were presented at the CERES/EADI Summerschool 2008.
I shall explore the dimensions of the concept of resilience and, in closing, try to formulate a conclusion about its applicability and usefulness in the context of ill health and adversity. In doing so I shall discuss the different ways the concept is applied to empirical data presented in several papers in the present issue, notably those of Kessy and Obrist, Matinga, Nombo and Niehof, Van der Geest, and Wiegers. The Dutch term for resilience is veerkracht, which literally means spring power. Upon release, a strong spring when stretched or compressed, easily resumes its previous state. It bounces back. In the literature resilience has indeed been referred to as ‘bounce-back-ability’ (e.g. Davies 1993). If applied to a context of ill health and adversity it would mean the ability of people to regain their health and well-being and recover from adversity, shock and stress. In the case of HIV/AIDS Loevinsohn and Gillespie (2003) call resilience the ability of people to recover from the devastating impacts of HIV/AIDS and regain an acceptable level of well-being.
These descriptions indicate that resilience as the ability to bounce back has to do both with internal strength and with external factors challenging it. Carpenter et al. (2001) emphasize that defining resilience requires looking at the resilience of what and to what. Since in this special issue we focus on the resilience of people and households, and not, for example, on the resilience of ecological systems (e.g. Berkes et al. 2003), it is more appropriate to rephrase the two sides of the coin of resilience of Carpenter et al. as resilience of whom to what.
The two dimensions of resilience raise a question about their interrelationship. How do they interface? Does the kind and severity of the challenges and shocks people face, affect their strength or do people have a given amount of strength irrespective of the challenges they are confronted with? This leads to further questions, such as: why are some people stronger than others and what are their sources of strength? What causes some people to be able to bounce back even under the most adverse circumstances while others cannot? Phrased in a slightly different way, these questions are also part of the discourse on vulnerability in the livelihood literature, where a distinction is made between the vulnerability of a livelihood system and its external physical and institutional vulnerability context (see Brons et al. 2007). Davies (1993) distinguishes sensitivity and resilience. The intensity with which a system experiences a shock (sensitivity) determines the degree to which it is difficult to bounce back (resilience). It can further be assumed that if the vulnerability context affects people’s own vulnerability or sensitivity, it will also play a role in shaping the scope and impacts of shocks and stress and the opportunities for dealing with them. Hence, the two dimensions of resilience in fact hide a multi-layered process (cf. Kessy & Obrist, this issue). People’s agency can be positioned as a mediating factor in the interface between their sensitivity/vulnerability and the external vulnerability context, which – at the end of the day – together determine the outcome of the process in terms of resilience. Agency has been defined as “reflexively monitored flows of conduct” (Carter 1995: 65, my italics). To me, it means conscious and purposive behaviour of actors who consider the appropriateness and efficacy of their behaviour in a certain normative and institutional context.
For social scientists a key question would be how the layers that can be peeled from resilience as a process are socially, culturally, and institutionally embedded, which also comprises the question of the way resilience is gendered. In the following section I will address these questions, using the empirical data provided by the five papers on resilience in the present issue.
Sensitivity, agency and their social and cultural embeddedness
Using sensitivity and vulnerability interchangeably, the five papers deal with people who in several ways are sensitive to shock and stress. The people in Dodoma, Tanzania, in the paper by Kessy and Obrist, are poor and have no access to safe water and sanitation, which jeopardizes their health. The women in Eastern Cape, South Africa, in the paper by Matinga, daily have to carry heavy burdens of fuel wood on their backs because they cannot afford more expensive sources of energy for their households. The households in the village of Mkamba, Morogoro, Tanzania, described in the paper by Nombo and Niehof, are poor, deprived of their land, and struggling with HIV/AIDS affliction. The woman in a village in Southern Ghana in the case study by Van der Geest is poor, old and handicapped. She is blind and lost one leg. Wiegers’ paper on households in Zambia also focuses on the impacts of HIV/AIDS, as does the paper by Nombo and Niehof, and highlights in particular the plight of widows in a context of high prevalence of HIV/AIDS. Poverty as component of vulnerability is a common denominator in all papers. Additionally, emotional stress plays a role in making people more sensitive, especially in the case of old-age dependency (Van der Geest) and when afflicted with HIV/AIDS (Nombo & Niehof, Wiegers).
