Abstract
Anthropology is a discipline often considered wary of or even antagonistic towards quantitative data. In the book Cooking Data: Culture and Politics in an African Research World, C Biruk addresses this in a nuanced way, not by merely showing how quantitative data “gets it wrong,” but rather by rendering visible the knowledge it produces and how this in turn produces our shared world: “Numbers – and the standards by which they are evaluated – not only misrepresent real worlds but make new ones” (p. 212). Based on thorough ethnographic fieldwork on the production of HIV-AIDS demographics in quantitative health survey data in Malawi, Biruk explores these new worlds by studying the “social lives of numbers” (p. 3), troubling their status as “clean,” “raw” data and objective, pristine proof. This opens up an analytical space to reflect on broader questions of knowledge production, fundamentally unequal research worlds, and international research collaborations. Firmly rooted in the anthropology of global health, this accessible and well-written book speaks, however, to anthropology and academia as a whole.
Each of the five chapters addresses a different element that plays a role in how demographers, who are taken as representative of quantitative scientists, create “data” as well as how these data are subsequently used. In chapter 1, Biruk describes the preparatory work that is required to make “the field” in Malawi fit the survey questions that have been designed by foreign demographers. As it turns out, the production of “objective facts” necessitates translations, both linguistically and conceptually, prompting Biruk to engage with epistemological questions, (historical) processes of othering, and the fundamental inequalities inherent in research “collaborations.” Elaborating on these inequalities, chapter 2 focuses on the Malawians who administer the surveys in the field. Generally regarded as unskilled labourers who simply collect data that are already out there, Biruk calls them “knowledge workers” to highlight their primary role in data creation: “local knowledge comes to exist – and to gain value – because of them” (p. 83). As many (aim to) make a career out of administering surveys for different (international) projects, Biruk argues that it is not just the data but the knowledge workers themselves too who are produced in particular ways: “the fates of data and their creators are linked” (p. 98).
Although voluntary participation is generally considered a precondition for obtaining “clean data,” survey participants in Malawi are compensated with two bars of soap. Considered by both the research project and Malawi’s ethics board an “ethical gift” befitting the poverty-stricken context, chapter 3 details how different actors ascribed different meanings to the soap, depending on their position in the research world. However, these struggles over meaning, although very present during interactions in the field, fell firmly outside the scope of the survey itself and thus did not become “data.” This obscures “how it is through research transactions that people (and data) are made and unmade” (p. 124), leading Biruk to argue that “clean data can only materialise within and through messy social relations and transactions” (p. 125). This, importantly, holds true for both quantitative and qualitative research. Chapter 4 builds on this insight through a focus on how standards for data collection are performed by knowledge workers in the field. Using ethnographic vignettes, Biruk shows how “clean data” have in fact been carefully crafted to appear as such, which makes it “inherently cooked” (p. 139), exposing raw data as “an imagined fiction” (pp. 164–165).
The ways in which these data are subsequently turned into numbers and statistics that inform policymaking are the focus of chapter 5. Biruk convincingly shows how “the numbers underlying evidence-based claims in the policy-research nexus are never stable and always subject to processes of cooking, even in finished form” (p. 169). The social life of numbers, the underlying inequalities that structure their production as well as their (re)interpretations and alterations along their “life course” (p. 184), are not only essential for understanding the gap between data producers (researchers) and data users (policymakers) but also reflect inequalities in research collaborations and access to funding: without funding, for example, no numbers can be generated to bolster claims.
In line with the aim of the book, Biruk concludes not by celebrating qualitative research over quantitative work, but rather by reflecting on knowledge production and ethnographic data in particular. Where anthropologists cherish the image of being critics who render visible what others fail to capture (numerically), inadvertently obscure, or simply do not want to see, they are at the same time complicit in perpetuating the existing power structures that shape our (research) world. By destabilising notions of “the field” and fieldwork as the basis of knowledge production and truth claims, Biruk directs our attention to anthropology’s own data cooking practices, deploring how research often “reproduces the asymmetries it seeks to redress” (p. 209). Placing Cooking Data in line with work done by Vincanne Adams and Sally Engle Merry, C Biruk calls for anthropology’s renewed engagement with numbers and knowledge production. As relief interventions reliant on quantitative demographic (health) data become more and more frequent in Malawi and the current global COVID-19 pandemic has the world in its grip, Biruk’s work highlights pertinent questions that need answers – which are yet to be given or put into practice.
The proverbial proof being in the pudding, Biruk’s book contributes chiefly to the start of much-needed broader discussions on quantitative data, (anthropological) knowledge production, and its relations to policymaking. It is essential reading for scholars interested in epistemological questions within the discipline of anthropology, but also beyond, and for scholars involved in interdisciplinary research projects as well as international collaborations and research partnerships. Ultimately, Cooking Data urges us to ask what it is that data and numbers render visible and invisible, but also to go beyond that: how to make what you cannot count or capture in numerical form not just visible but of consequence – both in policy and research worlds.