Biruk has produced an account of ‘the social life of numbers’, with a focus on the administration of questionnaires by African field researchers for the international NGOs that constitute the ‘AIDS enterprise’ in Malawi. The act of making numbers and then making them count, or not, lies at the heart of postcolonial surveillance through global health. This book should be read by demographers who imagine that the monitoring of data collection by Malawian academics is the point at which indigenous knowledge is deployed. They will learn of the cultural distance between these academics and the field, and of the ways in which local field researchers themselves learn from villagers before translating this information into their own indispensable know-how. Members of institutional review boards who have oversight over the research ethics of rich academics working in poor countries should also read this book. They might see folly in the notion that information freely given is fairly given. They will see the ways in which local ethical review processes try to suppress the compensation of survey participants. They will learn that, for participants, several hours on a survey withdraw them from the work of making food – a sacrifice that, for many, is imagined as entailing a promise of further development assistance at a local, even a household, level. Without meaningful recompense, mere tokens for labour insult those who do not need it and short-change those who might.

From Cooking Data, policy analysts will learn the force of linking development assistance to programmatic commitments. They will learn that it was global pressure that urged Malawi to recognize its first AIDS case in 1985; it was the prospect of Western assistance that prompted Malawi’s first AIDS response plan in 1987; and it was only the promise of World Bank funding that saw the establishment in 2001 of a national AIDS commission. Surveys and data are a vital part of surveillance by NGOs in the global health sector, but Biruk shows in devastating detail how those numbers alone fail to direct the policies being funded. An imaginary geography of ‘advanced’ city and ‘backward’ countryside shaped an official understanding of AIDS as a problem produced by primitive culture under progressive siege from the metropolis. Nevertheless, research on HIV transmission through transactional and intergenerational sex found it to be of limited impact, while there was no evidence at all of significant HIV transmission through such cultural practices as widows being passed to brothers-in-law as sexual partners. In fact, there was no systematic evidence on the prevalence of these practices. Yet transactional and intergenerational sex and ‘harmful cultural practices’ were targeted by the national AIDS programme. On the other hand, despite official evidence of HIV transmission among men who have sex with other men since at least 1998, it was not mentioned in Malawi AIDS policy until 2003, and no funds were allocated for prevention or treatment until 2013.

This excellent book should also prompt reflection among anthropologists themselves. Biruk observes that the subject formation of an anthropologist used to be framed by long stays in a single locale. In contrast to the project-hopping demographers, anthropologists still benefit from slow study; but now they are as likely as demographers to be linked to the work of NGOs and dependent on their associated mobilities. In the case of critical studies in global health, this often places the anthropologist among those administering biomedical programmes and not, as perhaps was the case before, among the ‘beneficiaries’ of such programmes. Such an approach gives anthropologists such as Biruk the benefit of observing intercultural interfaces at multiple scales, but it requires more effort to take up the perspective of the most disadvantaged. This is certainly an obligation that Biruk discharges with diligence. However, it is notable that in identifying the contradiction between critical and useful studies, and in advocating a form of care to bridge that gap, it is the demographer and the field researcher, rather than the people of the villages, who are the primary objects of critical and useful care.

I wonder if there is scope for a ‘social life of numbers’ that steps over the household threshold and interrogates subject formation among the objects of the surveys. Many households have been surveyed multiple times as part of longitudinal studies. Indeed, the ‘projectification of the African landscape’ (p. 14), as the author argues, has helped to produce research fatigue and led to understandable resentment about the tokenistic bar of soap given in return for participation. Clearly, being surveyed creates an expectation that the information being collected will guide state policies and ultimately bring about tangible benefits. Biruk is acute in noting how bloodsucker rumours can circulate out of such disappointed hopes. Yet I wonder if the substance of the questions also shapes subjectivities. If people are asked about men who have sex with men, does it normalize or scapegoat this new focus of state curiosity?