While these context-neutral observations reveal factors that increase people’s vulnerability (or sensitivity), they do not tell us much about the way these factors are socially and culturally embedded and how people use their agency to deal with them. At this point the papers diverge. In the paper by Van der Geest the key word is reciprocity. Culturally underpinned reciprocity between parents and children provides a counterbalance to old-age vulnerability. However, reciprocity does not work as an automatic mechanism. It needs nurturing and it is ‘whimsical’. Not acknowledging it is a recipe against old-age insecurity, but when parents abide by the norm it does not guarantee that they will be cared for in old age by their children. Old persons may use their agency for a moral and normative appeal to their children, which is their cultural right, but in the case of Maame Mercy Ofori it works for her son but not for the other adult children. Similarly, the paper by Nombo and Niehof shows how reciprocal social relationships, part of people’s social capital, need nurturing and material investments and cannot be taken for granted. Women left behind when their husbands die of AIDS struggle with the fear that they are infected themselves, may lose access to land and may become the victim of property grabbing by their in-laws (see Wiegers). However, the case of Emily in Wiegers’ paper shows that such developments can be prevented if the widow has been able to build a strong social network and is able to sustain it. Social capital plays a key role in reducing sensitivity in a situation of high HIV/AIDS-prevalence, as the papers by Wiegers and Nombo and Niehof show. But the paper by Nombo and Niehof also makes painfully clear how women’s agency in building and sustaining social capital is curbed by their lack of means. In the paper by Kessy and Obrist, households’ access to safe water is likewise obstructed by their inability to pay user fees. As said above, poverty is a common denominator, apparently across cultural contexts.
Culture as a defining factor of vulnerability plays a key role in the paper by Matinga. She introduces the concept of ‘hardiness’. Hardiness can be read as toughness, stamina and ability to endure, rather than resilience; not bouncing back but reducing sensitivity and making yourself less vulnerable. In this process, women’s culturally underpinned value as collectors, connoisseurs and users of wood for the benefit of their family bolsters their hardiness. The women interviewed acknowledge this and take pride in, but – at the same time – they complain that they have no other choice than fulfilling their back-breaking duty and inhaling the smoke while cooking, and that they would like to have the means to use alternative, less harmful, sources of energy. Then hardiness becomes hardship, something that – whatever one’s agency – one cannot do much about but enduring.
People’s vulnerability/sensitivity is not only determined by their own characteristics and agency, but also by contextual factors that increase or decrease their vulnerability and affect their agency. The livelihood literature refers to this as the vulnerability context (e.g. Brons et al. 2007). If we look at the five papers under discussion there are several significant aspects of the vulnerability context that merit attention.
The (bio-)physical context
The physical infrastructure (or the lack of it) plays an important role in the vulnerability of the households in the study site described by Kessy and Obrist. The area is the largest and poorest unplanned area in Dodoma Municipality. The overflowing waste water from pit latrines contaminates the ground water in the shallow wells that households use for daily water consumption (Kessy & Obrist, this issue). For the wood collectors in the paper by Matinga, the ecological environment of the forest, on which they depend, has its own vulnerability dynamics. Not only does going into the forest expose them to hazards (such as snakes), degradation and thinning of the forest by wood collection might in future require women to venture further into it and take more risks.
The socio-demographic context
The changing demographic and social context may negatively affect people’s vulnerability in areas characterized by a high level of migration and HIV/AIDS, as in the paper by Nombo and Niehof and in Wiegers’ paper. Migration leads to fragmentation of kinship networks. People might still abide by norms of reciprocity and mutual help among kin, but they fail in enacting these norms because they are living too far apart. Interestingly, in the paper by Van der Geest the son returned to his mother’s village so as to take care of her. In an area with a lot of in-migration of labourers, migration also results in ethnic diversity. In the case of the village of Mkamba, as described in the paper by Nombo and Niehof, ethnic diversity negatively affects the vulnerability context because people don’t easily relate to other people who do not belong to their ethnic group. They cannot count on these ‘others’ and sometimes even mistrust them, at least in the case of this village.
