One of the most important moments in the development of cartography occurred when an obscure doctor, John Snow, produced his so-called ghost map of the cholera outbreak in the London neighbourhood of Soho in 1854.
By pinpointing the location of cases, and showing where they clustered, the doctor was able to identify the water pumps that were responsible for the outbreak, and to consign to the dustbin of medical history the “miasma” model of transmission – according to this theory, diseases were caused by some kind of bad air emanating from putrescent organic matter. Ever since, mapmaking has played a significant role in epidemiology, with ever more sophisticated cartographical techniques being used to capture and represent the big data of disease.
Because the cholera bacterium is transmitted through infected water and food, and not directly from person to person, it is relatively easy to trace its sources and take appropriate remedial action.
Not so, of course, with other infectious diseases from colds, flu, measles and mumps to smallpox, Ebola, and now, of course, Covid-19. Even when using geo-locating apps, it is almost impossible to track all the movements of coronavirus human carriers, many of whom may initially be symptom free. This challenge generates a pervasive contagion anxiety.
If it may not always be possible to map the “known unknown” – in this case, the full extent of transmission – it is much more difficult to represent what the psychoanalyst Christopher Bollas has called the “unthought known”, the deeper more unconscious responses to the pandemic, known in some way to us, but elusive in their meaning, sometimes repressed, sometimes emerging in coded forms.
We are dealing here with what psychoanalyst Sigmund Freud named the “other scene” of everyday life, where rational calculations of self-interest and limited altruism give way to structures of feeling and belief dominated by fears and fantasies of the “other” (other class, other gender, other ethnicity) about which we are often in denial.
This is fertile ground for the popular imagination of disease. These inhabitants of our inner world live in a country of the mind which remains foreign to us, making us strangers to ourselves, but which nevertheless we can still own and, after a fashion, assemble cognitively into some kind of map. Indeed it is one of the undeclared functions of this social imaginary to transform the fluid relays of communicable disease into fixed signs – omens or portents, variously associated with hidden malevolent forces or more visible pariah groups.
How then can these imaginary geographies of risk be represented in a way which does not underpin conspiracy theories, or justify attacks on vulnerable groups who find themselves in the frontline, not just of viral infection but also of vigilantism?
This task is usually left to poets, novelists, visual artists, choreographers and psychoanalysts. In particular, there has been an explosion of interest amongst artists and writers in creating imaginary maps, maps which represent other possible worlds, or support fictional narratives. Literary cartography has been established as a distinctive field of study. However, such mappings remain confined to the imaginative worlds created by their authors, and they rarely adventure into the field of ethno-cartography, to investigate how the cognitive maps (e.g. how people perceive certain features of the environment in their minds) of different social groups are shaped by their cultural formations.
At the same time, conventional cartography, with its strong commitment to an allegedly faithful representation of the physical territory which the map refers to, has developed ever more precise and high powered digital technologies for representing the real. From this standpoint, any mapping of the “other scene” carries with it a strong health warning: “there be monsters”.
Certainly, this is why the mapping of epidemics has largely been confined to the physical geography of its spread, not its emotional impact. The latter is supposedly the field of medical anthropology or social psychology, yet, paradoxically, cultural epidemiology – which aims to include social and cultural factors of disease transmission and control such as for instance public health measures’ social acceptance and the stigma related to disease – has not so far drawn on the creative resources or methods of participatory community mapping.
One of the most frequently reiterated phrases in official pronouncements about Covid-19 is that “we are entering uncharted territory”. So, what would it take to map this unknown landscape properly?
At one level, it requires a major, international, initiative in cartography to correlate the epidemiology of the virus with its patterns of dissemination in different societies across the world, comparing where it has taken hold in a major way with evidence from countries where the impact has been less. This would involve the commitment and coordination of technical and organisational resources on an unprecedented scale and is unlikely to happen.
But it is still valuable to entertain this grander vision as a thought experiment, and to reflect on the conceptual and methodological apparatus required. This would help also with more geographically narrow applications for similar purposes.
If the actual chains of physical transmission are impossible to map accurately once a disease reaches epidemic proportions, its social, cultural, economic and political impacts are much more easily subjected to mapping.
So, for example it is possible to show not only the spatial distribution and density of recorded cases at different points in time, but associated changes in levels and patterns of economic and social activity, geographical mobility, local travel, and public behaviour. Anyone familiar with geographic information systems (GIS) will be able to plot the links between disease hotspots, panic shopping, school closures and loss of public amenity. The global GIS company Esri provides the most comprehensive curation of GIS mappings of Covid-19 I know of. PolicyLab’s interactive maps of the US are also relevant, and their website provides some detailed instructions on how to carry out such mappings.
However, despite claims that the cartographical applications of artificial intelligence (AI) can map the spread of Covid-19 in real time, mapping spatial contiguities and even statistical correlations does not, in itself, amount to an adequate explanation of such a complex phenomenon as an epidemic. Statistical correlations are in fact just that: they show the existence of a relationship between certain variables, but cannot indicate by themselves causality – it is easy to mistake effects for causes.
Although mathematical modelling can yield elegant diagrams showing the sensitive dependence of complex non-linear systems (like the weather) on their initial conditions, and how small changes in local circumstances can produce sudden and large scale changes elsewhere in the system (the so-called butterfly effect), the attempt to apply chaos or catastrophe theory to understand tipping points in the climate of public opinion have so far failed to produce plausible accounts.
