Paper short abstract:

Reflecting on the idea of ‘connecting the dots’ to understand a novel pathogen in social and epidemiological contexts, this paper draws on conversations between modellers involved in COVID-19 responses in the UK and anthropologists to imagine future interdisciplinary collaborations.

Paper long abstract:

Early in the COVID-19 pandemic, mathematical modelling captured global attention, seen as a privileged tool to make sense of early epidemiological data. Assessing pathogenic, epidemiological, clinical and socio-behavioural characteristics of an outbreak to inform policies in real time is also known as ‘nowcasting’. We interviewed experts for a study documenting the role of modelling in policy responses to COVID-19 in the UK. Throughout the research, ‘connecting the dots’ emerged as a strong rationale to develop models which could help understand a novel pathogen. Our conversations with modellers resonated with our own experiences navigating uncertainty while doing research on epidemic responses. As anthropologists, we accept the fact that data is relational, and that what is made visible in ethnographic accounts necessarily mask other ways of knowing. To some extent, ‘connecting the dots’ between what is rendered visible and invisible by the anthropologist’s presence is part of the ethnographic enterprise.

In this paper, we contemplate uncertainty as an epistemological stance and a methodological effort which can be politically mobilised when systems are disrupted. In contrast with an anthropological critique of models, we offer reflections on how conversations between modellers and anthropologists can productively explore uncertainty. Such an approach allows for a dialogic interrogation of numeric and narrative representations of people, pathogens and flows when intervening in an epidemic. We do not intend to reconcile methodologies which are embedded in radically different epistemological traditions, but rather seize the opportunity of revisiting encounters between anthropology and modelling to imagine future interdisciplinary collaborations.

Paper short abstract:

Our paper analyses a collection of articles which include pandemic models from two Anglo-American daily newspapers to discuss the power of models in storying futures and sketching characters.

Paper long abstract:

The COVID-19 pandemic created an abrupt increase in uncertainty. This led the general public and decision makers alike to turn to expert knowledge of the past and present, in an attempt to plot the ongoing crisis and post-pandemic trajectories. Epidemiological and infection models are forms of expert knowledge often reported by mass-media, informing public debates, policies, and the collective imagination.

Models sketch narrative characters, explicitly or implicitly (Polletta, 2015). They define types of people and populations through parameters of contagion and survival, and imply evaluations of social and moral worth. They quantify (Espeland and Stevens, 2008) alternative social trajectories and commensurate (Espeland and Stevens, 1998) diverse interests and activities. Models also enact a discursive production of temporality, through time work (Flaherty, 2003), shaping readers’ invocations of history, definitions of the present situation, and anticipations of pandemic and post-pandemic futures.

How do models formulate the pandemic timeline and, consequently, project the future? How do they rely on scientific knowledge to lump and split (Zerubavel, 1996) risk categories? Our paper analyses a collection of articles which include pandemic models from two Anglo-American daily newspapers, The New York Times and The Guardian, published between January-June 2020 and January-June 2021, to discuss the power of models in storying futures and sketching characters.

Paper short abstract:

Based on qualitative depth interviews with 28 mathematical modellers engaged in the UK COVID-19 response, and selecting two modelling case examples, we trace how the making of pandemics ‘big’ and ‘small’, as well as ‘taller’ and ‘flatter’, is not simply a matter of method but of politics too.

Paper long abstract:

In this paper, we trace how the making of pandemics ‘big’ and ‘small’ is a critical element in the politics of evidencing. We draw on depth qualitative interviews with 28 mathematical modellers and others engaged in the UK COVID-19 response. Inspired by the field of science and technology studies we orientate around ideas of evidence-making assemblage. To make our story, we select for attention two particular devices linked to modelling which eventuate the imagined pandemic in different ways, and specifically, in different shapes (tall and flat) and sizes (big and small). Our first example concerns the estimation of ‘doubling time’. Our second example concerns the estimation of ‘individual variation’ in susceptibility. Both innovations challenge ‘consensus’ of the time, but in different, as well as in bigger and smaller, ways. One of these innovations breaks through while the other does not. This concentrates our attention on how the boundary work around the constitution of modelling evidence in a time of pandemic might tell us something about the political assemblages of evidence-making which afford some models more life, agency and sustainability than others. How might it be that some evidence-making innovations break through but others get held back or even shut down? Such boundary work is not without consequence for science or for the scientists concerned. Perhaps when pandemics are made big and tall, models and modellers which enact them smaller and flatter can become troublesome, especially when experimenting with less familiar methods which break from the routine.