Reciprocity and social capital
As Nombo and Niehof show, people may also be unable to abide by the norms of reciprocity because they are too poor to fulfil the obligations these entail. Then stating those norms may become empty rhetoric and practice becomes ‘whimsical’ (Van der Geest). As a consequence of the impotence in abiding by norms of reciprocity, these norms, however audible in oral exchanges and sayings, will fade away, resulting in an anomie-like situation. In order to establish to what extent this is the case, more in-depth research is needed, but if that is happening, collective social capital is eroded to such a degree that it can no longer be a source of support. As Berry (1993) found, in a situation of widespread poverty and stress the social obligations embodied by kinship and other social institutions can no longer be honoured. The evidence in the literature shows that in communities with high HIV/AIDS-prevalence, where the majority of households is in some way affected, households “fall through” the vulnerability threshold (Donahue et al. 2001: 9). In other words, their resilience is irreversibly affected.
The role of social capital seems to be ambiguous. On the one hand social capital, especially the kinship network, is assumed to be the last and crucial resort of poor people who lack other kinds of capital. On the other hand, it appears that people who lack material means also lack social capital (see, for example, Sauerborn et al. 1996). The paper by Wiegers testifies to both the significance of social capital and its fragility in a situation of high HIV/AIDS-prevalence. Nombo and Niehof draw attention to the erosion of community-level social capital and personal networks in their study area because of the fatal synergy between HIV/AIDS, poverty and ethnic diversity. Also in this case social capital appears to be fragile and increasingly out of reach of the people who need it most. However, if these people would not benefit materially from social capital, it could still provide moral and emotional support. In the cases described by Nombo (2007) this sometimes applies, but also with regard to moral and emotional support, proximity (in the case of kin) and mutual trust (in the case of neighbours from different ethnic groups) are important enabling conditions.
The cultural part of the vulnerability context
In the different situations reported by the papers, the cultural aspect of the vulnerability context works out in various ways. For the Xhosa women in Matinga’s paper their ‘value is in wood’. A woman’s ability to provide adequately for her family is measured by the size of the pile of wood (igoqo) she keeps outside her hut. Similar gendered and culturally underpinned desirable and prescribed behaviour that in fact heightens women’s individual vulnerability is visible in women’s role as caregivers in a context of HIV/AIDS.
In the cultural part of the vulnerability context of HIV/AIDS the secrecy and stigma often surrounding HIV and AIDS add to the contextual vulnerability. Because of secrecy and stigma, HIV and AIDS are – as it were – placed outside the cultural and social order and their discourses. There is no open communication about it and AIDS becomes like a monster lurking in the shade. AIDS is then easily linked to witchcraft, an – equally shadowy – alternative discourse that is fuelled by suspicion, fear, and secrecy (Nombo, 2007). In such a situation HIV/AIDS undermines the social and cultural fabric of society. People know this but do not want to acknowledge it, which renders powerless the agency of the infected and affected people. The reverse seems to apply as well. The case of the village of Kidatu that Nombo and Niehof refer to, shows that in a socially strong community – less stricken than the research site by poverty, ethnic diversity, anomie and mistrust, HIV/AIDS can be given a place in the order of things and affected people use their agency to get help and help others. This reduces contextual vulnerability for the afflicted and affected households and enhances their resilience because it provides space for their agency.
Resilience as process: the temporal aspect
Above I already alluded to resilience as the outcome of a process by drawing attention to the factors affecting people’s own sensitivity or vulnerability. These factors increase contextual vulnerability and affect people’s agency in mediating between these and the stress or shock that they are exposed to. There is a temporal dimension to the interfaces comprised by this process.
In most papers under discussion the stressor is not a one-time shock but rather a condition developing over time. In the paper by Matinga, the stressor even acquires a certain permanency, given that wood is indispensable for fuel and for the culturally framed equation of women’s value and wood. In the paper by Van der Geest, the stress of dependency induced by old age and disability is part of the human life cycle and can be anticipated, which explains why it is part of the normative paradigm of reciprocity. Although a cholera epidemic, which might easily strike in the research site of Kessy and Obrist, could be called a one-time shock, given the prevailing conditions it is also a shock that can be expected. Diseases like cholera and malaria can become endemic in certain areas because local conditions, including poverty, nurture their vectors (vulnerability context). This also applies to HIV/AIDS and poverty, as convincingly described by Stillwaggon (2006).