The reason is that although attitudes and behaviour can to some extent be measured through social surveys or direct observation, they are mediated by cultural value orientations and narratives that are much more difficult to pin down, let alone quantify. Equally in the case of an epidemic we are dealing with a constantly changing set of different indicators, entangled with a whole lot of factors and effects whose impact is difficult to predict or control.
These often-delayed knock-on effects are beyond the range of impact studies, and can in many cases only be studied in retrospect. Think, for instance, about the impact of Covid-19 on levels and forms of sexual intimacy, and hence indirectly on the birth rate, with couples being reluctant to bring a new baby into the world under these circumstances. We can hypothesise that this could be one effect of the pandemic, but we will have to wait much longer to be able to measure the phenomenon – and to accurately evaluate if it is indeed the case.
Should we then fall back on capturing local knowledge, and focus our mapping on finding out what meanings different groups attach to the situation, how this shapes and is shaped by their mental maps, and how they navigate the city?
As an urban ethnographer who uses participatory mapping as part of my research toolkit, I am obviously in favour of this approach. But here I need to issue another health warning. It is one thing to use such autobiographical material to document and analyse the significance of an epidemic event for the people who provide such data. Quite another to credit it as providing a “true account” without supplementary information and analysis.
Yet we still urgently need to find ways of exploring and charting this unknown territory if we are to develop a coherent response to the threat of a pandemic. One reason for the incipient panic around Covid-19 is that, however mild and localised the illness might be for many, it evokes a phantasm of death. The hidden and the unknown – for example, those now iconic “asymptomatic” carriers, but also the fact that we do not really know if we are likely to catch the light or bad version of the disease – are directly associated with a boundless mortal threat.
In other words, the invisible replication of the virus comes unconsciously to represent our contradictory and varied responses to the fear (but also the embracing) of death, well captured by Freud’s concept of the “death-drive”. This imagined or real encounter with death – whether our own, or that of close or not so close others in our societies – gives rise to a variety of psychological defences, ranging from outright denial (as in Trump’s initial refusal to give credence to the risk posed by Covid-19) to depressive withdrawal from the world (the rich hunkering down to self-isolate in remote luxury quarantine retreats) and manic panic (shops emptied of necessities).
If epidemiological maps do not directly represent the deep patterns of associative feeling and belief which constitute the mass psychology of an epidemic, they can unwittingly serve as their support. By providing representations of physical space and locations, maps are used by government agencies in modelling strategies to contain, delay, or mitigate the spread of a disease.
Yet, in many cases the dissemination of such maps through public media only serves to communicate that the situation is not under control, thus undermining the very messages of public reassurance they are designed to convey. For example, in the early phases of the pandemic many British newspapers carried maps in which a whole continent is painted red, signifying that it is host to Covid-19, even though only a tiny proportion of the population had tested positive. So, while such maps claim to simply present statistical facts, they actually unconsciously convey a worst-case scenario that is far from being a “fact”.
Such maps can be and certainly are used by public health authorities to frighten people into taking precautions, and we need to be aware that this is their purpose. If a critical approach to cartography has taught us anything, it is that maps are not to be taken at face value as objective statements of fact. What they omit and render invisible through their design parameters is often just as significant as what they show.
On the other hand, GIS cartography has undoubtedly helped to give a realistic picture of the rate of spread of the virus. Still, the general rule remains: a map is only as accurate as the data it visualises, and, in many countries, deaths from Covid-19 are drastically underreported, either because proper recording procedures are not in place, or for reasons of political impression management.
The imposition of lockdown and social distancing has radically transformed the way hundreds of millions of people relate to and navigate their physical environment, and the personal geographies of risk which they construct, both in response to official measures and their own social and cultural norms.
There have been a number of initiatives to document the cognitive mappings generated by the pandemic. CityLab invited Bloomberg readers to send in maps describing the impact of lockdown on their everyday lives. The Livingmaps Network, which I co-direct, has put out a similar call through its journal, with an added focus on how the lockdown is impacting our dream lives, and our hopes and fears for the future – if you want to contribute, you can read more about it here.
Important though such initiatives are, this kind of material needs to be interpreted with a careful attention to local social and cultural factors, if it is to contribute to our understanding of how different cultures and communities have responded to the virus, and to the measures which governments have taken to try to contain or suppress it. Only then will we begin to have an evidence base from which to construct a system of test and trace informed by real knowledge on the ground, as opposed to the top down command and control strategies which are currently favoured by the UK and some other governments.
The content and arguments presented in this blog post have been developed in more depth in a longer academic article available here on the cartographies of coronavirus recently published in Livingmaps Review.
Phil Cohen is the research director of the Livingmaps Network, and editor at large of its online journal Livingmaps Review. He is also an Emeritus Professor at the Centre for Cultural Studies Research, University of East London. His latest books are Waypoints: towards and ecology of political mindfulness (eyeglass books, 2019) and Archive That, Comrade: Left legacies and the counter culture of remembrance (PM Press 2018). "New Directions in Radical Cartography: Or why the map is not the territory" (co-edited with Michael Duggan) is forthcoming from Rowman & Littlefield.
The views expressed in this article are those of the author and do not necessarily reflect Corona Times' editorial stance, or the position of any institution or association.