This study examines social constructions of infertility in Malawi. The literature on infertility consists of epidemiological studies, describing patterns of infertility in terms of its incidence, causes and health seeking behaviour; studies of the psychological correlates of infertility; and ethnographic studies which describe experiences, perceptions and management of infertility within specific socio-cultural contexts. In addition, some studies discuss social aspects of medical practice in relation to infertility. Overall, studies of infertility in developing countries emphasize its many serious psychological and social consequences, usually attributed to cultural norms mandating parenthood. There appear to be several lacunae in the literature: men with fertility problems are rarely included, an in-depth examination of practitioners’ views is missing, and no qualitative study has been conducted on infertility in Malawi, which has a considerable secondary infertility rate. Furthermore, although ethnographic studies highlight the interpersonal (related to others’ judgements), normative (related to ideas about what ‘ought’ to be) and moral (related to ideas about what is good or bad) issues involved in infertility, no study has investigated how these issues are managed in situ, in verbal interactions. However, it has been argued that ‘talk’ is a prime site for the management of issues such as blaming and deflecting responsibility. Hence, this study addresses several gaps in the literature. It focuses on Malawi, and includes a wide range of participants: women and men with a fertility problem, significant others, indigenous and (Malawian and expatriate) biomedical practitioners. Semi-structured interviews with 63 participants were recorded and transcribed, and translations were obtained of interviews in which interpreters were used. For the analysis, I used discourse analysis (DA), informed by conversation analysis (CA). This analytic approach, novel in infertility studies, examines the interpersonal functions of statements in interactions, such as blaming or justifying. Use of DA and CA has led to novel insights into how respondents construct infertility, its causes, solutions (sought and offered), and consequences, and how they thereby manage interpersonal, normative, and moral issues, revolving around accountability, blame and justification, and attribution of (problematic) identity categories. For instance, I have shown how respondents construct childbearing as a cultural, normative requirement, and how this can be used to justify practices like extramarital affairs, or polygamy, as necessary solutions. In addition, identifying causes appears to be problematic for people with a fertility problem due to certain interpersonal and interactional issues, such as the idea that they are not entitled to medical knowledge. Practitioners can be seen to work up and bolster an identity of professional, competent expert in constructions of causes of infertility, and by attributing problems in helping infertility clients to external factors, including patients’ intelligence. This study has several theoretical, practical, and methodological implications, although I discuss some thorny methodological issues, especially those concerning the use of translations and the transferability of the analytic findings. A first contribution pertains to methodological debates and developments in conversation analysis, and in studies of infertility and other health issues which rely upon people’s self-reports. Second, my study contributes to theoretical developments in health psychology and health promotion. My analysis points to the relevance of social and normative considerations for engagement in ‘risky’ behaviours, such as extramarital affairs. This challenges cognition models which treat health behaviour as the outcome of individualistic decision-making processes, and see providing information as the main way of changing people’s behaviour. Therefore, a third set of implications is of a practical nature: some of the findings can contribute to health promotion, as well as to improvement of health services. For example, practitioners’ attribution of failures and (communication) problems to their patients, may prevent them from reflecting critically on, and addressing, their own contributions to problems. Overall, this thesis shows that when one wants to ‘give voice’ to people who are suffering from infertility, it is valuable to examine what they say in detail, within its interactional context, and the concerns they themselves make relevant, in their own terms.