HIV/AIDS represents a shock with temporal dynamics of its own, particularly because of the long time (six to eight years) that elapses between the moment at which a person is infected and the moment s/he begins to experience the first signs of infection. After that, the development and duration of full-blown AIDS will depend on a person’s condition and on the quality of care and medication. Likewise, impacts of AIDS morbidity and mortality on households and communities are temporally structured in ways that are difficult to predict (see Barnett & Whiteside 2002 for a discussion of these issues). Baylies (2002) has called HIV/AIDS a long wave disaster that unfolds in stages. Resilience may still be high in the first stage but declines as affected households are forced to sell assets to cater for their needs and have less people they can call on for help. The latter is also a consequence of the clustering of impacts in households and communities (Barnett & Whiteside 2002). Measuring resilience is difficult, but in the case of resilience to HIV/AIDS even more so, because of the complex temporal and spatial dynamics of the epidemic.
In their paper Kessy and Obrist take resilience as a process one step further by distinguishing between ex-ante and ex-post resilience. So far, most of my discussion has dealt with what Kessy and Obrist call ex-ante resilience. They refer to resilience to the impact of cholera after the disease has struck as ex-post resilience. It is mitigated by the availability of appropriate health services (emergency units) and the level of sensitization about proper health-seeking behaviour. Lack of such institutional support structures negatively affects ex-post resilience. Their argument can be extended to include ex-post resilience as part of resilience to AIDS (Wiegers, Nombo & Niehof) and old-age dependency (Van der Geest). In the case of HIV/AIDS ex-post resilience will be influenced by institutional health structures and availability of medication, as well as by supportive community structures and social and economic safety nets. For resilience to the effects of old-age dependency the same kind of institutions and structures are important, including normative frameworks that specify care obligations and entitlements. The concept of ex-post resilience is not applicable to the situation of the women in Matinga’s paper, because there the stressor is more or less permanent. Women cope with that by enduring and being tough, even if it means that their health complaints become a chronic condition.
At the beginning of this introduction I posed the question of the relevance of the concept of resilience in the context of health-related adversity. The discussion so far has shown resilience to be multi-layered, gendered, and temporally structured. Provided this is taken into account, applying the concept to contexts of health-related adversity can gain us depth in understanding people’s exposure and responses to such adversity. One could subsequently question whether the relevance of the concept is merely an academic issue or whether using it can contribute to prevention of health-related adversity and to mitigating its effects by policies and programmes. I think the latter question can be answered in the affirmative.
Policies and programmes can bolster people’s resilience when facing ill health and adversity. They can do so by improvements in the vulnerability context of ex-ante resilience as well as by setting up or maintaining appropriate functional structures and carrying out adequate measures to improve ex-post resilience. Apart from the need for analysis to identify the significant factors in the vulnerability context and assess the adequacy of the institutional structures, implementing policies and programmes requires resources. So, also at this level poverty is an overriding issue, in much the same way as it is the common denominator in the vulnerability of the people portrayed in the papers. However, in a situation of scarce resources it becomes even more important to strengthen people’s resilience by finding the right entry points and addressing key dimensions of vulnerability in an informed and efficacious way.
Finally, we may assume that emotions, personal experience and personality characteristics play a role in shaping resilience, but the papers do not provide much information on these issues. The paper by Van der Geest pays most attention to the emotional and personal side of vulnerability and resilience. However, the paper also shows that emotions can be, or at least seem, contradictory. In the same interview, the old lady says that life is not difficult for her (because of her son’s care for her) and that she is unhappy (because she is handicapped). Presumably, both feelings are authentic. Perhaps the fact that in spite of being unhappy the old lady realizes that she is lucky to have her son taking care of her indicates strength of character that reduces vulnerability. Another point to be made with regard to this paper is the following. Van der Geest frames the care relationship between Maame Mercy Ofori and her son within the discourse of reciprocity, notably the reciprocal relationship between parents and children. There is another way to look at the son’s care for his mother, which might explain the old lady’s ambiguity. Care can also be conceptualized as a gift, and as such it “has no agenda or programme apart from […] responding to the ‘other’ in the care encounter” (Walsh 2007: 220). The receiver of such a gift should be grateful, but can be unhappy at the same time. Furthermore, family responsibilities, including those relating to care, are always negotiated, albeit within a certain normative framework, rather than following blue-printed norms (Finch & Mason 1993). Even if care is a gift, there is space for giving or not giving and receiving or not receiving. In the case of Maame Mercy Ofori it is clear that her adult children have positioned themselves differently in this space, but we don’t know what role the old lady’s own agency played in this.
Emic narratives can reveal how people use their agency in negotiating assistance to reduce their vulnerability and increase their resilience. Fragments of emic narratives can be seen in the statements made by people in the papers under discussion. The emotions expressed in the statements vary from worry, resignation and acquiescence (Nombo and Niehof) to belligerence and protest (the women in Matinga’s paper). Mostly, they reveal the informants’ grudging understanding of the situation they find themselves in and have to accept because there is no other option. Only Emily’s narrative (Wiegers) clearly reveals positive agency and ability to respond. The role of emotions and personality characteristics in vulnerability and resilience seems to be a bit of a blind spot. This could explain why researchers in a longitudinal study in Uganda were “surprised” to find households much more resilient to the devastating impacts of HIV and AIDS than they had anticipated (Seeley et al. 2008: 1444). The questions raised at the beginning of this paper of ‘why are some people stronger than others and what are their sources of strength?’ and ‘what causes some people to be able to bounce back even under the most adverse circumstances and others not?’ clearly need more research to answer. Such research on resilience should include emic approaches and focus on people’s agency.
Anke Niehof (firstname.lastname@example.org) holds the Chair of Sociology of Consumers and Households at Wageningen University and Research Centre, the Netherlands. She wishes to thank Jeff van Exel for the English editing of this paper.
2002 The impact of AIDS on rural households in Africa: A shock like any other? Development and Change 33(4): 611-32.
Barnett, T. & A. Whiteside
2002 AIDS in the Twenty-first Century: Disease and globalization. New York: Palgrave.
Berkes, F., C. Folke & J. Colding
2003 Navigating socio-ecological systems: Building resilience for complexity and change. Cambridge UK: Cambridge University Press.
1993 No condition is permanent: The social dynamics of agrarian change in sub-Saharan Africa. Madison: University of Wisconsin Press.
Brons, J.et al.
2007 Dimensions of vulnerability of livelihoods in less-favoured areas: Interplay between the individual and the collective. In: R. Ruben, J. Pender & A. Kuyvenhoven (eds), A sustainable poverty reduction in less-favoured areas. Wallingford: CAB International, pp. 91-110.
Carpenter, S.R. et al.
2001 From metaphor to measurement: resilience of what to what? Ecosystems 4: 765-81.
1995 Agency and fertility: For an ethnography of practice. In: S. Greenhalgh (ed.), Situating fertility: Anthropology and demographic inquiry. Cambridge UK: Cambridge University Press, pp. 55-86.
1993 Are coping strategies a cop out? IDS Bulletin 24(4): 60-72.
Donahue, J., K. Kubbocho & S. Osinde
2001 Responding to a silent economic crisis among microfinance clients in Kenya and Uganda. Nairobi, Kenya: Microsave. www.microsave.com (accessed on 25/6/07).
Finch, J. & J. Mason
1993 Negotiating family responsibilities. London and New York: Tavistock/Routledge.
Loevinsohn, M. & S. Gillespie
2003 HIV/AIDS, food security and rural households: Understanding and responding. Regional Network on HIV/AIDS, Rural Livelihoods and Food Security. Working paper No. 2. Washington DC: RENEWAL.
2007 When AIDS meets poverty: Implications for social capital in a village in Tanzania. AWLAE Series No. 5. Wageningen: Wageningen Academic Publishers.
Sauerborn, R., A. Adams & M. Hien
1996 Household strategies to cope with the economic costs of illness. Social Science & Medicine 43(3): 291-301.
Seeley, J. et al.
2008 Using in-depth qualitative data to enhance our understanding of quantitative results regarding the impact of HIV and AIDS on households in rural Uganda. Social Science & Medicine 67: 1434-46.
2006 AIDS and the ecology of poverty. Oxford: Oxford University Press.
2007 A birth centre’s encounters with discourses of childbirth: how resistance led to innovation. Sociology of Health & Illness 29(2): 216-32.